From harvester at lists.evolt.org Mon Dec 11 00:00:08 2006 From: harvester at lists.evolt.org (harvester at lists.evolt.org) Date: 11 Dec 2006 00:00:08 -0600 Subject: [thelist] Tip Harvest for the Week of Monday Dec 04, 2006 Message-ID: <20061211060008.18589.qmail@tempest.evolt.org> The tip harvest for the Week of Monday Dec 04, 2006 has been added to the lists.evolt.org site. Get it at: http://lists.evolt.org/harvest/show.cgi?w=20061204 Week at a glance listing at: http://lists.evolt.org/harvest/week.cgi?w=20061204 Harvest Summary --------------- Number of messages: 153 Number of tips : 3 Tip Authors ----------- M. Seyon (1) Stephen Rider (2) Tip Types --------- Be aware of your full audience (1) Excellent Mac CSS Editor (1) Flash and progessive JPEG images (1) ------------------------------------------------------------------------------ What: Tips are how a list subscriber "pays" for making an off-topic post to the list (see List Info for more information). Of course, this does not make off-topic posts acceptable. :) How : You include a tip in your posting by using the [tip]...[/tip] tag (replace the square brackets with angle brackets). The tip tag takes the optional attributes of 'type' and 'author'. From web at ryanrushton.com Mon Dec 11 00:15:28 2006 From: web at ryanrushton.com (Ryan Rushton) Date: Sun, 10 Dec 2006 22:15:28 -0800 Subject: [thelist] Workaround for setAttribute bug in IE7? Message-ID: <457CF780.8060004@ryanrushton.com> I have a table that lists items (1 per row). When the user clicks on an item I want to insert a new row directly below the clicked item and pull in a summary from a database. My code works fine in Firefox, but in IE the line: newCell.setAttribute("colspan",8); doesn't seem to set the colspan. Any suggestions for a workaround? the HTML: ... ... the JavaScript: // Display/hide the summary for a given item function toggleSummary(recID) { var thisTable = document.getElementById('myList'); var itemRowId = 'record' + recID; var itemRow = document.getElementById(itemRowId); var itemRowCells = itemRow.getElementsByTagName("td"); var itemBgColor = itemRowCells[1].style.backgroundColor; var summaryRowId = 'summary' + recID; var summaryRow = document.getElementById(summaryRowId); // toggle off if (summaryRow != null) { thisTable.deleteRow(summaryRow.rowIndex); } // toggle on else { var newRowIndex = itemRow.rowIndex + 1; var newRow = thisTable.insertRow(newRowIndex); newRow.setAttribute("id",summaryRowId); var newCell = newRow.insertCell(0); newCell.setAttribute("colspan",8); newCell.style.backgroundColor = itemBgColor; newCell.innerHTML = ''; } return false; } From steven.pierce at gmail.com Mon Dec 11 00:57:28 2006 From: steven.pierce at gmail.com (Steven Pierce) Date: Sun, 10 Dec 2006 22:57:28 -0800 Subject: [thelist] Windows Server Administrator In-Reply-To: <456B51CF.9060602@galleryrobinson.com> References: <456B51CF.9060602@galleryrobinson.com> Message-ID: <6e0cf1cc0612102257r6072c4e9t271b80837dec1d56@mail.gmail.com> Try Hostgo.com. I know that they have servers that are Windows. They do have Admins that can help also. Steven On 11/27/06, Hershel Robinson wrote: > I am unable to locate a Dedicated Windows host with a good admin who can > install and configure software. I need specifically these: > > 1. QuickBooks 2007 > > 2. QODBC Web Server Edition ( qodbc.com/QODBCweb.htm ) to use with PHP > MySQL 4.1.15 PHP 5.0.4 > > At this point I am splitting up my search. I have places (like goDaddy) > who seem to have a fine server to provide, but no admin. > > So maybe some Evolter will step up to this task and be my admin? I am > talking about someone to install, configure and administer these tools > over the next few months on the server. Just until we can finish > transferring data between QuickBooks and MySQL. > > Thanks, > Hershel > > -- > Web Site Construction by Gallery Robinson > http://galleryrobinson.com/ > -- > > * * Please support the community that supports you. * * > http://evolt.org/help_support_evolt/ > > For unsubscribe and other options, including the Tip Harvester > and archives of thelist go to: http://lists.evolt.org > Workers of the Web, evolt ! > From hershel at galleryrobinson.com Mon Dec 11 02:09:22 2006 From: hershel at galleryrobinson.com (Hershel Robinson) Date: Mon, 11 Dec 2006 10:09:22 +0200 Subject: [thelist] Windows Server Administrator In-Reply-To: <6e0cf1cc0612102257r6072c4e9t271b80837dec1d56@mail.gmail.com> References: <456B51CF.9060602@galleryrobinson.com> <6e0cf1cc0612102257r6072c4e9t271b80837dec1d56@mail.gmail.com> Message-ID: <457D1232.2080109@galleryrobinson.com> Steven Pierce wrote: > Try Hostgo.com. I know that they have servers that are Windows. They do > have Admins that can help also. We are trying RackSpace. So far so good. :) Thanks. -- Web Site Construction by Gallery Robinson http://galleryrobinson.com/ From barney at textmatters.com Mon Dec 11 03:19:52 2006 From: barney at textmatters.com (Barney Carroll) Date: Mon, 11 Dec 2006 09:19:52 +0000 Subject: [thelist] Coding standards.... [headers] In-Reply-To: <7613e9690612090632t3abdaec4y5ac906f11ccead7d@mail.gmail.com> References: <72E9FAA171D63B48AAC707C72900E6B4A71978@ireland.spinhead.com> <7613e9690612090632t3abdaec4y5ac906f11ccead7d@mail.gmail.com> Message-ID: <457D22B8.3050109@textmatters.com> Julian Rickards wrote: > My position is that, in the same way that you can't have multiple > tags on a page to represent its multiple purposes, you shouldn't have > multiple <h1> tags. Are you saying that h1 is a reinstatement of the title? ...But somehow not superfluous? I'm playing a bit of the devil's advocate here because that's by and large how I use h1s. But I've also been known to have <title>[site name] - [page name] ... ...

[page name]

... Joel D Canfield wrote: >> it is fine to use H1, H3, H2 if that's how your document flows. > > how would a document flow like that, semantically? how would you get > to the third most important point before the second most important > point? 'Headers' are not 'points'. Someone earlier made a joking reference to the use of h3 to make text bold. The fact that HTML has (often very confusing - and rightfully so) distinctions like em, strong, ul, ol, h# etc. are to help it accommodate all the various aspects of document writing from human culture, even prior to the w3c. I feel generally there is too much of a feeling that the form of writing that must/can be used on the web must be sensitive to the way the DOM is constructed, to logical XML trees etc... And this is spitting in the face of the fact that SGML and DHTML have come so far to accommodate writing as an inherently complex thing. 1) No matter of intuition as to web standards and technologies will teach you good English. I think vice-versa would be closer to the truth, but still a fundamentally over-simplistic concept. 2) There is no 'tree' of header elements. The DOM can be perceived as a tree, but p objects are not 'contained' within their over-arching h#, nor are lower h#s contained within the next lowest h# above. Regards, Barney From costelloe at gmail.com Mon Dec 11 04:05:53 2006 From: costelloe at gmail.com (Dawson Costelloe) Date: Mon, 11 Dec 2006 10:05:53 +0000 Subject: [thelist] Acceptance Criteria Help Message-ID: <606cbdff0612110205x303487c8p1ea9bd624a5ff8aa@mail.gmail.com> Hello Evolters, I was wondering if anyone had copy of Acceptance Criteria that you may have supplied to a client previously? I'm putting forward a proposal and it's part of the requirement. Below is a brief of the kind of thing I'm looking for... "Acceptance Criteria: In this section we should outline any conditions involved with *client approving or disapproving our work. Here we will mention what happens if something is done that the client doesn't like and the process that is involved in changing it. We should also touch on any issues involving drastic changes to project scope and requirements." Thanks everyone, D. From lee.kowalkowski at googlemail.com Mon Dec 11 06:57:29 2006 From: lee.kowalkowski at googlemail.com (Lee kowalkowski) Date: Mon, 11 Dec 2006 12:57:29 +0000 Subject: [thelist] Help me think in MVC In-Reply-To: References: Message-ID: <610592c90612110457t40651013w263381a62a39d99a@mail.gmail.com> On 08/12/06, Joe Crawford wrote: > I discovered the Model-View-Controller pattern earlier this year. > Well, not discovered, I had read about it in OO programming articles > years ago. > > I think the reason I never got it was because I'd never encountered a > system that used it very well, or forced you to use it. I don't mind, but when I learned MVC, it wasn't an OO topic, it was more relevant to programming graphics / GUI applications. I don't think OO can take the credit for MVC, perhaps those articles are yet more on the topic "let's show how to do MVC using OO". When I was a graphics programmer writing pilot displays 10 years ago, MVC was the only real option (well, if you wanted to write maintainable code). Imagine a computer game, you want any single frame to represent the exact same snapshot in time. We can't allow changes to the model to influence a view during its rendition, or it could be distorted. MVC doesn't solve this, but it helps identify where you ought to solve it. The model is just an encapsulation of the real-world information, how/wher-ever it is stored. The view is responsible for rendering the model. The controller is pretty much responsible for everything else, like handling input, informing the model of changes, invoking the view. The model and view are usually pretty straight forward, yet in the real world it's not unusual to encounter some overly complicated varieties, with systems full of layers that essentially populate one type of object with the values from another, abstraction for the sake of it, or "just in case". In a web environment, it can be a tricky job to think in MVC, because potentially all three are distributed, or there are several and they are chained. OO text-books tend to contain trivialised conceptualisations. Chaining MVC systems typically means one system uses another system's view for its model. So whether something is a model or a view now depends on context. You even get views in a database. You could have the understanding that on the server, the generated HTML document is the view, but from the client-side perspective this is the model, the browser will no doubt be rendering the view of the model by applying CSS and the browser's default rendering rules. You can have several different controllers in a web application; JavaScript (especially in AJAX applications), the user's browser, the HTTP server, your server-side code that receives the requests. It looks like a right mess, but principally, any system can be defined by its input, process and output, MVC is just a different approach to system breakdown. -- Lee From austin at dotmail.co.uk Mon Dec 11 05:03:03 2006 From: austin at dotmail.co.uk (Austin Harris) Date: Mon, 11 Dec 2006 11:03:03 +0000 (GMT) Subject: [thelist] Coding standards.... [headers] In-Reply-To: <457D22B8.3050109@textmatters.com> Message-ID: <10229202.131165834983945.JavaMail.root@redroom.customer.ultraspeed.co.uk> I always thought that headers should flow correctly ie: h1 , h2, h3 etc and not h1, h3, h2. You can go back up ie: h1, h2 , h3, h4, h2, h3 tc Appears to back me up on this. It can in my experience be nearly impossible to achieve this in the "real world" though... Also I've never needed to use more than one h1 tag in a page. Austin Julian Rickards wrote: > My position is that, in the same way that you can't have multiple > tags on a page to represent its multiple purposes, you shouldn't have > multiple <h1> tags. Are you saying that h1 is a reinstatement of the title? ...But somehow not superfluous? I'm playing a bit of the devil's advocate here because that's by and large how I use h1s. But I've also been known to have <title>[site name] - [page name] ... ...

[page name]

... Joel D Canfield wrote: >> it is fine to use H1, H3, H2 if that's how your document flows. > > how would a document flow like that, semantically? how would you get > to the third most important point before the second most important > point? 'Headers' are not 'points'. Someone earlier made a joking reference to the use of h3 to make text bold. The fact that HTML has (often very confusing - and rightfully so) distinctions like em, strong, ul, ol, h# etc. are to help it accommodate all the various aspects of document writing from human culture, even prior to the w3c. I feel generally there is too much of a feeling that the form of writing that must/can be used on the web must be sensitive to the way the DOM is constructed, to logical XML trees etc... And this is spitting in the face of the fact that SGML and DHTML have come so far to accommodate writing as an inherently complex thing. 1) No matter of intuition as to web standards and technologies will teach you good English. I think vice-versa would be closer to the truth, but still a fundamentally over-simplistic concept. 2) There is no 'tree' of header elements. The DOM can be perceived as a tree, but p objects are not 'contained' within their over-arching h#, nor are lower h#s contained within the next lowest h# above. Regards, Barney -- * * Please support the community that supports you. * * http://evolt.org/help_support_evolt/ For unsubscribe and other options, including the Tip Harvester and archives of thelist go to: http://lists.evolt.org Workers of the Web, evolt ! From lee.kowalkowski at googlemail.com Mon Dec 11 08:54:27 2006 From: lee.kowalkowski at googlemail.com (Lee kowalkowski) Date: Mon, 11 Dec 2006 14:54:27 +0000 Subject: [thelist] Coding standards.... [headers] In-Reply-To: <10229202.131165834983945.JavaMail.root@redroom.customer.ultraspeed.co.uk> References: <457D22B8.3050109@textmatters.com> <10229202.131165834983945.JavaMail.root@redroom.customer.ultraspeed.co.uk> Message-ID: <610592c90612110654iafc71efq735f7e88103f4fa2@mail.gmail.com> Folks, To understand the semantic angle. Let's suppose there was a CSS property you could apply to your heading elements that automatically numbered your headings using the notation 1.1.1... etc. A level 1 heading, would start a new section. On the web, you could say a new section should start a new page, just like when authoring non-web documents. Therefore, you would put something in your coding standards to this effect. Your numbering solution would also need to be able to start numbering from any arbitrary number in this case. Level 2 headings would be direct subsections of the level 1 section, and have numbers like 1.1, 1.2, 2.1, 8.5, and so on for the rest of the levels. I think most people are familiar with this system of numbered headings. HTML heading elements aren't automatically numbered, but it is not uncommon for people to apply a similar semantic view. Which means it doesn't make sense to have a level 3 heading without a level 2 (or a level 1). That is, it doesn't make sense to have a section 1.3.2 if there's no section 1.3 or no section 1. But this is just a traditional semantic view. It may be feasible in another design for the different heading levels to have an alternative semantic meaning, e.g., level 1 = Emergency Notices, level 2 = Warning Notices, level 3 = Information Notices. Both systems can be implemented with valid HTML, although you do get a telling-off if you've skipped headings and have checked "show outline" in the W3C validator. It's just a matter of ensuring the coding standards support/enforce the semantic view of the chosen design. -- Lee From julian.rickards at gmail.com Mon Dec 11 09:02:52 2006 From: julian.rickards at gmail.com (Julian Rickards) Date: Mon, 11 Dec 2006 10:02:52 -0500 Subject: [thelist] Coding standards.... [headers] In-Reply-To: <610592c90612110654iafc71efq735f7e88103f4fa2@mail.gmail.com> References: <457D22B8.3050109@textmatters.com> <10229202.131165834983945.JavaMail.root@redroom.customer.ultraspeed.co.uk> <610592c90612110654iafc71efq735f7e88103f4fa2@mail.gmail.com> Message-ID: <7613e9690612110702p177418e7mc01fe67e9ae106fa@mail.gmail.com> I don't look at heading level as a degree of importance but rather as a summary of what is discussed following the heading. H2 is no more important than H3 and so on, it is just that H2 covers a broader topic (not necessarily longer either) than H3 which is more focussed. Other than an ordered list, I don't know that HTML provides for degrees of importance. Recognize that my opinions are just that, opinions and I am willing to be convinced differently: to date, this hasn't yet happened but I am pleased to see this being discussed. Jules On 11/12/06, Lee kowalkowski wrote: > > > But this is just a traditional semantic view. It may be feasible in > another design for the different heading levels to have an alternative > semantic meaning, e.g., level 1 = Emergency Notices, level 2 = Warning > Notices, level 3 = Information Notices. -- Website: http://jrickards.ca Blog: http://pen-and-ink.ca E-mail: julian dot rickards at gmail dot com From julian.rickards at gmail.com Mon Dec 11 07:31:56 2006 From: julian.rickards at gmail.com (Julian Rickards) Date: Mon, 11 Dec 2006 08:31:56 -0500 Subject: [thelist] Coding standards.... [headers] In-Reply-To: <457D22B8.3050109@textmatters.com> References: <72E9FAA171D63B48AAC707C72900E6B4A71978@ireland.spinhead.com> <7613e9690612090632t3abdaec4y5ac906f11ccead7d@mail.gmail.com> <457D22B8.3050109@textmatters.com> Message-ID: <7613e9690612110531h3f581874k57f4345954443780@mail.gmail.com> Yes, I believe it is. The H1, as I see it, is to identify the content or purpose of the current page and it is the visible title (because the tag contents are not visible with the "web page" per se, I consider it to be somewhat invisible). However, when you look at the results of a search engine, having the same text in the H1 as part of the TITLE means that the contents or purpose of the page is clearly visible because the link to the page uses the TITLE content, even if the search terms are not found in the TITLE. One other comment about your title content as shown below, I would recommend that the [page name] be placed to the left of the [site name] so that if the title is going to be cut off, it will be the site name that is cut off, not the page name. Also, it is easier to find if a page is the one you are looking for if the page name is first and not second. Jules On 11/12/06, Barney Carroll <barney at textmatters.com> wrote: > > > Are you saying that h1 is a reinstatement of the title? ...But somehow > not superfluous? I'm playing a bit of the devil's advocate here because > that's by and large how I use h1s. > > <title>[site name] - [page name] > -- Website: http://jrickards.ca Blog: http://pen-and-ink.ca E-mail: julian dot rickards at gmail dot com From barney at textmatters.com Mon Dec 11 09:26:46 2006 From: barney at textmatters.com (Barney Carroll) Date: Mon, 11 Dec 2006 15:26:46 +0000 Subject: [thelist] Coding standards.... [headers] In-Reply-To: <7613e9690612110702p177418e7mc01fe67e9ae106fa@mail.gmail.com> References: <457D22B8.3050109@textmatters.com> <10229202.131165834983945.JavaMail.root@redroom.customer.ultraspeed.co.uk> <610592c90612110654iafc71efq735f7e88103f4fa2@mail.gmail.com> <7613e9690612110702p177418e7mc01fe67e9ae106fa@mail.gmail.com> Message-ID: <457D78B6.9030102@textmatters.com> Forgive me for saying so... But it's almost as if we're trying to decide for ourselves what these things might be from a vacuum and the guidance of w3c... What about books? Seriously, the culture of literature is a massive thing with plenty of examples - memory can join hands with idle theory!

Putting HTML in context

In this chapter we will look at the ontology of markup languages from SGML to present day XML and XHTML. It is important, before we delve into a history of the 'ML', to draw attention to two key recent conceptions which must be made distinct in order to better understand the world without them.

The Document Object Model

The Document Object Model (DOM) took a long while to cement itself as the key method for browser clients to treat markup, and in turn it took a strong role in shaping how developers and authors thought of extensible markup.

It is, however, only a recent model for processing markup, and even at level 1 was not as malleable as we now rely on it being.

DHTML

DHTML (Dynamic HyperText Markup Language) has fallen out of use as a term recently - for good reasons. It is not a markup language proper in the sense that XHTML, HTML and XML are distinct - rather DHTML represents the use of HTML, CSS and ECMAscript (more commonly known as javascript) - but its implication, even of the division of such languages, is very significant to the way attitudes to markup differed before and after its inception.

SGML: The beginning of everything

The notion of Markup Languages as we know them - that is to say... ... 'Emergency'... 'level of importance'... Strange terms. A header is a header. I'm no professor of English but I'm very concerned at how abstract these perfectly ingrained human terms seem to be... Regards, Barney Julian Rickards wrote: > I don't look at heading level as a degree of importance but rather as a > summary of what is discussed following the heading. H2 is no more important > than H3 and so on, it is just that H2 covers a broader topic (not > necessarily longer either) than H3 which is more focussed. > > Other than an ordered list, I don't know that HTML provides for degrees of > importance. > > Recognize that my opinions are just that, opinions and I am willing to be > convinced differently: to date, this hasn't yet happened but I am pleased to > see this being discussed. > > Jules > > On 11/12/06, Lee kowalkowski wrote: >> >> But this is just a traditional semantic view. It may be feasible in >> another design for the different heading levels to have an alternative >> semantic meaning, e.g., level 1 = Emergency Notices, level 2 = Warning >> Notices, level 3 = Information Notices. From joel at spinhead.com Mon Dec 11 09:21:46 2006 From: joel at spinhead.com (Joel D Canfield) Date: Mon, 11 Dec 2006 07:21:46 -0800 Subject: [thelist] Coding standards.... [headers] Message-ID: <72E9FAA171D63B48AAC707C72900E6B4A71987@ireland.spinhead.com> > Are you saying that h1 is a reinstatement of the title? page title plays a big role in SEO and should be fairly inclusive;

should be more concise since (afaik) it's lumped in with the rest of the page content from an SEO perspective (SEO experts are welcome to correct me if that's a misperception) joel From julian.rickards at gmail.com Mon Dec 11 09:35:40 2006 From: julian.rickards at gmail.com (Julian Rickards) Date: Mon, 11 Dec 2006 10:35:40 -0500 Subject: [thelist] Coding standards.... [headers] In-Reply-To: <457D78B6.9030102@textmatters.com> References: <457D22B8.3050109@textmatters.com> <10229202.131165834983945.JavaMail.root@redroom.customer.ultraspeed.co.uk> <610592c90612110654iafc71efq735f7e88103f4fa2@mail.gmail.com> <7613e9690612110702p177418e7mc01fe67e9ae106fa@mail.gmail.com> <457D78B6.9030102@textmatters.com> Message-ID: <7613e9690612110735v1edf7792tce82ec72598d5dab@mail.gmail.com> Based on my opinion of this topic, I would wonder what larger topic do your H3's belong to. On 11/12/06, Barney Carroll wrote: > > Forgive me for saying so... But it's almost as if we're trying to decide > for ourselves what these things might be from a vacuum and the guidance > of w3c... What about books? Seriously, the culture of literature is a > massive thing with plenty of examples - memory can join hands with idle > theory! > >

Putting HTML in context

>

In this chapter we will look at the ontology of markup languages from > SGML to present day XML and XHTML. It is important, before we delve into > a history of the 'ML', to draw attention to two key recent conceptions > which must be made distinct in order to better understand the world > without them.

> >

The Document Object Model

>

The Document Object Model (DOM) took a long while to cement itself as > the key method for browser clients to treat markup, and in turn it took > a strong role in shaping how developers and authors thought of > extensible markup.

>

It is, however, only a recent model for processing markup, and even > at level 1 was not as malleable as we now rely on it being.

> >

DHTML

>

DHTML (Dynamic HyperText Markup Language) has fallen out of use as a > term recently - for good reasons. It is not a markup language proper in > the sense that XHTML, HTML and XML are distinct - rather DHTML > represents the use of HTML, CSS and ECMAscript (more commonly known as > javascript) - but its implication, even of the division of such > languages, is very significant to the way attitudes to markup differed > before and after its inception.

> >

SGML: The beginning of everything

>

The notion of Markup Languages as we know them - that is to say... > > ... > > 'Emergency'... 'level of importance'... Strange terms. A header is a > header. I'm no professor of English but I'm very concerned at how > abstract these perfectly ingrained human terms seem to be... > > Regards, > Barney -- Website: http://jrickards.ca Blog: http://pen-and-ink.ca E-mail: julian dot rickards at gmail dot com From julian.rickards at gmail.com Mon Dec 11 09:50:13 2006 From: julian.rickards at gmail.com (Julian Rickards) Date: Mon, 11 Dec 2006 10:50:13 -0500 Subject: [thelist] Coding standards.... [headers] In-Reply-To: <72E9FAA171D63B48AAC707C72900E6B4A71987@ireland.spinhead.com> References: <72E9FAA171D63B48AAC707C72900E6B4A71987@ireland.spinhead.com> Message-ID: <7613e9690612110750g2065ba51j1358cb005c19cadd@mail.gmail.com> I'm not a so-called SEO expert but I don't think that you need to modify the page title for H1 or TITLE, just use the same. However, the TITLE can have additional information such as the section within which the page title falls and the web site name such as "Brown Leather Gloves - Accessories - Winter Wear - TipTop Tailors, Your Local Mens Clothiers". On 11/12/06, Joel D Canfield wrote: > > > Are you saying that h1 is a reinstatement of the title? > > page title plays a big role in SEO and should be fairly inclusive;

> should be more concise since (afaik) it's lumped in with the rest of the > page content from an SEO perspective (SEO experts are welcome to correct > me if that's a misperception) > > joel > -- > > * * Please support the community that supports you. * * > http://evolt.org/help_support_evolt/ > > For unsubscribe and other options, including the Tip Harvester > and archives of thelist go to: http://lists.evolt.org > Workers of the Web, evolt ! > -- Website: http://jrickards.ca Blog: http://pen-and-ink.ca E-mail: julian dot rickards at gmail dot com From lee.kowalkowski at googlemail.com Mon Dec 11 09:56:57 2006 From: lee.kowalkowski at googlemail.com (Lee kowalkowski) Date: Mon, 11 Dec 2006 15:56:57 +0000 Subject: [thelist] Coding standards.... [headers] In-Reply-To: <7613e9690612110702p177418e7mc01fe67e9ae106fa@mail.gmail.com> References: <457D22B8.3050109@textmatters.com> <10229202.131165834983945.JavaMail.root@redroom.customer.ultraspeed.co.uk> <610592c90612110654iafc71efq735f7e88103f4fa2@mail.gmail.com> <7613e9690612110702p177418e7mc01fe67e9ae106fa@mail.gmail.com> Message-ID: <610592c90612110756p6ef8a650i5d5413e61dd8cc81@mail.gmail.com> On 11/12/06, Julian Rickards wrote: > I don't look at heading level as a degree of importance but rather as a > summary of what is discussed following the heading. H2 is no more important > than H3 and so on, it is just that H2 covers a broader topic (not > necessarily longer either) than H3 which is more focussed. I tend to agree with you, after all a heading is semantically meaningless if it has no associated content. But the W3C HTML Specification mentions both importance and structure: http://www.w3.org/TR/html4/struct/global.html#h-7.5.5. > Other than an ordered list, I don't know that HTML provides for degrees of > importance. I don't think importance can even be implied in an ordered list. It depends on context, an ordered list may be used for chronological numbering, like the steps in a recipe. -- I often see encouragement for only one H1 element in a web page, which I also tend to agree with. However, in a "portal" application, one could view several disparate "documents" (portlets) confederated into the same page. Hmm, dubiously assuming portals provide good usability in the first place - I think there could be situations where it would be semantically meaningless to put potentially unrelated parts of a page under the same heading, without the said heading becoming just as meaningless. -- Lee From barney at textmatters.com Mon Dec 11 10:03:57 2006 From: barney at textmatters.com (Barney Carroll) Date: Mon, 11 Dec 2006 16:03:57 +0000 Subject: [thelist] Coding standards.... [headers] In-Reply-To: <7613e9690612110735v1edf7792tce82ec72598d5dab@mail.gmail.com> References: <457D22B8.3050109@textmatters.com> <10229202.131165834983945.JavaMail.root@redroom.customer.ultraspeed.co.uk> <610592c90612110654iafc71efq735f7e88103f4fa2@mail.gmail.com> <7613e9690612110702p177418e7mc01fe67e9ae106fa@mail.gmail.com> <457D78B6.9030102@textmatters.com> <7613e9690612110735v1edf7792tce82ec72598d5dab@mail.gmail.com> Message-ID: <457D816D.8010505@textmatters.com> Julian Rickards wrote: > Based on my opinion of this topic, I would wonder what larger topic do your > H3's belong to. Honestly? The purpose of the author's use of headers here seems pretty clear to me (then again I do know him and he isn't a published writer!): In this context, the h1 is a chapter title. It contains information about everything to follow up until the next h1. The chapter needs breaking up into smaller parts, for each instantiation of *ML technology, in chronological order. Prior to the first 'sub-chapter', there is an introduction (it would be terrible to dump the reader in without at least saying what is to follow) - but this introduction is a guide to the whole of what follows and, whereas not part of any sub-division, is definitely part of the h1. This is customary to the point where it is unnecessary to have what you might want to be an h2 stating 'Introduction', especially at this size. The introduction needs to give the reader advance warning of specific notions that may be needed to properly understand the content to follow. These are two distinct points which should certainly not be h2s - that would disrupt the clear semantic flow of h2s as they are presently used in the article. Neither do they want their own containing h2, which would do the same thing - and to put them within the next existent h2 defeats the point. Of course as web designers we should be concerned about every aspect of semantics, and need to look below the surface - but when this is rendered in HTML through most browsers without even any CSS, it would make perfect sense. Julian Rickards wrote: > One other comment about your title content as shown below, I would > recommend that the [page name] be placed to the left of the > [site name] so that if the title is going to be cut off, it will be > the site name that is cut off, not the page name. Also, it is easier > to find if a page is the one you are looking for if the page name is > first and not second. That makes a lot of sense actually, but could be construed as running counter to tradition in more significant a way than what I'm suggesting as running order of h#s... In terms of our existing culture of documents as opposed to that of the free-floating information designer. The impression I get is that what we have here is information architects become concerned with the job of information designers! I believe you need both to create a truly successful website of any complexity. It especially helps when they are two separate individuals! Regards, Barney From morrison.ben at gmail.com Mon Dec 11 09:39:47 2006 From: morrison.ben at gmail.com (ben morrison) Date: Mon, 11 Dec 2006 15:39:47 +0000 Subject: [thelist] Coding standards.... [headers] In-Reply-To: <72E9FAA171D63B48AAC707C72900E6B4A71987@ireland.spinhead.com> References: <72E9FAA171D63B48AAC707C72900E6B4A71987@ireland.spinhead.com> Message-ID: <6073aef90612110739m2405b11ctbbfdcbb0a16ff87b@mail.gmail.com> On 12/11/06, Joel D Canfield wrote: > > Are you saying that h1 is a reinstatement of the title? > > page title plays a big role in SEO and should be fairly inclusive;

> should be more concise since (afaik) it's lumped in with the rest of the > page content from an SEO perspective (SEO experts are welcome to correct > me if that's a misperception) I've always been uneder the impression that Title and H1 are similar in importance and it will carry more weight if keywords are the same in both. http://www.iso9001help.co.uk/_seo_tutorial.htm ben From lee.kowalkowski at googlemail.com Mon Dec 11 10:09:41 2006 From: lee.kowalkowski at googlemail.com (Lee kowalkowski) Date: Mon, 11 Dec 2006 16:09:41 +0000 Subject: [thelist] Coding standards.... [headers] In-Reply-To: <457D78B6.9030102@textmatters.com> References: <457D22B8.3050109@textmatters.com> <10229202.131165834983945.JavaMail.root@redroom.customer.ultraspeed.co.uk> <610592c90612110654iafc71efq735f7e88103f4fa2@mail.gmail.com> <7613e9690612110702p177418e7mc01fe67e9ae106fa@mail.gmail.com> <457D78B6.9030102@textmatters.com> Message-ID: <610592c90612110809o4f13d182l2715d4300a41c204@mail.gmail.com> On 11/12/06, Barney Carroll wrote: > 'Emergency'... 'level of importance'... Strange terms. A header is a > header. I'm no professor of English but I'm very concerned at how > abstract these perfectly ingrained human terms seem to be... Wow, I'm rapidly losing you Barney! I'm no professor of English, but e.g. means "for example". LOL! That would be a professor of Latin anyway wouldn't it? I think you're missing the point a little, it's not the meaning of "heading" that we're discussing, as you rightly point out, we should all pretty much fathom that. It's the meaning or application of the levels 1 to 6 people seem to struggle with. What do they mean? Well, desparately trying to pull my content back to the topic of coding standards - it's up to the owner of the said coding standards to define their meaning, and oversee their use. There are some widespread opinions we have discussed for guidance: I.e. "Only 1 level 1 heading.", and "Do not skip levels.". -- Lee From julian.rickards at gmail.com Mon Dec 11 10:01:09 2006 From: julian.rickards at gmail.com (Julian Rickards) Date: Mon, 11 Dec 2006 11:01:09 -0500 Subject: [thelist] Coding standards.... [headers] In-Reply-To: <610592c90612110756p6ef8a650i5d5413e61dd8cc81@mail.gmail.com> References: <457D22B8.3050109@textmatters.com> <10229202.131165834983945.JavaMail.root@redroom.customer.ultraspeed.co.uk> <610592c90612110654iafc71efq735f7e88103f4fa2@mail.gmail.com> <7613e9690612110702p177418e7mc01fe67e9ae106fa@mail.gmail.com> <610592c90612110756p6ef8a650i5d5413e61dd8cc81@mail.gmail.com> Message-ID: <7613e9690612110801g2b2b10acx42dacbd3537b7f06@mail.gmail.com> With the appropriate leading text, an ordered list could be used as a list of importance. On 11/12/06, Lee kowalkowski wrote: > > > > Other than an ordered list, I don't know that HTML provides for degrees > of > > importance. > > I don't think importance can even be implied in an ordered list. It > depends on context, an ordered list may be used for chronological > numbering, like the steps in a recipe. -- Website: http://jrickards.ca Blog: http://pen-and-ink.ca E-mail: julian dot rickards at gmail dot com From lee.kowalkowski at googlemail.com Mon Dec 11 10:15:34 2006 From: lee.kowalkowski at googlemail.com (Lee kowalkowski) Date: Mon, 11 Dec 2006 16:15:34 +0000 Subject: [thelist] Coding standards.... [headers] In-Reply-To: <7613e9690612110801g2b2b10acx42dacbd3537b7f06@mail.gmail.com> References: <457D22B8.3050109@textmatters.com> <10229202.131165834983945.JavaMail.root@redroom.customer.ultraspeed.co.uk> <610592c90612110654iafc71efq735f7e88103f4fa2@mail.gmail.com> <7613e9690612110702p177418e7mc01fe67e9ae106fa@mail.gmail.com> <610592c90612110756p6ef8a650i5d5413e61dd8cc81@mail.gmail.com> <7613e9690612110801g2b2b10acx42dacbd3537b7f06@mail.gmail.com> Message-ID: <610592c90612110815p174b5ff1g7d56f8413c8f31e9@mail.gmail.com> On 11/12/06, Julian Rickards wrote: > With the appropriate leading text, an ordered list could be used as a list > of importance. That is what I'm saying, the leading text would be therefore providing context. -- Lee From joel at spinhead.com Mon Dec 11 09:47:47 2006 From: joel at spinhead.com (Joel D Canfield) Date: Mon, 11 Dec 2006 07:47:47 -0800 Subject: [thelist] Coding standards.... [headers] Message-ID: <72E9FAA171D63B48AAC707C72900E6B4A71988@ireland.spinhead.com> > of w3c... What about books? Seriously, the culture of literature is a > massive thing with plenty of examples - memory can join hands > with idle > theory! your example, coupled with the reference to print, looks as if you're using the heading tags for their appearance. if you like the visual formatting of

for your second level headings, style your References: <457D22B8.3050109@textmatters.com> <10229202.131165834983945.JavaMail.root@redroom.customer.ultraspeed.co.uk> <610592c90612110654iafc71efq735f7e88103f4fa2@mail.gmail.com> <7613e9690612110702p177418e7mc01fe67e9ae106fa@mail.gmail.com> <457D78B6.9030102@textmatters.com> <7613e9690612110735v1edf7792tce82ec72598d5dab@mail.gmail.com> <457D816D.8010505@textmatters.com> Message-ID: <7613e9690612110814j63e28b7cs836e9ca367f87b69@mail.gmail.com> I see the point you (and your friend?) are making but I still don't agree with skipping headers. Maybe an unordered list might be considered. On 11/12/06, Barney Carroll wrote: > > Julian Rickards wrote: > > Based on my opinion of this topic, I would wonder what larger topic do > your > > H3's belong to. > > Honestly? The purpose of the author's use of headers here seems pretty > clear to me (then again I do know him and he isn't a published writer!): > > In this context, the h1 is a chapter title. It contains information > about everything to follow up until the next h1. > > The chapter needs breaking up into smaller parts, for each instantiation > of *ML technology, in chronological order. > > Prior to the first 'sub-chapter', there is an introduction (it would be > terrible to dump the reader in without at least saying what is to > follow) - but this introduction is a guide to the whole of what follows > and, whereas not part of any sub-division, is definitely part of the h1. > This is customary to the point where it is unnecessary to have what you > might want to be an h2 stating 'Introduction', especially at this size. > > The introduction needs to give the reader advance warning of specific > notions that may be needed to properly understand the content to follow. > These are two distinct points which should certainly not be h2s - that > would disrupt the clear semantic flow of h2s as they are presently used > in the article. Neither do they want their own containing h2, which > would do the same thing - and to put them within the next existent h2 > defeats the point. -- Website: http://jrickards.ca Blog: http://pen-and-ink.ca E-mail: julian dot rickards at gmail dot com From julian.rickards at gmail.com Mon Dec 11 10:37:07 2006 From: julian.rickards at gmail.com (Julian Rickards) Date: Mon, 11 Dec 2006 11:37:07 -0500 Subject: [thelist] Coding standards.... [headers] In-Reply-To: <72E9FAA171D63B48AAC707C72900E6B4A71988@ireland.spinhead.com> References: <72E9FAA171D63B48AAC707C72900E6B4A71988@ireland.spinhead.com> Message-ID: <7613e9690612110837h6b514ad7n5a5eb0a35836d878@mail.gmail.com> I was going to raise a similar point but I decided against it: now I have to comment on yours. In the same way that it is "dangerous" to select an HTML element based on its appearance (default or CSS-styled), it also is "dangerous" to play with CSS. For example, by styling some H3s to look like H2s so although the HTML structure may use one pattern, the styling makes sighted people think the structure is different than it really is. Now, there are exceptions that I would be willing to consider. For example, let's say that you want a heading to precede a glossary at the bottom of the page. You may decide that H2 is the appropriate heading level but because the glossary is outside of the primary discussion, you want to style it differently and that would be fine by me. Jules On 11/12/06, Joel D Canfield wrote: > > > your example, coupled with the reference to print, looks as if you're > using the heading tags for their appearance. if you like the visual > formatting of

for your second level headings, style your

to > look however you want; appearance doesn't matter as much to screen > readers, etc. as *semantics* do. -- Website: http://jrickards.ca Blog: http://pen-and-ink.ca E-mail: julian dot rickards at gmail dot com From joel at spinhead.com Mon Dec 11 10:54:18 2006 From: joel at spinhead.com (Joel D Canfield) Date: Mon, 11 Dec 2006 08:54:18 -0800 Subject: [thelist] Coding standards.... [headers] Message-ID: <72E9FAA171D63B48AAC707C72900E6B4A71989@ireland.spinhead.com> > by styling some H3s to look like H2s not some, all; consistency is crucial and I didn't intend to imply otherwise. what I was saying was, if you like the visual appearance of

where semantically you would use

, CSS is the answer. but you have to style all your

s this way, and then style

differently enough to be of value to sighted users. inconsistency is the dangerous element ;) joel From beertastic at gmail.com Mon Dec 11 11:12:32 2006 From: beertastic at gmail.com (Tris) Date: Mon, 11 Dec 2006 17:12:32 +0000 Subject: [thelist] Creating zip files in PHP that are HUGE! Message-ID: <8b20a7490612110912s47330e0q75552f64fee01232@mail.gmail.com> I'm added this cool class to my site: http://www.phpconcept.net/pclzip/index.en.php It all works great, until I try and compile a large zip (80Mb) SO my question(s) are: 1. Anyone uesd tis before on large files 2. would a Linux server have an inbuilt setting that stops it processign internally if there was too much load? 3. Any other ideas for creating large zips on teh fly..?? Tris... (I'm currently trying to create them and email a link) From cditty at gmail.com Mon Dec 11 12:36:20 2006 From: cditty at gmail.com (Chris Ditty) Date: Mon, 11 Dec 2006 12:36:20 -0600 Subject: [thelist] Creating zip files in PHP that are HUGE! In-Reply-To: <8b20a7490612110912s47330e0q75552f64fee01232@mail.gmail.com> References: <8b20a7490612110912s47330e0q75552f64fee01232@mail.gmail.com> Message-ID: I just use the system() function and use the built in zip program that comes with linux. I zip files that are easily 200 meg in size daily. CDitty On 12/11/06, Tris wrote: > I'm added this cool class to my site: > http://www.phpconcept.net/pclzip/index.en.php > > It all works great, until I try and compile a large zip (80Mb) > > SO my question(s) are: > > 1. Anyone uesd tis before on large files > 2. would a Linux server have an inbuilt setting that stops it > processign internally if there was too much load? > 3. Any other ideas for creating large zips on teh fly..?? > > Tris... > > (I'm currently trying to create them and email a link) > -- > > * * Please support the community that supports you. * * > http://evolt.org/help_support_evolt/ > > For unsubscribe and other options, including the Tip Harvester > and archives of thelist go to: http://lists.evolt.org > Workers of the Web, evolt ! > From tim at hyperlinkage.com Mon Dec 11 14:45:59 2006 From: tim at hyperlinkage.com (tim) Date: Mon, 11 Dec 2006 20:45:59 +0000 Subject: [thelist] Coding standards.... [headers] In-Reply-To: <7613e9690612110702p177418e7mc01fe67e9ae106fa@mail.gmail.com> References: <457D22B8.3050109@textmatters.com> <10229202.131165834983945.JavaMail.root@redroom.customer.ultraspeed.co.uk> <610592c90612110654iafc71efq735f7e88103f4fa2@mail.gmail.com> <7613e9690612110702p177418e7mc01fe67e9ae106fa@mail.gmail.com> Message-ID: <457DC387.7090303@hyperlinkage.com> Julian Rickards wrote: > I don't look at heading level as a degree of importance but rather as a > summary of what is discussed following the heading. H2 is no more important > than H3 and so on, it is just that H2 covers a broader topic I've enjoyed following the discussion on this topic. Interesting that you don't regard headings as having different levels of importance, as in fact that is the only meaning given to the numbers by the HTML specification. Although I don't agree with your strict usage of heading tags for all situations, I can see how it could be a valuable addition to a set of coding standards. Tim Using CSS to replace text with a background image is widely thought to keep semantic meaning while retaining the graphical appearance of a document. However, I have recently noticed a serious usability problem with this technique. Many users with colour blindness or other visual impairment may configure their browser to use custom background colours. This has the effect of replacing all backgrounds (including background images) on a page with their choice of colour. This will removes the replaced image and, as all other CSS styles are still rendered, *the text alternative will not appear*. If not used with extreme caution, this could render a web site making heavy use of this technique totally unnavigable. -- www.hyperlinkage.com From edwin at bitstorm.org Mon Dec 11 16:48:50 2006 From: edwin at bitstorm.org (Edwin Martin) Date: Mon, 11 Dec 2006 23:48:50 +0100 Subject: [thelist] Workaround for setAttribute bug in IE7? In-Reply-To: <457CF780.8060004@ryanrushton.com> References: <457CF780.8060004@ryanrushton.com> Message-ID: <457DE052.9000809@bitstorm.org> Ryan Rushton wrote: > I have a table that lists items (1 per row). When the user clicks on an > item I want to insert a new row directly below the clicked item and pull > in a summary from a database. My code works fine in Firefox, but in IE > the line: newCell.setAttribute("colspan",8); doesn't seem to set the > colspan. Any suggestions for a workaround? > Have you tried newCell.colSpan=8? (I wonder why people use setAttribute so much, while the HTML-DOM has all these ready-to-use properties). http://www.w3.org/TR/REC-DOM-Level-1/ecma-script-language-binding.html More info about table manipulation in IE: http://msdn.microsoft.com/workshop/author/tables/buildtables.asp Edwin Martin > > the HTML: >

Item List
154 10490 Item One 12/8/06 Active John Doe edit
> > ... > > > > > > > > > > > ... > > the JavaScript: > // Display/hide the summary for a given item > function toggleSummary(recID) { > var thisTable = document.getElementById('myList'); > var itemRowId = 'record' + recID; > var itemRow = document.getElementById(itemRowId); > var itemRowCells = itemRow.getElementsByTagName("td"); > var itemBgColor = itemRowCells[1].style.backgroundColor; > var summaryRowId = 'summary' + recID; > var summaryRow = document.getElementById(summaryRowId); > > // toggle off > if (summaryRow != null) { > thisTable.deleteRow(summaryRow.rowIndex); > } > > // toggle on > else { > var newRowIndex = itemRow.rowIndex + 1; > var newRow = thisTable.insertRow(newRowIndex); > newRow.setAttribute("id",summaryRowId); > var newCell = newRow.insertCell(0); > newCell.setAttribute("colspan",8); > newCell.style.backgroundColor = itemBgColor; > newCell.innerHTML = ''; > } > return false; > } > -- http://www.bitstorm.org/ From evolt_org at striderweb.com Mon Dec 11 16:58:58 2006 From: evolt_org at striderweb.com (Stephen Rider) Date: Mon, 11 Dec 2006 16:58:58 -0600 Subject: [thelist] Workaround for setAttribute bug in IE7? In-Reply-To: <457DE052.9000809@bitstorm.org> References: <457CF780.8060004@ryanrushton.com> <457DE052.9000809@bitstorm.org> Message-ID: On Dec 11, 2006, at 4:48 PM, Edwin Martin wrote: > Have you tried newCell.colSpan=8? > > (I wonder why people use setAttribute so much, while the HTML-DOM has > all these ready-to-use properties). Because it's easier to remember one rule than a thousand little rules. Of course, thanks to IE we can't just do that. Steve From mwarden at gmail.com Mon Dec 11 17:09:20 2006 From: mwarden at gmail.com (Matt Warden) Date: Mon, 11 Dec 2006 18:09:20 -0500 Subject: [thelist] Workaround for setAttribute bug in IE7? In-Reply-To: <457DE052.9000809@bitstorm.org> References: <457CF780.8060004@ryanrushton.com> <457DE052.9000809@bitstorm.org> Message-ID: On 12/11/06, Edwin Martin wrote: > Ryan Rushton wrote: > > I have a table that lists items (1 per row). When the user clicks on an > > item I want to insert a new row directly below the clicked item and pull > > in a summary from a database. My code works fine in Firefox, but in IE > > the line: newCell.setAttribute("colspan",8); doesn't seem to set the > > colspan. Any suggestions for a workaround? > > > Have you tried newCell.colSpan=8? > > (I wonder why people use setAttribute so much, while the HTML-DOM has > all these ready-to-use properties). > > http://www.w3.org/TR/REC-DOM-Level-1/ecma-script-language-binding.html Why? Because it makes no sense that setAttribute works in some cases, unless you are talking about IE and attributes which have been turned into object properties. Here is the meaning of setAttribute from devguru: "The setAttribute method sets the value of the named attribute. If an attribute of the same name already exists in the element, then its value is changed to that of the 'value' parameter; *if not, one is created*." (emphasis mine) and from moz: "If the specified attribute already exists, then the value of that attribute is changed to the value passed to this function. *If it does not exist, then the attribute is created*." (emphasis mine) It is very clearly a DOM manipulation. Why would I think that setting .colSpan object property in IE would create an attribute node in the DOM? It is a really dumb idea, especially when you consider what happens when one does: foo.colSpan = null; Does this remove an existing attribute node (doubt it)? Does it just set its value to null (probably)? If the attribute node doesn't already exist, does it create it and set it to null (probably)? -- Matt Warden Cleveland, OH, USA http://mattwarden.com This email proudly and graciously contributes to entropy. From anthony at baratta.com Mon Dec 11 11:44:51 2006 From: anthony at baratta.com (Anthony Baratta) Date: Mon, 11 Dec 2006 09:44:51 -0800 Subject: [thelist] Creating zip files in PHP that are HUGE! Message-ID: There was a discussion about this last week and the consensus was to queue the zip creation and do it "off web server" by a cron job. That will save load on the web server and time for the user to wait properly. That said, if you are in a shared hosting environment, it's possible that your ISP is killing processes that go over a specific load for a specific amount of time. I've had that issue before, even with cron jobs where they would spike the CPU for too long and an automated process would come along and kill them. Might want to check with your ISP on what they are doing with the shared environment. -----Original message----- From: Tris beertastic at gmail.com Date: Mon, 11 Dec 2006 09:12:32 -0800 To: wdvltalk at lists.wdvl.com, thelist at lists.evolt.org Subject: [thelist] Creating zip files in PHP that are HUGE! > I'm added this cool class to my site: > http://www.phpconcept.net/pclzip/index.en.php > > It all works great, until I try and compile a large zip (80Mb) > > SO my question(s) are: > > 1. Anyone uesd tis before on large files > 2. would a Linux server have an inbuilt setting that stops it > processign internally if there was too much load? > 3. Any other ideas for creating large zips on teh fly..?? > > Tris... > > (I'm currently trying to create them and email a link) > -- > > * * Please support the community that supports you. * * > http://evolt.org/help_support_evolt/ > > For unsubscribe and other options, including the Tip Harvester > and archives of thelist go to: http://lists.evolt.org > Workers of the Web, evolt ! > From lists at neptunewebworks.com Mon Dec 11 11:48:12 2006 From: lists at neptunewebworks.com (Max Schwanekamp) Date: Mon, 11 Dec 2006 09:48:12 -0800 Subject: [thelist] Creating zip files in PHP that are HUGE! In-Reply-To: <8b20a7490612110912s47330e0q75552f64fee01232@mail.gmail.com> References: <8b20a7490612110912s47330e0q75552f64fee01232@mail.gmail.com> Message-ID: <457D99DC.7080201@neptunewebworks.com> Tris wrote: > http://www.phpconcept.net/pclzip/index.en.php > It all works great, until I try and compile a large zip (80Mb) > SO my question(s) are: > > 1. Anyone uesd tis before on large files > 2. would a Linux server have an inbuilt setting that stops it > processign internally if there was too much load? > 3. Any other ideas for creating large zips on teh fly..?? 1. Not I 2. Not the server, but your php.ini file does. Do a phpinfo() and check for max_execution and memory_limit values. 3. Use the built-in gnu tools, tar and gzip with the system() function[0]. What error are you getting? What's your PHP version? What exactly are you trying to do? What are you dumping that results in an 80MB file after compression? [0] http://us3.php.net/system -- Max Schwanekamp NeptuneWebworks.com 541-517-9064 From evolt at markgroen.com Mon Dec 11 14:32:51 2006 From: evolt at markgroen.com (evolt at markgroen.com) Date: Mon, 11 Dec 2006 12:32:51 -0800 Subject: [thelist] Creating zip files in PHP that are HUGE! In-Reply-To: <8b20a7490612110912s47330e0q75552f64fee01232@mail.gmail.com> References: <8b20a7490612110912s47330e0q75552f64fee01232@mail.gmail.com> Message-ID: <200612111232.52041.evolt@markgroen.com> On Monday 11 December 2006 09:12, Tris wrote: > I'm added this cool class to my site: > http://www.phpconcept.net/pclzip/index.en.php > > It all works great, until I try and compile a large zip (80Mb) > > SO my question(s) are: > > 1. Anyone uesd tis before on large files Not on a server myself, sorry. > 2. would a Linux server have an inbuilt setting that stops it > processign internally if there was too much load? There is no Linux "server" per se, you probably mean Apache and PHP software running with a Linux computer operating system. This class you have uses php, and the php.ini file has a 30 second default setting for processes and will time out if yours takes longer than that. As well, the default php file handling capacity is 8MB internally, and a 2MB upload limit - both can be changed if you have access to the php.ini file. > 3. Any other ideas for creating large zips on teh fly..?? Create them locally (I use Linux and type "zip", the file name and it's done in a twinkling) with a program that doesn't rely on other software than the computer's operating system and then FTP to the server. Methinks Winzip, the Linux and Mac compression softwares, all of them should be able to do large files locally? Then use the method that Jeff W. mentioned for distribution, having them answer an email to download later if it's a big one.... The other way is the "chunk" the files, beyond what you may be looking for as far as a simple answer, but it is another way to do it; cheers, Mark From beertastic at gmail.com Tue Dec 12 03:21:13 2006 From: beertastic at gmail.com (Tris) Date: Tue, 12 Dec 2006 09:21:13 +0000 Subject: [thelist] Creating zip files in PHP that are HUGE! In-Reply-To: <457D99DC.7080201@neptunewebworks.com> References: <8b20a7490612110912s47330e0q75552f64fee01232@mail.gmail.com> <457D99DC.7080201@neptunewebworks.com> Message-ID: <8b20a7490612120121k1d7617adtc0727d11d2d99702@mail.gmail.com> PHP 5 I'm getting no error, just a 0Kb zip appearing in my DIR (where the full one is supposed to be) and teh page stops loading instantly.. I've tried echoing some strings, just to get some content to the scren before it goes tits up.. but nothing.. But on smaller files, it's fine... We've upped all the max exection time and procesing memory etc, but still nothing... really weird.. on teh plus side, I'm collating all these emails into one, and sending it to our server dude...! :-p On 11/12/06, Max Schwanekamp wrote: > Tris wrote: > > http://www.phpconcept.net/pclzip/index.en.php > > It all works great, until I try and compile a large zip (80Mb) > > SO my question(s) are: > > > > 1. Anyone uesd tis before on large files > > 2. would a Linux server have an inbuilt setting that stops it > > processign internally if there was too much load? > > 3. Any other ideas for creating large zips on teh fly..?? > > 1. Not I > 2. Not the server, but your php.ini file does. Do a phpinfo() and check > for max_execution and memory_limit values. > 3. Use the built-in gnu tools, tar and gzip with the system() function[0]. > > What error are you getting? What's your PHP version? What exactly are > you trying to do? What are you dumping that results in an 80MB file > after compression? > > [0] http://us3.php.net/system > > > -- > Max Schwanekamp > NeptuneWebworks.com > 541-517-9064 > -- > > * * Please support the community that supports you. * * > http://evolt.org/help_support_evolt/ > > For unsubscribe and other options, including the Tip Harvester > and archives of thelist go to: http://lists.evolt.org > Workers of the Web, evolt ! > From anthony at baratta.com Mon Dec 11 12:52:51 2006 From: anthony at baratta.com (Anthony Baratta) Date: Mon, 11 Dec 2006 10:52:51 -0800 Subject: [thelist] Why does this JS code work the way it does??? Message-ID: <7dffad5c29868820178e7e7376247983@baratta.com> Howdy... Normally I can tear into a piece of code and figure out what it's doing. But I can't quite see why this JS function is working exactly the way it does. Here's the function, my comments follow: function CheckNumericKeyStroke() { var strValidKeys = "1234567890"; if (strValidKeys.indexOf(String.fromCharCode(event.keyCode)) == -1) return false; else return true; } Here's how it's put in use: NOTE: I'm working with an IE only environment. Now, here's what I don't understand - why do the backspace and delete keys work in the above text input element? They should comeback with "false" and not "fire". But they do. Why? IE quirk? When I isolate the key code part "String.fromCharCode(event.keyCode)", and send the result as an alert to the browser on the onkeypress event - both backsapce and delete do not fire on that event. Is this by standards design? Any pointers on this behaviour is appreciated. From mwarden at gmail.com Mon Dec 11 12:21:56 2006 From: mwarden at gmail.com (Matt Warden) Date: Mon, 11 Dec 2006 13:21:56 -0500 Subject: [thelist] Coding standards.... [headers] In-Reply-To: <457D78B6.9030102@textmatters.com> References: <457D22B8.3050109@textmatters.com> <10229202.131165834983945.JavaMail.root@redroom.customer.ultraspeed.co.uk> <610592c90612110654iafc71efq735f7e88103f4fa2@mail.gmail.com> <7613e9690612110702p177418e7mc01fe67e9ae106fa@mail.gmail.com> <457D78B6.9030102@textmatters.com> Message-ID: On 12/11/06, Barney Carroll wrote: > Forgive me for saying so... But it's almost as if we're trying to decide > for ourselves what these things might be from a vacuum and the guidance > of w3c... What about books? Seriously, the culture of literature is a > massive thing with plenty of examples - memory can join hands with idle > theory! > >

Putting HTML in context

>

In this chapter we will look at the ontology of markup languages from > SGML to present day XML and XHTML. It is important, before we delve into > a history of the 'ML', to draw attention to two key recent conceptions > which must be made distinct in order to better understand the world > without them.

> >

The Document Object Model

>

The Document Object Model (DOM) took a long while to cement itself as > the key method for browser clients to treat markup, and in turn it took > a strong role in shaping how developers and authors thought of > extensible markup.

>

It is, however, only a recent model for processing markup, and even > at level 1 was not as malleable as we now rely on it being.

> >

DHTML

>

DHTML (Dynamic HyperText Markup Language) has fallen out of use as a > term recently - for good reasons. It is not a markup language proper in > the sense that XHTML, HTML and XML are distinct - rather DHTML > represents the use of HTML, CSS and ECMAscript (more commonly known as > javascript) - but its implication, even of the division of such > languages, is very significant to the way attitudes to markup differed > before and after its inception.

> >

SGML: The beginning of everything

>

The notion of Markup Languages as we know them - that is to say... > All you are doing here is omitting the second-level header "Introduction" in the first section. Your paragraph 1 is not related to your h1 any more than any of the other paragraphs, even though you put that paragraph directly under the h1. There is an h2 there, whether you display it or not. If anything, your example makes it more clear that header levels should follow the typical outline document flow (h1, h2, h3, etc.). -- Matt Warden Cleveland, OH, USA http://mattwarden.com This email proudly and graciously contributes to entropy. From mwarden at gmail.com Tue Dec 12 00:37:53 2006 From: mwarden at gmail.com (Matt Warden) Date: Tue, 12 Dec 2006 01:37:53 -0500 Subject: [thelist] information architecture question (non technical) Message-ID: thelist, This question does not require any specific technical knowledge to answer. Suppose you are trying to present summary level data about tasks. Here's a quick rundown of the characteristics of the tasks: 1) Each task is assigned to a single worker (and only workers -- not supervisors, etc.) 2) Each task has a task type 3) Each task has a number of days until it must be completed (I am calling this "criticality") The idea is to display a summary of tasks assigned to a worker, broken down by task type and criticality level (levels are "past due", "0-10 days", "11-20 days", "20-30 days", "> 30 days"). This is relatively easy. The problem is that there also needs to be a display of the same summary for portions of the organization. So, for example, a manager manages two units of three workers, each led by a supervisor. So: manager |____Unit 1 Supervisor |_____ Worker 1 |_____ Worker 2 |_____ Worker 3 |____Unit 2 Supervisor |_____ Worker 4 |_____ Worker 5 |_____ Worker 6 |____Unit 3 Supervisor |_____ Worker 7 |_____ Worker 8 |_____ Worker 9 The summary for the manger should be the sum of all the summaries of the units below. The summary of each unit should be sum of the summary data for the workers in that unit. The number of levels could be anywhere from 1 to 6 or 7. All of the data does not need to be shown at the same time, but it is important to convey the data in a way that makes sense (i.e., it needs to be clear how Unit 3 Supervisor gets her total). I currently do have a solution, but I hate it. The whole point of this is to offer a quick overview of an individual worker's workload and the workload by organizational area -- my current solution falls short on the "quick overview" part, in my opinion. I'm wondering if I can tap into the creativity of some of you folks and come up with a better design. How would you model this data, hopefully keeping the same structure for both the individual worker view and the organizational view? (This is a hypothetical situation and does not necessarily relate to a current client.) -- Matt Warden Cleveland, OH, USA http://mattwarden.com This email proudly and graciously contributes to entropy. From anthony at baratta.com Tue Dec 12 10:57:30 2006 From: anthony at baratta.com (Anthony Baratta) Date: Tue, 12 Dec 2006 08:57:30 -0800 Subject: [thelist] Why does this JS code work the way it does??? Message-ID: <0bf212db265d26842934c40ffa789691@baratta.com> OK...That makes sense. Anywhere I can find info like that regarding down / up / press? e.g. Browser acts on down versus up, etc. I'm not interested in changing the behaviour because it works as intended - I just didn't want the code to be a "magic happens here" type of function. Besides, knowing the why helps in solving other problems later. ;-) -----Original message----- From: "Lee kowalkowski" lee.kowalkowski at googlemail.com Date: Tue, 12 Dec 2006 05:31:02 -0800 To: "Anthony Baratta" anthony at baratta.com, "thelist at lists.evolt.org" thelist at lists.evolt.org Subject: Re: [thelist] Why does this JS code work the way it does??? > On 11/12/06, Anthony Baratta wrote: > > Normally I can tear into a piece of code and figure out what it's doing. But I can't quite see why this JS function is working exactly the way it does. Here's the function, my comments follow: > > > id="FooElement" > > onkeypress="return CheckNumericKeyStroke();"> > > > > Now, here's what I don't understand - why do the backspace and delete keys work in the above text input element? They should comeback with "false" and not "fire". But they do. Why? IE quirk? > > The browser will receive and act upon the onkeydown events for > Backspace, Enter, F5, etc..., you're waiting for the onkeypress event, > too late I'm afraid. Change it to onkeydown. > > -- > Lee > From RPringle at aurora-il.org Tue Dec 12 10:20:15 2006 From: RPringle at aurora-il.org (Pringle, Ron) Date: Tue, 12 Dec 2006 10:20:15 -0600 Subject: [thelist] PHP Sessions not being passed between directories Message-ID: <4A27645B1A4F4D4181909FA5D213FEF402A0A829@coaexb.coa.aurora-il.org> I friend of mine is having an odd problem that I am trying to help him with. Using PHP 4.3xxx on a sharing hosting environment. I create a session and make some database calls and store session info from the calls, use session_write_close and then redirect to another page in the same directory. I print out the session vars on that redirected page and they show up just fine. >From that page is a link which takes the user to another page in a sub-directory. From that page, I can access the session_id but not any of the session variables. Now if I just create a session variable named Foo and assign it the value of "bar", it gets passed just fine and can be accessed from the page in the sub-directory. Is there something obvious I'm doing wrong? I can supply some pared down sample code if it will help. Ron From jason.handby at corestar.co.uk Tue Dec 12 11:19:34 2006 From: jason.handby at corestar.co.uk (jason.handby) Date: Tue, 12 Dec 2006 17:19:34 -0000 Subject: [thelist] FW: Why does this JS code work the way it does??? Message-ID: <9A50776858A21848A96469CDFCBCDEFF4CE64D@exch-be12.exchange.local> -----Original Message----- From: jason.handby Sent: 12 December 2006 13:38 To: 'thelist at lists.evolt.org' Subject: RE: [thelist] Why does this JS code work the way it does??? Hi Anthony, > Now, here's what I don't understand - why do the backspace and delete > keys work in the above text input element? They should comeback with > "false" and not "fire". But they do. > Why? IE quirk? Internet Explorer doesn't fire the onkeypress event when you press the backspace or delete keys. http://tinyurl.com/dpr2t I don't know if this is a quirk or a feature :-) I think you can use onkeydown to trap a wider range of keys. Jason From jason.handby at corestar.co.uk Tue Dec 12 07:37:58 2006 From: jason.handby at corestar.co.uk (jason.handby) Date: Tue, 12 Dec 2006 13:37:58 -0000 Subject: [thelist] Why does this JS code work the way it does??? In-Reply-To: <7dffad5c29868820178e7e7376247983@baratta.com> Message-ID: <9A50776858A21848A96469CDFCBCDEFF4CE5B0@exch-be12.exchange.local> Hi Anthony, > Now, here's what I don't understand - why do the backspace > and delete keys work in the above text input element? They > should comeback with "false" and not "fire". But they do. > Why? IE quirk? Internet Explorer doesn't fire the onkeypress event when you press the backspace or delete keys. http://tinyurl.com/dpr2t I don't know if this is a quirk or a feature :-) I think you can use onkeydown to trap a wider range of keys. Jason From mwarden at gmail.com Tue Dec 12 07:57:20 2006 From: mwarden at gmail.com (Matt Warden) Date: Tue, 12 Dec 2006 08:57:20 -0500 Subject: [thelist] Why does this JS code work the way it does??? In-Reply-To: <7dffad5c29868820178e7e7376247983@baratta.com> References: <7dffad5c29868820178e7e7376247983@baratta.com> Message-ID: On 12/11/06, Anthony Baratta wrote: > Here's how it's put in use: > > id="FooElement" > onkeypress="return CheckNumericKeyStroke();"> > > NOTE: I'm working with an IE only environment. > > Now, here's what I don't understand - why do the backspace and delete keys work in the above text input element? They should comeback with "false" and not "fire". But they do. Why? IE quirk? > Backspace and delete do not trigger onkeypress. If I recall correctly, you need to use onkeydown in order to listen to these keys. You may have a browser difference, according to docs. IE: "To invoke Press any alphanumeric keyboard key." http://msdn.microsoft.com/workshop/author/dhtml/reference/events/onkeypress.asp Moz: "The keypress event is raised when the user presses a key on the keyboard." http://developer.mozilla.org/en/docs/DOM:element.onkeypress -- Matt Warden Cleveland, OH, USA http://mattwarden.com This email proudly and graciously contributes to entropy. From steven.pierce at gmail.com Tue Dec 12 10:45:08 2006 From: steven.pierce at gmail.com (Steven Pierce) Date: Tue, 12 Dec 2006 08:45:08 -0800 Subject: [thelist] Windows Server Administrator In-Reply-To: <457D1232.2080109@galleryrobinson.com> References: <456B51CF.9060602@galleryrobinson.com> <6e0cf1cc0612102257r6072c4e9t271b80837dec1d56@mail.gmail.com> <457D1232.2080109@galleryrobinson.com> Message-ID: <6e0cf1cc0612120845x73d4753cxfbf229210d81049@mail.gmail.com> Good Luck with that.. On 12/11/06, Hershel Robinson wrote: > > Steven Pierce wrote: > > Try Hostgo.com. I know that they have servers that are Windows. They > do > > have Admins that can help also. > > We are trying RackSpace. So far so good. :) > > Thanks. > > -- > Web Site Construction by Gallery Robinson > http://galleryrobinson.com/ > -- > > * * Please support the community that supports you. * * > http://evolt.org/help_support_evolt/ > > For unsubscribe and other options, including the Tip Harvester > and archives of thelist go to: http://lists.evolt.org > Workers of the Web, evolt ! > From pturmel-webdev at turmel.org Tue Dec 12 14:17:52 2006 From: pturmel-webdev at turmel.org (Phil Turmel) Date: Tue, 12 Dec 2006 15:17:52 -0500 Subject: [thelist] information architecture question (non technical) In-Reply-To: References: Message-ID: <457F0E70.4040302@turmel.org> Matt Warden wrote: > thelist, > > This question does not require any specific technical knowledge to answer. > > Suppose you are trying to present summary level data about tasks. > Here's a quick rundown of the characteristics of the tasks: > > 1) Each task is assigned to a single worker (and only workers -- not > supervisors, etc.) > 2) Each task has a task type > 3) Each task has a number of days until it must be completed (I am > calling this "criticality") > > The idea is to display a summary of tasks assigned to a worker, broken > down by task type and criticality level (levels are "past due", "0-10 > days", "11-20 days", "20-30 days", "> 30 days"). > > This is relatively easy. > > The problem is that there also needs to be a display of the same > summary for portions of the organization. So, for example, a manager > manages two units of three workers, each led by a supervisor. So: > > manager > |____Unit 1 Supervisor > |_____ Worker 1 > |_____ Worker 2 > |_____ Worker 3 > |____Unit 2 Supervisor > |_____ Worker 4 > |_____ Worker 5 > |_____ Worker 6 > |____Unit 3 Supervisor > |_____ Worker 7 > |_____ Worker 8 > |_____ Worker 9 > > The summary for the manger should be the sum of all the summaries of > the units below. The summary of each unit should be sum of the summary > data for the workers in that unit. The number of levels could be > anywhere from 1 to 6 or 7. All of the data does not need to be shown > at the same time, but it is important to convey the data in a way that > makes sense (i.e., it needs to be clear how Unit 3 Supervisor gets her > total). > > I currently do have a solution, but I hate it.... Matt, It seems to me that you should format the summary for each supervisor the same way you format for a single worker, with counts for each type. Then include links up one level, and links to subordinates, as needed. I'm picturing something like this: Heading for mgr/supervisor name & dept/unit name +==========+==========+==========+==========+==========+ | Past Due | Current | 10 - 19 | 20 - 29 | 30+ | +==========+==========+==========+==========+==========+ | ## type | ## type | ## type | ## type | ## type | | | ## type | ## type | ## type | ## type | | | | ## type | ## type | | | | | ## type | | | +==========+==========+==========+==========+==========+ Link up one level Link to subordinate #1 Link to subordinate #2 Link to subordinate #3 Link to subordinate #4 A worker display would be identical, but without subordinate links. And it might list individual keywords for tasks instead of counts of tasks. For a printable report, I would expand slightly, to something like this: Heading for mgr/supervisor name & dept/unit name +==========+==========+==========+==========+==========+ | Subtotal for Subordinate #1 | | Past Due | Current | 10 - 19 | 20 - 29 | 30+ | +==========+==========+==========+==========+==========+ | ## type | ## type | ## type | ## type | ## type | | | ## type | ## type | ## type | ## type | | | | ## type | ## type | | | | | ## type | | | +==========+==========+==========+==========+==========+ | Subtotal for Subordinate #2 | | Past Due | Current | 10 - 19 | 20 - 29 | 30+ | +==========+==========+==========+==========+==========+ | ## type | ## type | ## type | ## type | ## type | | | ## type | ## type | ## type | ## type | | | | ## type | ## type | | | | | ## type | | | +==========+==========+==========+==========+==========+ | Subtotal for Subordinate #3 | | Past Due | Current | 10 - 19 | 20 - 29 | 30+ | +==========+==========+==========+==========+==========+ | ## type | ## type | ## type | ## type | ## type | | | ## type | ## type | ## type | ## type | | | | ## type | ## type | | | | | ## type | | | +==========+==========+==========+==========+==========+ | Subtotal for Subordinate #4 | | Past Due | Current | 10 - 19 | 20 - 29 | 30+ | +==========+==========+==========+==========+==========+ | ## type | ## type | ## type | ## type | ## type | | | ## type | ## type | ## type | ## type | | | | ## type | ## type | | | | | ## type | | | +==========+==========+==========+==========+==========+ +==========+==========+==========+==========+==========+ | Total for dept/unit name | | Past Due | Current | 10 - 19 | 20 - 29 | 30+ | +==========+==========+==========+==========+==========+ | ## type | ## type | ## type | ## type | ## type | | | ## type | ## type | ## type | ## type | | | | ## type | ## type | | | | | ## type | | | +==========+==========+==========+==========+==========+ As for generating these values, the big DBs have big advantages: MSSQL can do recursion with Common Table Expressions [1], and Oracle can return hierarchies with the CONNECT BY clause[2]. There's a patch for PostgeSQL to support the Oracle syntax[3]. Most everyone else is stuck with multiple UNIONS of multiple INNER JOINs, or games with hashcode columns and adjacency sets. Bleah! [1] http://msdn2.microsoft.com/en-us/library/ms175972.aspx [2] http://download-east.oracle.com/docs/cd/B19306_01/server.102/b14200/queries003.htm#i2053935 [3] http://gppl.moonbone.ru/ HTH, Phil From joel at spinhead.com Tue Dec 12 12:23:26 2006 From: joel at spinhead.com (Joel D Canfield) Date: Tue, 12 Dec 2006 10:23:26 -0800 Subject: [thelist] information architecture question (non technical) Message-ID: <72E9FAA171D63B48AAC707C72900E6B4A71991@ireland.spinhead.com> > How would you model this data, hopefully keeping the same structure > for both the individual worker view and the organizational view? off the top of my head: a polar area diagram? http://www.agnesscott.edu/LRIDDLE/WOMEN/nightpiechart.htm joel From dan.mccullough at gmail.com Tue Dec 12 13:44:22 2006 From: dan.mccullough at gmail.com (Dan McCullough) Date: Tue, 12 Dec 2006 14:44:22 -0500 Subject: [thelist] Windows Server Administrator In-Reply-To: <457D1232.2080109@galleryrobinson.com> References: <456B51CF.9060602@galleryrobinson.com> <6e0cf1cc0612102257r6072c4e9t271b80837dec1d56@mail.gmail.com> <457D1232.2080109@galleryrobinson.com> Message-ID: They are very very good, my only problem was their price point was out of line for my business. On 12/11/06, Hershel Robinson wrote: > Steven Pierce wrote: > > Try Hostgo.com. I know that they have servers that are Windows. They do > > have Admins that can help also. > > We are trying RackSpace. So far so good. :) > > Thanks. > > -- > Web Site Construction by Gallery Robinson > http://galleryrobinson.com/ > -- > > * * Please support the community that supports you. * * > http://evolt.org/help_support_evolt/ > > For unsubscribe and other options, including the Tip Harvester > and archives of thelist go to: http://lists.evolt.org > Workers of the Web, evolt ! > From barney at textmatters.com Mon Dec 11 11:37:10 2006 From: barney at textmatters.com (Barney Carroll) Date: Mon, 11 Dec 2006 17:37:10 +0000 Subject: [thelist] Coding standards.... [headers] In-Reply-To: <72E9FAA171D63B48AAC707C72900E6B4A71988@ireland.spinhead.com> References: <72E9FAA171D63B48AAC707C72900E6B4A71988@ireland.spinhead.com> Message-ID: <457D9746.9050503@textmatters.com> Joel D Canfield wrote: > picture me, standing in front of an audience speaking. "Our first item > is blah. Our third item is blah blah blah. There was no second item." But if the purpose of the number following 'h' is to denote sequence, then am I doubly flawed in having 5 '2nd' items? > I'm trying to understand your perspective, but it keeps coming back to > the appearance that you're focusing on, um, appearance. Yes? Not appearance... I know I pointedly disagreed with the notion of hs as a tree schema, and also with the use of the term 'importance'... But those are perhaps the best raw general terms to describe the semantic structure I'm talking about! Visual indication is what I always think about first. This is perhaps a flaw (more a focus, I re-assure myself... hehehe) of my being a visual designer with a background in typography and book design. I'm fascinated with alternative representation, semantics generally and especially screen readers, but yes, semantic sense must at some point be derived through visuals on by behalf. In fact, I think it's very healthy for sighted people to think this way. It gives some cultural reference points that aren't so abstract. If that document were to be a speech, I would deliver my h1 with a marked pause afterwards - it would also be my first utterance except for maybe a personal introduction. My introduction would be in the same key as all the actual nitty gritty. "However I would like to draw your attention to... The Document Object Model" - this would be said with some emphasis, but no way near as much as the h2 to follow, which is clearly separate from the introduction. It makes absolute sense, too. Lists? Bulleted lists should never contain that much free prose. There is a distinct header and two separate phrases beneath. Impossible with ul and ol - possible but incorrect (I believe) with dds - that would be unanimously non-semantic to the core. Believe me, written material in this format has existed for ages and attracts no raised eye-brows. It is essential for educational material especially, I believe. Regards, -- Barney Carroll Text Matters Information design: we help explain things using language | design | systems | process improvement _______________________________________________________ phone +44 (0)118 918 2382 email barney at textmatters.com web http://www.textmatters.com From joel at spinhead.com Tue Dec 12 17:11:26 2006 From: joel at spinhead.com (Joel D Canfield) Date: Tue, 12 Dec 2006 15:11:26 -0800 Subject: [thelist] Coding standards.... [headers] Message-ID: <72E9FAA171D63B48AAC707C72900E6B4A71997@ireland.spinhead.com> > Joel D Canfield wrote: > > picture me, standing in front of an audience speaking. "Our > first item > > is blah. Our third item is blah blah blah. There was no > second item." > > But if the purpose of the number following 'h' is to denote sequence, > then am I doubly flawed in having 5 '2nd' items? eh, poorly worded on my part. should say more like "our most important item is blah; our third most important item is blah blah blah. there is no second most important item." which isn't logical. > > I'm trying to understand your perspective, but it keeps > coming back to > > the appearance that you're focusing on, um, appearance. Yes? > > Not appearance... I know I pointedly disagreed with the > notion of hs as > a tree schema, and also with the use of the term 'importance'... But > those are perhaps the best raw general terms to describe the semantic > structure I'm talking about! So, are you agreeing in spite of yourself, or still disagreeing? ;) I'm enjoying this discussion; forces me to think about what I think, and I'm a good meta-mental mood right now. joel From barney at textmatters.com Tue Dec 12 04:31:08 2006 From: barney at textmatters.com (Barney Carroll) Date: Tue, 12 Dec 2006 10:31:08 +0000 Subject: [thelist] Coding standards.... [headers] In-Reply-To: <457DC387.7090303@hyperlinkage.com> References: <457D22B8.3050109@textmatters.com> <10229202.131165834983945.JavaMail.root@redroom.customer.ultraspeed.co.uk> <610592c90612110654iafc71efq735f7e88103f4fa2@mail.gmail.com> <7613e9690612110702p177418e7mc01fe67e9ae106fa@mail.gmail.com> <457DC387.7090303@hyperlinkage.com> Message-ID: <457E84EC.3000601@textmatters.com> tim wrote: > Although I don't agree with your strict usage of heading tags for all > situations, I can see how it could be a valuable addition to a set of > coding standards. I think we got off valuable additions to coding standards a long time ago! Hehehe. Regards, Barney From matthew at chickenskinners.com Tue Dec 12 03:35:08 2006 From: matthew at chickenskinners.com (Matthew Pollard) Date: Tue, 12 Dec 2006 09:35:08 -0000 Subject: [thelist] Workaround for setAttribute bug in IE7? Message-ID: <000701c71dd0$d08cf410$60640a0a@1055bb8af96945c> Ryan I amp pretty sure you can access the colspan as a property rather than an attribute - if that makes sense. Something along the lines of: newCell.colSpan = 8; I think you should really access the id in the same way: newCell.id = blah; Cheers Matthew From mwarden at gmail.com Tue Dec 12 17:27:54 2006 From: mwarden at gmail.com (Matt Warden) Date: Tue, 12 Dec 2006 18:27:54 -0500 Subject: [thelist] Coding standards.... [headers] In-Reply-To: <457D9746.9050503@textmatters.com> References: <72E9FAA171D63B48AAC707C72900E6B4A71988@ireland.spinhead.com> <457D9746.9050503@textmatters.com> Message-ID: On 12/11/06, Barney Carroll wrote: > Joel D Canfield wrote: > > picture me, standing in front of an audience speaking. "Our first item > > is blah. Our third item is blah blah blah. There was no second item." > > But if the purpose of the number following 'h' is to denote sequence, > then am I doubly flawed in having 5 '2nd' items? It is clearly not sequence. I think Joel just made a poor example and indeed was trying to convey depth rather than sequence. > If that document were to be a speech, I would deliver my h1 with a > marked pause afterwards - it would also be my first utterance except for > maybe a personal introduction. My introduction would be in the same key > as all the actual nitty gritty. "However I would like to draw your > attention to... The Document Object Model" - this would be said with > some emphasis, but no way near as much as the h2 to follow, which is > clearly separate from the introduction. It makes absolute sense, too. This is all presentation. It has nothing to do with the document structure, only how you're deciding to deliver it. > Lists? Bulleted lists should never contain that much free prose. There > is a distinct header and two separate phrases beneath. Impossible with > ul and ol - possible but incorrect (I believe) with dds - that would be > unanimously non-semantic to the core. Bullets are again entirely presentational. You use lists when you have a list of things, regardless of whether you believe it makes "sense" to show a bullet next to it. If I want to present a list of the 10 best paragraphs ever written, this would be an ordered list. I would probably hide the bullet, but that is presentation decision. H1-H6 indicate depth in a document. Saying you can skip a header is akin to saying you can do this: 1 Sandwiches 1._.1 Introductory point 1 1._.2 Introductory point 2 1.2 Burgers 1.2.1 Cheeseburgers are... 1.2.2 Hamburgers are... 1.3 Subs ... What the heck is 1._.1?! There is a 1.1. If you choose to hide it, that is a presentation decision. -- Matt Warden Cleveland, OH, USA http://mattwarden.com This email proudly and graciously contributes to entropy. From lee.kowalkowski at googlemail.com Tue Dec 12 04:13:36 2006 From: lee.kowalkowski at googlemail.com (Lee kowalkowski) Date: Tue, 12 Dec 2006 10:13:36 +0000 Subject: [thelist] Workaround for setAttribute bug in IE7? In-Reply-To: References: <457CF780.8060004@ryanrushton.com> <457DE052.9000809@bitstorm.org> Message-ID: <610592c90612120213i1037cf1ds2c056b52267edafa@mail.gmail.com> On 11/12/06, Matt Warden wrote: > Why would I think that setting > .colSpan object property in IE would create an attribute node in the > DOM? It is a really dumb idea, ...because you're invoking an assignment operator on a DOM object in the first place? In JavaScript, the assignment operator often creates new properties: window.foo = "bar"; - If I thought that wasn't capable of creating foo if it didn't exist, I ought to read the ECMAScript language specification again. If the browser is exposing its DOM via a convenient JavaScript API, I'd have every right to expect to be able to create attributes using a straight-forward assignment operator. I don't think it's a dumb idea in the slightest. I personally find elem.colSpan more maintainable than elem.setAttribute("colspan", ...). With a few exceptions, the choice between setAttribute and an assignment is developer preference. Incidentally, the colSpan attribute node would not be created either way because it will already exist. Table cells are given a default colSpan attribute of "1", the browser creates a lot of DOM attributes by default. > especially when you consider what > happens when one does: > > foo.colSpan = null; > > Does this remove an existing attribute node (doubt it)? Does it just > set its value to null (probably)? If the attribute node doesn't > already exist, does it create it and set it to null (probably)? None of the above in IE - you get an invalid argument exception. Now it *is* a really dumb idea to expect assigning an property to null will delete the property! -- Lee From Ron.Luther at hp.com Tue Dec 12 14:00:32 2006 From: Ron.Luther at hp.com (Luther, Ron) Date: Tue, 12 Dec 2006 14:00:32 -0600 Subject: [thelist] information architecture question (non technical) In-Reply-To: Message-ID: <1FF0647605921940BCB22C2A454AD7CE01194728@cceexc17.americas.cpqcorp.net> Matt Warden asked about IA: >>This question does not require any specific technical knowledge to answer. Whew! Good thing for me! ;-) >>Suppose you are trying to present summary level data about [stuff] >>How would you model this data, hopefully keeping the same >>structure for both the individual worker view and the organizational view? Hi Matt, Well ... I've done some very similar kinds of reporting and the way I've set that up is to have my hierarchy data set off in a second table. You've got your 'transactional data'; worker number, job/task number, due date (or number of days). [If you really have 'task type' I would argue that should be set that off in a separate hierarchy / attribute table and linked in as a categorical control. E.g Select all 'painting' tasks.] I would set the data rollup hierarchy off in a separate table. Worker --> supervisor --> department --> plant --> division --> State / Province --> Region --> Planet --> Galaxy ... Etc. Now set up your report as if reporting for an individual worker bee. Link the worker bee back into the hierarchy table and add report controls to let users report at whatever level they want. Done. [Visually folks often use a collapsed/expanded folder control where you initialize the report at 'all' and let users drill into the lower levels with the control. E.g. - All Apparel - Sweaters - Overcoats - Shoes - Running Shoes - Hiking Shoes - Boat Shoes + Shirts] Same kind of structure works well for geographic hierarchies, or product hierarchies, or organizational hierarchies. {This is similar to past due purchase order reporting where you can either link the individual buyer back to a 'buyer org' to a supervisor to an organization / location / etc ....... or you can link the part number back to a product or commodity hierarchy. Past due sales order reporting has similar hierarchical linkages.} Now ... If you have LOTS of data and really want your reporting to fly ... {I wasn't completely sure if you were asking about structure or about speed} ... then you might want to think about preprocessing an aggregate table and running your reporting off that. E.g. at the end of the day sum up and write out a single 'level zero' record that has sums for *all* 0-10 day late tasks, 11-20 day late tasks, etc. Then write out the 'level one' records by plant or division. Then write out the 'level two' records by supervisor and the 'level three' records by worker. Make that a control in the report and things will move pretty fast. (A user wanting the world-wide view is retrieving exactly one record ... even with tricky caching strategies it don't get much faster than that!) HTH, RonL. * I used to get into some _really_ neat arguments with a co-worker in Cali about 'Federated' versus 'Non-Federated' data models and whether that hierarchical reference data should be managed in a stand alone environment or whether it should flow through the transactional systems along with the raw data. Fun! From lee.kowalkowski at googlemail.com Tue Dec 12 07:31:02 2006 From: lee.kowalkowski at googlemail.com (Lee kowalkowski) Date: Tue, 12 Dec 2006 13:31:02 +0000 Subject: [thelist] Why does this JS code work the way it does??? In-Reply-To: <7dffad5c29868820178e7e7376247983@baratta.com> References: <7dffad5c29868820178e7e7376247983@baratta.com> Message-ID: <610592c90612120531w90655b8ge31b3a81012e676b@mail.gmail.com> On 11/12/06, Anthony Baratta wrote: > Normally I can tear into a piece of code and figure out what it's doing. But I can't quite see why this JS function is working exactly the way it does. Here's the function, my comments follow: > id="FooElement" > onkeypress="return CheckNumericKeyStroke();"> > > Now, here's what I don't understand - why do the backspace and delete keys work in the above text input element? They should comeback with "false" and not "fire". But they do. Why? IE quirk? The browser will receive and act upon the onkeydown events for Backspace, Enter, F5, etc..., you're waiting for the onkeypress event, too late I'm afraid. Change it to onkeydown. -- Lee From tim at hyperlinkage.com Tue Dec 12 19:37:27 2006 From: tim at hyperlinkage.com (tim) Date: Wed, 13 Dec 2006 01:37:27 +0000 Subject: [thelist] Why does this JS code work the way it does??? In-Reply-To: <7dffad5c29868820178e7e7376247983@baratta.com> References: <7dffad5c29868820178e7e7376247983@baratta.com> Message-ID: <457F5957.5000103@hyperlinkage.com> Anthony Baratta wrote: > var strValidKeys = "1234567890"; > if (strValidKeys.indexOf(String.fromCharCode(event.keyCode)) == -1) I desperately hope that you have a noscript and non-keyboard alternative to the above code listing. Also, which keys are you hoping to catch? indexOf looks like the wrong method to use for this. > if ( ...stuff... == -1) > return false; > else > return true; Grrrrrrr...... You've used four lines of code which could have simply been: return ! (strValidKeys.indexOf(String.fromCharCode(event.keyCode)) == -1) -- www.hyperlinkage.com From mwarden at gmail.com Tue Dec 12 20:02:04 2006 From: mwarden at gmail.com (Matt Warden) Date: Tue, 12 Dec 2006 21:02:04 -0500 Subject: [thelist] Why does this JS code work the way it does??? In-Reply-To: <457F5957.5000103@hyperlinkage.com> References: <7dffad5c29868820178e7e7376247983@baratta.com> <457F5957.5000103@hyperlinkage.com> Message-ID: On 12/12/06, tim wrote: > You've used four lines of code which could have simply been: > > return ! (strValidKeys.indexOf(String.fromCharCode(event.keyCode)) == -1) ... which saves you what, exactly? Certainly not smacks upside the head. You've cut the lines of code down from 5 to 2. If I have any complaints about the code Anthony brought for discussion, it's that it's too short. I would want (formatting only -- ignoring the programming choices): function CheckNumericKeyStroke() { var strValidKeys = "1234567890"; // if there keyCode is not one of the numbers in validkeys... if (strValidKeys.indexOf(String.fromCharCode(event.keyCode)) == -1) { return false; // do not allow the keystroke } else { return true; // allow the keystroke } } Remember: every line you save for the sake of saving it almost certainly costs your company money in maintenance costs. This is a concept some programmers never seem to get and they instead focus on filesize (which costs a company next to nothing) and execution speed (which costs the company very little in 90% of the cases). -- Matt Warden Cleveland, OH, USA http://mattwarden.com This email proudly and graciously contributes to entropy. From tim at hyperlinkage.com Tue Dec 12 20:03:18 2006 From: tim at hyperlinkage.com (tim) Date: Wed, 13 Dec 2006 02:03:18 +0000 Subject: [thelist] Coding standards.... [headers] In-Reply-To: References: <72E9FAA171D63B48AAC707C72900E6B4A71988@ireland.spinhead.com> <457D9746.9050503@textmatters.com> Message-ID: <457F5F66.6090304@hyperlinkage.com> Matt Warden wrote: > If I want to present a list of the 10 > best paragraphs ever written, this would be an ordered list. I would > probably hide the bullet, but that is presentation decision. The bullet? In an ordered list? If it was a "top 10" list it would entirely depend on if your list was ordered or unordered - i.e. sequential or not. Of course, the actual display is a question of style, but UL vs OL is a question of semantics. Using CSS to change the display of a list would require some text making the meaning clear. I've enjoyed the discussion. Interestingly XHTML 2 provides the following example, which will upset us all: "The series h1 h2 h1 is acceptable, while h1 h3 h1 is not, since the heading level h2 has been skipped." Ohhhhh.... Tim -- www.hyperlinkage.com From mwarden at gmail.com Tue Dec 12 20:14:51 2006 From: mwarden at gmail.com (Matt Warden) Date: Tue, 12 Dec 2006 21:14:51 -0500 Subject: [thelist] Coding standards.... [headers] In-Reply-To: <457F5F66.6090304@hyperlinkage.com> References: <72E9FAA171D63B48AAC707C72900E6B4A71988@ireland.spinhead.com> <457D9746.9050503@textmatters.com> <457F5F66.6090304@hyperlinkage.com> Message-ID: On 12/12/06, tim wrote: > "The series h1 h2 h1 is acceptable, while h1 h3 h1 is not, since the > heading level h2 has been skipped." This will upset half of us. This is exactly what the other half of us have been saying. h1 h2 h1 h2 is ok. h1 h3 makes no sense. Do you have a link, btw? -- Matt Warden Cleveland, OH, USA http://mattwarden.com This email proudly and graciously contributes to entropy. From tim at hyperlinkage.com Tue Dec 12 20:19:41 2006 From: tim at hyperlinkage.com (tim) Date: Wed, 13 Dec 2006 02:19:41 +0000 Subject: [thelist] Why does this JS code work the way it does??? In-Reply-To: References: <7dffad5c29868820178e7e7376247983@baratta.com> <457F5957.5000103@hyperlinkage.com> Message-ID: <457F633D.5070101@hyperlinkage.com> Matt Warden wrote: > if ( ...some booleans statement... ) { > return false; > } > else { > return true; > } This is a big frustration for many many many many many many programmers. You are evaluating a boolean only to return a boolean value, so just return the result of the original comparison. -- www.hyperlinkage.com From tim at hyperlinkage.com Tue Dec 12 20:27:22 2006 From: tim at hyperlinkage.com (tim) Date: Wed, 13 Dec 2006 02:27:22 +0000 Subject: [thelist] Coding standards.... [headers] In-Reply-To: References: <72E9FAA171D63B48AAC707C72900E6B4A71988@ireland.spinhead.com> <457D9746.9050503@textmatters.com> <457F5F66.6090304@hyperlinkage.com> Message-ID: <457F650A.5060703@hyperlinkage.com> Matt Warden wrote: > This will upset half of us. This is exactly what the other half of us > have been saying. > > h1 > h2 > h1 > h2 I'm pretty sure that some have been against more than a single H1 tag per document. > Do you have a link, btw? http://www.w3.org/TR/xhtml2/mod-structural.html#sec_8.5. And yes, this does mention that skipping levels is "considered to be" bad practice. However, it does not use the wording (must or should) that usually appears in a W3C document. Hence, I argue that it's still an option for our style guides rather than set in stone. Tim -- www.hyperlinkage.com From ftarzwell at fayec.com Tue Dec 12 20:28:37 2006 From: ftarzwell at fayec.com (Flavia Tarzwell (FayeC)) Date: Tue, 12 Dec 2006 21:28:37 -0500 Subject: [thelist] Webserver separate from mail server? Message-ID: <457F6555.90505@fayec.com> Can anybody tell me if it's possible to run a mail server separately from the web server? A friend of mine has a client who is planning to host their own mail server (issues with the current host) but they are not planning on moving the site to their own web server. Ie just the mail that would be separate from the rest of the site. I have no concrete information on if this is even possible to accomplish. And I thought that since the domain is pointing to the web server's IP that it might be a little more complicated to set up the mail to be redirectd to a different IP?? Can someone clarify that (if possible or not?)? Thanks in advance, FayeC From tim at hyperlinkage.com Tue Dec 12 20:40:56 2006 From: tim at hyperlinkage.com (tim) Date: Wed, 13 Dec 2006 02:40:56 +0000 Subject: [thelist] Webserver separate from mail server? In-Reply-To: <457F6555.90505@fayec.com> References: <457F6555.90505@fayec.com> Message-ID: <457F6838.10202@hyperlinkage.com> Flavia Tarzwell (FayeC) wrote: > Can anybody tell me if it's possible to run a mail server separately > from the web server? You need full control over the site's DNS records. Usually a domain is configured with an "A" record, which points the name to an IP. What you need to do is create (or change) an "MX" record to point to the IP address of the mail server. MX means "mail exchange". If you don't have DNS control then ask the hosting company to: "Switch the MX record to [...IP ADDRESS...]" Hope this helps, Tim -- www.hyperlinkage.com From ftarzwell at fayec.com Tue Dec 12 20:28:37 2006 From: ftarzwell at fayec.com (Flavia Tarzwell (FayeC)) Date: Tue, 12 Dec 2006 21:28:37 -0500 Subject: [thelist] Webserver separate from mail server? Message-ID: <457F6555.90505@fayec.com> Can anybody tell me if it's possible to run a mail server separately from the web server? A friend of mine has a client who is planning to host their own mail server (issues with the current host) but they are not planning on moving the site to their own web server. Ie just the mail that would be separate from the rest of the site. I have no concrete information on if this is even possible to accomplish. And I thought that since the domain is pointing to the web server's IP that it might be a little more complicated to set up the mail to be redirectd to a different IP?? Can someone clarify that (if possible or not?)? Thanks in advance, FayeC From volkan.ozcelik at gmail.com Tue Dec 12 23:19:59 2006 From: volkan.ozcelik at gmail.com (=?ISO-8859-9?Q?VOLKAN_=D6Z=C7EL=DDK?=) Date: Wed, 13 Dec 2006 07:19:59 +0200 Subject: [thelist] Proper commenting and code maintainance [was: Why does this JS code work the way it does???] Message-ID: Then let us create a hybrid of both views: // if there keyCode is not one of the numbers in validkeys // do not allow the keystroke // otherwise allow it: return (strValidKeys.indexOf(String.fromCharCode(event.keyCode)) == -1); And then compress your source with somehting like packer [1], js crunchinator [2] or EasyPack [3] [1] http://dean.edwards.name/packer*/* [2] http://www.brainjar.com/js/crunch/ [3] http://www.codeproject.com/useritems/EasyPack.asp (latest source code and binaries on http://www.sarmal.com/sardalya/Samples.aspx) HTH, -- Volkan Ozcelik +>Yep! I'm blogging! : http://www.volkanozcelik.com/cre8/blog/ +> Going solo in Turkish: http://www.volkanozcelik.com/donkisot/ +> My projects/studies/trials/errors : http://www.sarmal.com/ +> Sardalya JavaScript Library: http://www.sarmal.com/sardalya/ From volkan.ozcelik at gmail.com Tue Dec 12 23:22:32 2006 From: volkan.ozcelik at gmail.com (=?ISO-8859-9?Q?VOLKAN_=D6Z=C7EL=DDK?=) Date: Wed, 13 Dec 2006 07:22:32 +0200 Subject: [thelist] Proper commenting and code maintainance [was: Why does this JS code work the way it does???] In-Reply-To: References: Message-ID: Sorry, I forgot to quote the original post " Matt Warden wrote: > if ( ...some booleans statement... ) { > return false; > } > else { > return true; > } This is a big frustration for many many many many many many programmers. " On 12/13/06, VOLKAN ?Z?EL?K wrote: > > > Then let us create a hybrid of both views: > > // if there keyCode is not one of the numbers in validkeys > // do not allow the keystroke > // otherwise allow it: > return (strValidKeys.indexOf(String.fromCharCode (event.keyCode)) == -1); > > And then compress your source with somehting like packer [1], js > crunchinator [2] or EasyPack [3] > > > [1] http:// dean.edwards.name/packer*/* > [2] http://www.brainjar.com/js/crunch/ > [3] http:// www.codeproject.com/useritems/EasyPack.asp > (latest source code and binaries on http://www.sarmal.com/sardalya/Samples.aspx > ) > > Cheers, Volkan. From skquinn at speakeasy.net Wed Dec 13 00:07:12 2006 From: skquinn at speakeasy.net (Shawn K. Quinn) Date: Wed, 13 Dec 2006 00:07:12 -0600 Subject: [thelist] Webserver separate from mail server? In-Reply-To: <457F6555.90505@fayec.com> References: <457F6555.90505@fayec.com> Message-ID: <1165990032.8537.18.camel@afterburner> On Tue, 2006-12-12 at 21:28 -0500, Flavia Tarzwell (FayeC) wrote: > Can anybody tell me if it's possible to run a mail server separately > from the web server? It's definitely possible, and in some cases it's even unavoidable when a Web host does not handle e-mail. I'll explain with a real-life example ("$" is a GNU bash command prompt, edited to remove extra info): $ host evolt.org evolt.org has address 67.19.100.194 evolt.org mail is handled by 10 mail.evolt.org. evolt.org mail is handled by 20 relay.evolt.org. $ host mail.evolt.org mail.evolt.org is an alias for aa.dallas.tx.us.evolt.org. aa.dallas.tx.us.evolt.org has address 67.19.100.194 mail.evolt.org is an alias for aa.dallas.tx.us.evolt.org. mail.evolt.org is an alias for aa.dallas.tx.us.evolt.org. $ host www.evolt.org www.evolt.org is an alias for aa.dallas.tx.us.evolt.org. aa.dallas.tx.us.evolt.org has address 67.19.100.194 www.evolt.org is an alias for aa.dallas.tx.us.evolt.org. www.evolt.org is an alias for aa.dallas.tx.us.evolt.org. Currently, mail.evolt.org and www.evolt.org happen to point to the same computer. However, this need not be the case. The MX record ("mail is handled by" above) could just as easily point to just about any other machine on the planet (assuming sufficient bandwidth and a static IP address). It is worth noting that in the past, full-time Internet connections were not as common as they are today, and this is the MX record's primary reason for existence. Note also, lists.evolt.org points to a different computer, so if you were to offer mailing lists under a different hostname in the same domain. In fact, since I know I owe at least two tips: It's quite possible to register a domain name and use it only for e-mail or other purposes besides a Web site, so don't assume a domain is not in active use just because the host "www" under it doesn't resolve. -- Shawn K. Quinn From delusr at gmail.com Wed Dec 13 00:14:35 2006 From: delusr at gmail.com (del usr) Date: Wed, 13 Dec 2006 16:14:35 +1000 Subject: [thelist] Webserver separate from mail server? In-Reply-To: <1165990032.8537.18.camel@afterburner> References: <457F6555.90505@fayec.com> <1165990032.8537.18.camel@afterburner> Message-ID: <57da305a0612122214x3411a421mea7af29a9de96a00@mail.gmail.com> Not to sound like a trouble maker but with the number of issues relating to hosting a web site/email server, if you don't know the answer to this I would get another company to do your system administration. There are just to many things that can go wrong. Try setting up a few old boxes and play around with them. On 12/13/06, Shawn K. Quinn wrote: > > On Tue, 2006-12-12 at 21:28 -0500, Flavia Tarzwell (FayeC) wrote: > > Can anybody tell me if it's possible to run a mail server separately > > from the web server? > > It's definitely possible, and in some cases it's even unavoidable when a > Web host does not handle e-mail. I'll explain with a real-life example > ("$" is a GNU bash command prompt, edited to remove extra info): > > $ host evolt.org > evolt.org has address 67.19.100.194 > evolt.org mail is handled by 10 mail.evolt.org. > evolt.org mail is handled by 20 relay.evolt.org. > $ host mail.evolt.org > mail.evolt.org is an alias for aa.dallas.tx.us.evolt.org. > aa.dallas.tx.us.evolt.org has address 67.19.100.194 > mail.evolt.org is an alias for aa.dallas.tx.us.evolt.org. > mail.evolt.org is an alias for aa.dallas.tx.us.evolt.org. > $ host www.evolt.org > www.evolt.org is an alias for aa.dallas.tx.us.evolt.org . > aa.dallas.tx.us.evolt.org has address 67.19.100.194 > www.evolt.org is an alias for aa.dallas.tx.us.evolt.org. > www.evolt.org is an alias for aa.dallas.tx.us.evolt.org. > > Currently, mail.evolt.org and www.evolt.org happen to point to the same > computer. However, this need not be the case. The MX record ("mail is > handled by" above) could just as easily point to just about any other > machine on the planet (assuming sufficient bandwidth and a static IP > address). It is worth noting that in the past, full-time Internet > connections were not as common as they are today, and this is the MX > record's primary reason for existence. > > Note also, lists.evolt.org points to a different computer, so if you > were to offer mailing lists under a different hostname in the same > domain. In fact, since I know I owe at least two tips: > > > It's quite possible to register a domain name and use it only for e-mail > or other purposes besides a Web site, so don't assume a domain is not in > active use just because the host "www" under it doesn't resolve. > > > -- > Shawn K. Quinn > > -- > > * * Please support the community that supports you. * * > http://evolt.org/help_support_evolt/ > > For unsubscribe and other options, including the Tip Harvester > and archives of thelist go to: http://lists.evolt.org > Workers of the Web, evolt ! > -- PRIVILEGED - PRIVATE AND CONFIDENTIAL This email and any files transmitted with it are intended solely for the use of the addressee(s) and may contain information which is confidential or privileged. If you receive this email and you are not the addressee(s) [or responsible for delivery of the email to the addressee(s)], please disregard the contents of the email, delete the email and notify the author immediately. From jens.brueckmann at gmail.com Wed Dec 13 02:27:23 2006 From: jens.brueckmann at gmail.com (Jens Brueckmann) Date: Wed, 13 Dec 2006 09:27:23 +0100 Subject: [thelist] Coding standards.... [headers] In-Reply-To: <457F650A.5060703@hyperlinkage.com> References: <72E9FAA171D63B48AAC707C72900E6B4A71988@ireland.spinhead.com> <457D9746.9050503@textmatters.com> <457F5F66.6090304@hyperlinkage.com> <457F650A.5060703@hyperlinkage.com> Message-ID: I would like to draw your attention to a Priority 2 Accessibility checkpoint of WCAG, which clearly states [1]: "HTML, H2 elements should follow H1 elements, H3 elements should follow H2 elements, etc. Content developers should not 'skip' levels (e.g., H1 directly to H3)." [1] http://www.w3.org/TR/WCAG10-HTML-TECHS/#document-headers Cheers, jens. -- Jens Brueckmann http://www.yalf.de From barney at textmatters.com Wed Dec 13 05:38:01 2006 From: barney at textmatters.com (Barney Carroll) Date: Wed, 13 Dec 2006 11:38:01 +0000 Subject: [thelist] Coding standards.... [headers] In-Reply-To: References: <72E9FAA171D63B48AAC707C72900E6B4A71988@ireland.spinhead.com> <457D9746.9050503@textmatters.com> <457F5F66.6090304@hyperlinkage.com> <457F650A.5060703@hyperlinkage.com> Message-ID: <457FE619.9040208@textmatters.com> Jens Brueckmann wrote: > "HTML, H2 elements should follow H1 elements, H3 elements should > follow H2 elements, etc. Content developers should not 'skip' levels > (e.g., H1 directly to H3)." > > [1] http://www.w3.org/TR/WCAG10-HTML-TECHS/#document-headers > > Cheers, > > jens. This seals it as far as best practice is concerned! - Authors as I imagine them might feel constrained by this tiered tree-branch pattern that all web documents should be able to be represented by. - Information architects doubtless love this idea, but then they have no business inside the document. - 'Content developers', on the other hand... (heeheehee) The truth is, the w3 has the best intentions here, and I think it'd just be counter-productive for me to seriously advocate a post-modernist approach. What I'm ultimately concerned with is that notions of what is clean and sensible and ordered in terms of a coded document's structure are being used as key guides as to how a human document should be arranged and labeled. However, things aren't as bad as I might make them out; if the 'content developer' does not have a distinct notion of how to go about writing, compiling or editing a document, this is a good generic piece of advice. In fact, in any kind of authorship apart from poetry, fiction and post-modern prose, the ability to adhere to this model is compulsory for the sake of everything (except actually reading the work!). Those people who have a strong enough notion of what they're writing to go against this advice should do so of that conviction and not from a general lack of guidance. I still find it a bit tragic that the w3 should take it upon itself to tell us how to write, and that as a consequence the shape of the web to come is made that bit more uniform (or 'standardised's evil twin) - essentially this dictates what kind of content can appear on the web. It has no practical applications, and is not, in the purest sense, a coding issue. Great thread guys, gave me a lot of insights and allowed me to elucidate my own ideas that bit more. Regards, Barney From austin at dotmail.co.uk Wed Dec 13 05:48:38 2006 From: austin at dotmail.co.uk (Austin Harris) Date: Wed, 13 Dec 2006 11:48:38 +0000 (GMT) Subject: [thelist] Coding standards.... [headers] In-Reply-To: <457FE619.9040208@textmatters.com> Message-ID: <26675601.211166010518335.JavaMail.root@redroom.customer.ultraspeed.co.uk> >>Jens Brueckmann wrote: >> "HTML, H2 elements should follow H1 elements, H3 elements should >> follow H2 elements, etc. Content developers should not 'skip' levels >> (e.g., H1 directly to H3)." >> >> [1] http://www.w3.org/TR/WCAG10-HTML-TECHS/#document-headers >> >> Cheers, >> >> jens. >This seals it as far as best practice is concerned! >- Authors as I imagine them might feel constrained by this tiered >tree-branch pattern that all web documents should be able to be >represented by. >- Information architects doubtless love this idea, but then they have no >business inside the document. >- 'Content developers', on the other hand... (heeheehee) Not sure how to take this last comment, (as a contract developer) (hohoho, to add a bit of christmas cheer ;) I'm sure project managers / full time developers are sick of me banging on about following standards but then that is what they pay me for... It's always one that is tricky to get right IMHO. I seem to spend most of my time "cleaning" other peoples code to get it up to scratch before I even get to things like this - STILL taking tables out of layouts on very big sites... Austin (Great thread btw). From lee.kowalkowski at googlemail.com Wed Dec 13 06:06:35 2006 From: lee.kowalkowski at googlemail.com (Lee kowalkowski) Date: Wed, 13 Dec 2006 12:06:35 +0000 Subject: [thelist] Coding standards.... [headers] In-Reply-To: <457FE619.9040208@textmatters.com> References: <72E9FAA171D63B48AAC707C72900E6B4A71988@ireland.spinhead.com> <457D9746.9050503@textmatters.com> <457F5F66.6090304@hyperlinkage.com> <457F650A.5060703@hyperlinkage.com> <457FE619.9040208@textmatters.com> Message-ID: <610592c90612130406m47559a4fw80f07e1c529ef88c@mail.gmail.com> On 13/12/06, Barney Carroll wrote: > Jens Brueckmann wrote: > > [1] http://www.w3.org/TR/WCAG10-HTML-TECHS/#document-headers > This seals it as far as best practice is concerned! > > - Authors as I imagine them might feel constrained by this tiered > tree-branch pattern that all web documents should be able to be > represented by. Well they shouldn't, there's nothing wrong with having different classes of h2 headings. It must be OK, the accessibility guidelines themselves have [1, above]. Basically, if it's a subtopic heading, then increment the level, if it's a new subject heading, use the previous level (or higher). If you need headings of the same level to look different, classify your headings. If you wish to avoid multiple H1s, then consider a creating a new page. -- Lee From mark.macinnes at awakeanddreaming.net Wed Dec 13 07:27:38 2006 From: mark.macinnes at awakeanddreaming.net (Mark A. MacInnes) Date: Wed, 13 Dec 2006 13:27:38 +0000 Subject: [thelist] Linux Dedicated Server Message-ID: <457FFFCA.4070304@awakeanddreaming.net> Hello, I'm sure I remember reading some posts about this recently, but I can't find in the archives - I'm in a rush so excuse me if I missed them. Can people recommend some hosts of linux dedicated servers? Requirements: - Preferably UK-based, but open to especially good non-UK-based - Excellent and fast customer support, via 'phone, e-mail and online. - PHP & MySQL are the only techs that *have* to work - Ideally they would be willing to take over most of the management, but if not then support would need to help us out. - Unfortunately can't say whether 1 site or multiple sites - that's something that hasn't been decided as of yet. I know my requirements are fairly open, but then we're looking to get more info in order to make some important decisions. Thanks in advance! Mark A. MacInnes -- - Awake & Dreaming - http://www.awakeanddreaming.net From beertastic at gmail.com Wed Dec 13 09:25:38 2006 From: beertastic at gmail.com (Tris) Date: Wed, 13 Dec 2006 15:25:38 +0000 Subject: [thelist] Adding 15 mins to PHP time... Message-ID: <8b20a7490612130725r6e1171bagf8640446984cc2e0@mail.gmail.com> I'm creating a timer for todays lucky client... They wanna hit start and have a timer on screen across all pages.. No prob doing that in javascript, so I'm gonna create a session called timer with the date value: date("H:i:s"); Giving: 15:42:34 etc.. (the precise users start time) What I need to do is on each page, check the current server time, and if it's 15 mis higher than my session, redirect the user to a thank you page and destroythetimer session.. soooo... How can I add 15 mins to a date function? (I'm also gonna have to fee different times to the javasript ticker, but that'll be easiy once I kno whow to knock off time from um, well, time :-p ) Thoughts? Tris... From caseyc at IntelliSoftmn.com Wed Dec 13 09:39:08 2006 From: caseyc at IntelliSoftmn.com (Casey Crookston) Date: Wed, 13 Dec 2006 09:39:08 -0600 Subject: [thelist] PHP Date question Message-ID: <24233F1426CDC547B0CC54EEA4938520030434@ngsv31.my-netgain.com> I'm de-bugging an app I didn't code. On a report, two date variables are set as follows: $this->Search->Fields["DateStart"]->Value .= " 12:00:00 AM"; $this->Search->Fields["DateEnd"]->Value .= " 11:59:59 PM"; So, the data for the report runs from 12:00:00 AM on the start date selected by the user to 11:59:59 on the end date selected by the user. Trouble is, their bank/merchant account is several time zones off. We need to adjust the times to run from 8:00:00 PM of the PREVIOUS DAY (one day before the user selects as the start date) to 7:59:59 PM of the end date. Setting the end time to 7:59:59 is simple enough, but I'm not sure how to adjust the start date back one day. Something like: $date-> Fields["DateStart"]->modify("-1 day"); But that doesn't work. I'm sure it's simple, but I know nothing about PHP. Thanks, Casey From mwarden at gmail.com Wed Dec 13 09:32:49 2006 From: mwarden at gmail.com (Matt Warden) Date: Wed, 13 Dec 2006 10:32:49 -0500 Subject: [thelist] Adding 15 mins to PHP time... In-Reply-To: <8b20a7490612130725r6e1171bagf8640446984cc2e0@mail.gmail.com> References: <8b20a7490612130725r6e1171bagf8640446984cc2e0@mail.gmail.com> Message-ID: On 12/13/06, Tris wrote: > How can I add 15 mins to a date function? http://us3.php.net/strtotime -- Matt Warden Cleveland, OH, USA http://mattwarden.com This email proudly and graciously contributes to entropy. From eduardok at gmail.com Wed Dec 13 09:46:57 2006 From: eduardok at gmail.com (Eduardo Kienetz) Date: Wed, 13 Dec 2006 13:46:57 -0200 Subject: [thelist] Adding 15 mins to PHP time... In-Reply-To: <8b20a7490612130725r6e1171bagf8640446984cc2e0@mail.gmail.com> References: <8b20a7490612130725r6e1171bagf8640446984cc2e0@mail.gmail.com> Message-ID: <3e7d042a0612130746p11928b1br9d30e6384978a85a@mail.gmail.com> On 12/13/06, Tris wrote: > I'm creating a timer for todays lucky client... > They wanna hit start and have a timer on screen across all pages.. > No prob doing that in javascript, so I'm gonna create a session called > timer with the date value: > date("H:i:s"); Giving: 15:42:34 etc.. (the precise users start time) > > What I need to do is on each page, check the current server time, and > if it's 15 mis higher than my session, redirect the user to a thank > you page and destroythetimer session.. > > soooo... > > How can I add 15 mins to a date function? > (I'm also gonna have to fee different times to the javasript ticker, > but that'll be easiy once I kno whow to knock off time from um, well, > time :-p ) $startDate = "01/08/2006 08:04:20"; echo date("d/m/Y H:i:s", strtotime("+15 minutes",strtotime($startDate))); Will print: 08/01/2006 08:19:20 -- Eduardo Bacchi Kienetz LPI Certified - Level 2 http://www.noticiaslinux.com.br/eduardo/ From mwarden at gmail.com Tue Dec 12 18:37:47 2006 From: mwarden at gmail.com (Matt Warden) Date: Tue, 12 Dec 2006 19:37:47 -0500 Subject: [thelist] Workaround for setAttribute bug in IE7? In-Reply-To: <610592c90612120213i1037cf1ds2c056b52267edafa@mail.gmail.com> References: <457CF780.8060004@ryanrushton.com> <457DE052.9000809@bitstorm.org> <610592c90612120213i1037cf1ds2c056b52267edafa@mail.gmail.com> Message-ID: On 12/12/06, Lee kowalkowski wrote: > ...because you're invoking an assignment operator on a DOM object in > the first place? In JavaScript, the assignment operator often creates > new properties: window.foo = "bar"; - If I thought that wasn't capable > of creating foo if it didn't exist, I ought to read the ECMAScript > language specification again. Yes, it creates an *object property* in JavaScript. If you read my entire message (not just the portion you replied to), you'll see I'm saying that having an assignment with the side effect of creating DOM nodes is a really dumb idea. Question: does this create an attribute node or not? domnode.length = 4; Answer: Thanks to this horrible design decision, you don't know unless you know what domnode is and whether .length is just an object property or an object property tied to an attribute node in the DOM. > None of the above in IE - you get an invalid argument exception. Now > it *is* a really dumb idea to expect assigning an property to null > will delete the property! Again, please read my message carefully. I am not talking about properties, I'm talking about attribute nodes in the DOM. Assuming you understand this, your argument is: Great idea: assigning value to object property creates DOM attribute node Really dumb idea: removing assignment to object property deletes DOM attribute node ?! -- Matt Warden Cleveland, OH, USA http://mattwarden.com This email proudly and graciously contributes to entropy. From beertastic at gmail.com Wed Dec 13 10:05:29 2006 From: beertastic at gmail.com (Tris) Date: Wed, 13 Dec 2006 16:05:29 +0000 Subject: [thelist] Adding 15 mins to PHP time... In-Reply-To: <3e7d042a0612130746p11928b1br9d30e6384978a85a@mail.gmail.com> References: <8b20a7490612130725r6e1171bagf8640446984cc2e0@mail.gmail.com> <3e7d042a0612130746p11928b1br9d30e6384978a85a@mail.gmail.com> Message-ID: <8b20a7490612130805h75847205j6452d8a5a9b2a646@mail.gmail.com> MAny thaks to all.. I ended up doing: $_SESSION[timeStart] = date("H:i:s"); $_SESSION[timeEnd] = date('H:i:s',strtotime('+15 minute')); and just putting:: if ($_SESSION[timeEnd] <= $_SESSION[timeStart]) { $pageDisplay = "TIMES UP!"; } on each page... Hopefully it'll work.. right? ;-p o rdo I have to convert it to a unix time stamp blah blah blah.. Tris... On 13/12/06, Eduardo Kienetz wrote: > On 12/13/06, Tris wrote: > > I'm creating a timer for todays lucky client... > > They wanna hit start and have a timer on screen across all pages.. > > No prob doing that in javascript, so I'm gonna create a session called > > timer with the date value: > > date("H:i:s"); Giving: 15:42:34 etc.. (the precise users start time) > > > > What I need to do is on each page, check the current server time, and > > if it's 15 mis higher than my session, redirect the user to a thank > > you page and destroythetimer session.. > > > > soooo... > > > > How can I add 15 mins to a date function? > > (I'm also gonna have to fee different times to the javasript ticker, > > but that'll be easiy once I kno whow to knock off time from um, well, > > time :-p ) > > $startDate = "01/08/2006 08:04:20"; > echo date("d/m/Y H:i:s", strtotime("+15 minutes",strtotime($startDate))); > Will print: 08/01/2006 08:19:20 > > -- > Eduardo Bacchi Kienetz > LPI Certified - Level 2 > http://www.noticiaslinux.com.br/eduardo/ > -- > > * * Please support the community that supports you. * * > http://evolt.org/help_support_evolt/ > > For unsubscribe and other options, including the Tip Harvester > and archives of thelist go to: http://lists.evolt.org > Workers of the Web, evolt ! > From kenrbnsn at rbnsn.com Wed Dec 13 10:13:31 2006 From: kenrbnsn at rbnsn.com (Ken Robinson) Date: Wed, 13 Dec 2006 11:13:31 -0500 Subject: [thelist] Adding 15 mins to PHP time... In-Reply-To: <8b20a7490612130805h75847205j6452d8a5a9b2a646@mail.gmail.com> References: <8b20a7490612130725r6e1171bagf8640446984cc2e0@mail.gmail.com> <3e7d042a0612130746p11928b1br9d30e6384978a85a@mail.gmail.com> <8b20a7490612130805h75847205j6452d8a5a9b2a646@mail.gmail.com> Message-ID: <20061213111331.whmbc0um1ls08s0g@www.rbnsn.com> Quoting Tris : > MAny thaks to all.. > I ended up doing: > > $_SESSION[timeStart] = date("H:i:s"); > $_SESSION[timeEnd] = date('H:i:s',strtotime('+15 minute')); > > and just putting:: > > if ($_SESSION[timeEnd] <= $_SESSION[timeStart]) { > $pageDisplay = "TIMES UP!"; > } > > on each page... > > Hopefully it'll work.. right? ;-p > o rdo I have to convert it to a unix time stamp blah blah blah.. Remember, you're comparing strings here. If you convert to a UNIX timestamp you will be comparing integers. Ken From rick.denhaan at gmail.com Wed Dec 13 09:51:15 2006 From: rick.denhaan at gmail.com (Rick den Haan) Date: Wed, 13 Dec 2006 16:51:15 +0100 Subject: [thelist] Adding 15 mins to PHP time... In-Reply-To: <8b20a7490612130725r6e1171bagf8640446984cc2e0@mail.gmail.com> References: <8b20a7490612130725r6e1171bagf8640446984cc2e0@mail.gmail.com> Message-ID: <45802173.4070402@gmail.com> Tris schreef: > How can I add 15 mins to a date function [in PHP, ed.]? > (I'm also gonna have to fee different times to the javasript ticker, > but that'll be easiy once I kno whow to knock off time from um, well, > time :-p ) > Tris, When I need to do something like this in PHP, I tend to use a timestamp, and add (or deduct) the necessary amount of time in seconds. I.e.: $now = time(); $earlier = $now - (60 * 15); // 60 seconds * 15 minutes You can then push $earlier to the date function, or strftime, or your choice of magickery to turn it into a date, e.g.... $then = date("Y-m-d H:i:s", $earlier); ...to get a string you can paste into a MySQL-query. HTH, Rick. From kenrbnsn at rbnsn.com Wed Dec 13 09:41:57 2006 From: kenrbnsn at rbnsn.com (Ken Robinson) Date: Wed, 13 Dec 2006 10:41:57 -0500 Subject: [thelist] Adding 15 mins to PHP time... In-Reply-To: <8b20a7490612130725r6e1171bagf8640446984cc2e0@mail.gmail.com> References: <8b20a7490612130725r6e1171bagf8640446984cc2e0@mail.gmail.com> Message-ID: <20061213104157.h5zjbllk3l44ckw0@www.rbnsn.com> Quoting Tris : > I'm creating a timer for todays lucky client... > They wanna hit start and have a timer on screen across all pages.. > No prob doing that in javascript, so I'm gonna create a session called > timer with the date value: > date("H:i:s"); Giving: 15:42:34 etc.. (the precise users start time) > > What I need to do is on each page, check the current server time, and > if it's 15 mis higher than my session, redirect the user to a thank > you page and destroythetimer session.. > > soooo... > > How can I add 15 mins to a date function? > (I'm also gonna have to fee different times to the javasript ticker, > but that'll be easiy once I kno whow to knock off time from um, well, > time :-p ) To add times in PHP, you need to convert everything to the number of seconds since 1-1-1970 (zero date in PHP) To make it simple, just store the integer value in your session variable: $_SESSION['users_start_time'] = time(); 15 minutes is 900 seconds, so to see whether the current time is more than 15 minutes later, do something like this: $time_dif = time() - $_SESSION['users_start_time']; if ($time_dif > 900) { unset($_SESSION['users_start_time')); header('location: thankyou.php'); } Ken From beertastic at gmail.com Wed Dec 13 10:22:44 2006 From: beertastic at gmail.com (Tris) Date: Wed, 13 Dec 2006 16:22:44 +0000 Subject: [thelist] Adding 15 mins to PHP time... In-Reply-To: <8b20a7490612130820l1062d9b0uc5190d21502ec0b8@mail.gmail.com> References: <8b20a7490612130725r6e1171bagf8640446984cc2e0@mail.gmail.com> <3e7d042a0612130746p11928b1br9d30e6384978a85a@mail.gmail.com> <8b20a7490612130805h75847205j6452d8a5a9b2a646@mail.gmail.com> <20061213111331.whmbc0um1ls08s0g@www.rbnsn.com> <8b20a7490612130820l1062d9b0uc5190d21502ec0b8@mail.gmail.com> Message-ID: <8b20a7490612130822y7c473bcm77e520cc8e1138c8@mail.gmail.com> I'm a tool!!! I was comparing it to 15 mins ago. NOT now... Monkey man!! fixed, movingon to bug 471.. ;-p CHEERS!!! On 13/12/06, Tris wrote: > Just this sec realised that.. cut cheers! > > My code is not working... > I've changeds the 15misn to 10 seconds for testing.. but it's not > working, any thoughts? > > > ============================ > > if ((!isset($_SESSION[timeStart])) && (!isset($_SESSION[timeEnd]))) { > > $_SESSION[timeStart] = strtotime('now'); > $_SESSION[timeEnd] = strtotime('+10 second'); > > } else { > > $now = strtotime('now'); > if ($_SESSION[timeStart] > $_SESSION[timeEnd]) { > $pageDisplay = "TIMES UP!"; > } else { > $pageDisplay = "Keep going!"; > } > > $pageDisplay .= " >

Current Time: $now >

>

Time to end: $_SESSION[timeEnd] >

start Time: $_SESSION[timeStart] > "; > > } > > ============================ > > > > On 13/12/06, Ken Robinson wrote: > > Quoting Tris : > > > > > MAny thaks to all.. > > > I ended up doing: > > > > > > $_SESSION[timeStart] = date("H:i:s"); > > > $_SESSION[timeEnd] = date('H:i:s',strtotime('+15 minute')); > > > > > > and just putting:: > > > > > > if ($_SESSION[timeEnd] <= $_SESSION[timeStart]) { > > > $pageDisplay = "TIMES UP!"; > > > } > > > > > > on each page... > > > > > > Hopefully it'll work.. right? ;-p > > > o rdo I have to convert it to a unix time stamp blah blah blah.. > > > > Remember, you're comparing strings here. If you convert to a UNIX > > timestamp you > > will be comparing integers. > > > > Ken > > > > > > -- > > > > * * Please support the community that supports you. * * > > http://evolt.org/help_support_evolt/ > > > > For unsubscribe and other options, including the Tip Harvester > > and archives of thelist go to: http://lists.evolt.org > > Workers of the Web, evolt ! > > > From beertastic at gmail.com Wed Dec 13 10:20:41 2006 From: beertastic at gmail.com (Tris) Date: Wed, 13 Dec 2006 16:20:41 +0000 Subject: [thelist] Adding 15 mins to PHP time... In-Reply-To: <20061213111331.whmbc0um1ls08s0g@www.rbnsn.com> References: <8b20a7490612130725r6e1171bagf8640446984cc2e0@mail.gmail.com> <3e7d042a0612130746p11928b1br9d30e6384978a85a@mail.gmail.com> <8b20a7490612130805h75847205j6452d8a5a9b2a646@mail.gmail.com> <20061213111331.whmbc0um1ls08s0g@www.rbnsn.com> Message-ID: <8b20a7490612130820l1062d9b0uc5190d21502ec0b8@mail.gmail.com> Just this sec realised that.. cut cheers! My code is not working... I've changeds the 15misn to 10 seconds for testing.. but it's not working, any thoughts? ============================ if ((!isset($_SESSION[timeStart])) && (!isset($_SESSION[timeEnd]))) { $_SESSION[timeStart] = strtotime('now'); $_SESSION[timeEnd] = strtotime('+10 second'); } else { $now = strtotime('now'); if ($_SESSION[timeStart] > $_SESSION[timeEnd]) { $pageDisplay = "TIMES UP!"; } else { $pageDisplay = "Keep going!"; } $pageDisplay .= "

Current Time: $now

Time to end: $_SESSION[timeEnd]

start Time: $_SESSION[timeStart] "; } ============================ On 13/12/06, Ken Robinson wrote: > Quoting Tris : > > > MAny thaks to all.. > > I ended up doing: > > > > $_SESSION[timeStart] = date("H:i:s"); > > $_SESSION[timeEnd] = date('H:i:s',strtotime('+15 minute')); > > > > and just putting:: > > > > if ($_SESSION[timeEnd] <= $_SESSION[timeStart]) { > > $pageDisplay = "TIMES UP!"; > > } > > > > on each page... > > > > Hopefully it'll work.. right? ;-p > > o rdo I have to convert it to a unix time stamp blah blah blah.. > > Remember, you're comparing strings here. If you convert to a UNIX > timestamp you > will be comparing integers. > > Ken > > > -- > > * * Please support the community that supports you. * * > http://evolt.org/help_support_evolt/ > > For unsubscribe and other options, including the Tip Harvester > and archives of thelist go to: http://lists.evolt.org > Workers of the Web, evolt ! > From Ron.Luther at hp.com Wed Dec 13 10:14:17 2006 From: Ron.Luther at hp.com (Luther, Ron) Date: Wed, 13 Dec 2006 10:14:17 -0600 Subject: [thelist] Proper commenting and code maintainance [was: Why doesthis JS code work the way it does???] In-Reply-To: Message-ID: <1FF0647605921940BCB22C2A454AD7CE0119473F@cceexc17.americas.cpqcorp.net> Hi Gang, Since this thread has morphed into 'proper commenting and code maintenance' ... a question: Anybody playing with loading code comments, documentation, or training info into custom object properties? With object reuse in modules developed by different folks or even different teams does it make sense to create some common 'comment' attributes that you can add to give other folks a clue about intended use, purpose, range, restrictions, etc? [Kind of a cheapie meta-data repository if you will.] Good Idea? Bad Idea? Another of Ron's whack-a-doodle ideas? Regards, RonL. {What if you combined it with an in-house 'bookmarklet' kind of thing to read these back out to you as you navigated through the site? Does that make it more sensible?} From Ron.Luther at hp.com Wed Dec 13 10:40:46 2006 From: Ron.Luther at hp.com (Luther, Ron) Date: Wed, 13 Dec 2006 10:40:46 -0600 Subject: [thelist] PHP Date question In-Reply-To: <24233F1426CDC547B0CC54EEA4938520030434@ngsv31.my-netgain.com> Message-ID: <1FF0647605921940BCB22C2A454AD7CE01194741@cceexc17.americas.cpqcorp.net> Casey Crookston asked about moving two dates from midnight to midnight ... To 8pm to 8pm: Hi Casey, If these fields are really timestamps then it looks like you only need to subtract 4 hours from each one. [Google for links to all PHP functions.] >>$date-> Fields["DateStart"]->modify("-1 day"); How about: $newDateStart = $oldDateStart; // whatever the old date was called $newDateStart->modify("-4 hours"); echo $newDateStart; // to see if it worked HTH, RonL. From lists at neptunewebworks.com Wed Dec 13 10:54:42 2006 From: lists at neptunewebworks.com (Max Schwanekamp) Date: Wed, 13 Dec 2006 08:54:42 -0800 Subject: [thelist] PHP Date question In-Reply-To: <24233F1426CDC547B0CC54EEA4938520030434@ngsv31.my-netgain.com> References: <24233F1426CDC547B0CC54EEA4938520030434@ngsv31.my-netgain.com> Message-ID: <45803052.2050709@neptunewebworks.com> Casey Crookston wrote: > $this->Search->Fields["DateStart"]->Value .= " 12:00:00 AM"; > $this->Search->Fields["DateEnd"]->Value .= " 11:59:59 PM"; > So, the data for the report runs from 12:00:00 AM on the start date > selected by the user to 11:59:59 on the end date selected by the user. > Trouble is, their bank/merchant account is several time zones off. We > need to adjust the times to run from 8:00:00 PM of the PREVIOUS DAY (one > day before the user selects as the start date) to 7:59:59 PM of the end > date. Setting the end time to 7:59:59 is simple enough, but I'm not > sure how to adjust the start date back one day. > $date-> Fields["DateStart"]->modify("-1 day"); > But that doesn't work. I'm sure it's simple, but I know nothing about > PHP. Thanks, If you absolutely must work with date strings instead of unix timestamps (much easier in most cases), then strtotime() is your friend. But again, it's way easier to just convert user inputted date strings into unix timestamps and work with those, since they're integers. -- Max Schwanekamp NeptuneWebworks.com 541-517-9064 From mwarden at gmail.com Wed Dec 13 10:55:42 2006 From: mwarden at gmail.com (Matt Warden) Date: Wed, 13 Dec 2006 11:55:42 -0500 Subject: [thelist] Proper commenting and code maintainance [was: Why doesthis JS code work the way it does???] In-Reply-To: <1FF0647605921940BCB22C2A454AD7CE0119473F@cceexc17.americas.cpqcorp.net> References: <1FF0647605921940BCB22C2A454AD7CE0119473F@cceexc17.americas.cpqcorp.net> Message-ID: On 12/13/06, Luther, Ron wrote: > Anybody playing with loading code comments, documentation, > or training info into custom object properties? Can you give an example of what you mean? I'm not sure I'm following. Do you mean: $ println(MyAwesomeClass.purpose); MyAwesomeClass is used to represent awesome things and... ? -- Matt Warden Cleveland, OH, USA http://mattwarden.com This email proudly and graciously contributes to entropy. From lee.kowalkowski at googlemail.com Wed Dec 13 10:58:53 2006 From: lee.kowalkowski at googlemail.com (Lee kowalkowski) Date: Wed, 13 Dec 2006 16:58:53 +0000 Subject: [thelist] Workaround for setAttribute bug in IE7? In-Reply-To: References: <457CF780.8060004@ryanrushton.com> <457DE052.9000809@bitstorm.org> <610592c90612120213i1037cf1ds2c056b52267edafa@mail.gmail.com> Message-ID: <610592c90612130858s2d36c717s637f6ad55f64d5b0@mail.gmail.com> On 13/12/06, Matt Warden wrote: > On 12/12/06, Lee kowalkowski wrote: > Yes, it creates an *object property* in JavaScript. If you read my > entire message (not just the portion you replied to), you'll see I'm > saying that having an assignment with the side effect of creating DOM > nodes is a really dumb idea. I understood, and I'm saying in a JavaScript binding to a browser-supplied object, the difference between attribute and property should ideally be transparent as a matter of utmost convenience. My emphasis is the fact that the JS and DOM are bound, therefore if it *didn't* create the attribute - *that* would be a side-effect. > Really dumb idea: removing assignment to object property deletes DOM > attribute node I didn't actually say that because you can't remove assignments to DOM object properties without using removeAttribute (In IE). The assignment operator is no way to go about removing objects or properties. If anything should do the removing of attributes in the JS binding it should be the delete operator, but unfortunately IE's browser DOM doesn't support that action, although other browsers do. What I said was assigning a property to null or undefined doesn't remove the property itself, just changes its value. It could even create the property, if it doesn't exist, as stated in ECMA-262 section 8.6.2.2. So why shouldn't I expect it to create a DOM attribute? -- Lee From caseyc at IntelliSoftmn.com Wed Dec 13 11:13:59 2006 From: caseyc at IntelliSoftmn.com (Casey Crookston) Date: Wed, 13 Dec 2006 11:13:59 -0600 Subject: [thelist] PHP Date question Message-ID: <24233F1426CDC547B0CC54EEA4938520030435@ngsv31.my-netgain.com> >>$newDateStart->modify("-4 hours");<< Hmmm. $dStart .= modify("-4 hours"); $dStart = $dStart->modify("-4 hours"); $dStart->modify("-4 hours"); All three of these error out. And, I am not given an error. The page load just stops. Errg. Casey From caseyc at IntelliSoftmn.com Wed Dec 13 11:14:42 2006 From: caseyc at IntelliSoftmn.com (Casey Crookston) Date: Wed, 13 Dec 2006 11:14:42 -0600 Subject: [thelist] PHP Date question Message-ID: <24233F1426CDC547B0CC54EEA4938520177972@ngsv31.my-netgain.com> >>If you absolutely must work with date strings instead of unix timestamps (much easier in most cases), then strtotime() is your friend.<< Will this work against an MSSQL database? From anthony at baratta.com Wed Dec 13 11:35:32 2006 From: anthony at baratta.com (Anthony Baratta) Date: Wed, 13 Dec 2006 09:35:32 -0800 Subject: [thelist] Proper commenting and code maintainance [was: Whydoesthis JS code work the way it does???] Message-ID: <3b08b67f20365d8d60f0e823c273f879@baratta.com> > On 12/13/06, Luther, Ron wrote: > > Anybody playing with loading code comments, documentation, > > or training info into custom object properties? I'm primarily coding in dotNet C# right now and when we build Assemblies, I've been forcing my team to use the XML comments format for adding info about the NameSpace, Class, and Function into the compiled code. We then turn on XML documentation at compile time to generate and XML "help" document and use NDOC to convert that into a compiled help file and web pages. That format works pretty well for us right now. From caseyc at IntelliSoftmn.com Wed Dec 13 11:52:29 2006 From: caseyc at IntelliSoftmn.com (Casey Crookston) Date: Wed, 13 Dec 2006 11:52:29 -0600 Subject: [thelist] PHP Date question Message-ID: <24233F1426CDC547B0CC54EEA4938520177973@ngsv31.my-netgain.com> >>If you absolutely must work with date strings instead of unix timestamps (much easier in most cases), then strtotime() is your friend.<< Solved. $dStart = strtotime("-4 hours",strtotime($dStart)); $dStart = date('m/d/Y H:i:s A',$dStart); Thank you! From Ron.Luther at hp.com Wed Dec 13 12:17:27 2006 From: Ron.Luther at hp.com (Luther, Ron) Date: Wed, 13 Dec 2006 12:17:27 -0600 Subject: [thelist] Proper commenting and code maintainance [was: Whydoesthis JS code work the way it does???] In-Reply-To: Message-ID: <1FF0647605921940BCB22C2A454AD7CE01194742@cceexc17.americas.cpqcorp.net> Matt Warden asked: >>Can you give an example of what you mean? I'm not sure I'm following. Hi Matt, What I was wondering was whether something like this might be useful: $ println(ReturnFunction.DocNotes); The qty field passed to the ReturnFunction object should always be a positive integer. Do not EVER send a negative number even though this is the quantity being returned. What I have actually done is more along the lines of embedding user help hints: $ println(RadioButton4.RonHelp); Press this radio button if you want to see fiscal year results. Warning: will not work for fiscal years prior to 1642! That a little clearer? RonL. From Ron.Luther at hp.com Wed Dec 13 12:24:24 2006 From: Ron.Luther at hp.com (Luther, Ron) Date: Wed, 13 Dec 2006 12:24:24 -0600 Subject: [thelist] Proper commenting and code maintainance [was:Whydoesthis JS code work the way it does???] In-Reply-To: <3b08b67f20365d8d60f0e823c273f879@baratta.com> Message-ID: <1FF0647605921940BCB22C2A454AD7CE01194743@cceexc17.americas.cpqcorp.net> Anthony Baratta noted: >>I'm primarily coding in dotNet C# right now and when we build >>Assemblies, I've been forcing my team to use the XML comments >>format for adding info about the NameSpace, Class, and Function >>into the compiled code. We then turn on XML documentation at >>compile time to generate and XML "help" document and use NDOC >>to convert that into a compiled help file and web pages. >>That format works pretty well for us right now. Hi Anthony, That sounds pretty neat - and much more comprehensive than what I was thinking about! ;-) Actually, I was kinda thinking back to the old days when I seem to recall you doing some neat things with ASP dictionary objects. I was just kinda wondering if it might make some sense to leave a helpful hint or two to future code maintainers on how to use those boogers ... and whether it might be cool to leave that note in a custom attribute on the object itself cuz paper and 'outfile' documentation tends to disappear over time. RonL. From mwarden at gmail.com Wed Dec 13 12:42:43 2006 From: mwarden at gmail.com (Matt Warden) Date: Wed, 13 Dec 2006 13:42:43 -0500 Subject: [thelist] Workaround for setAttribute bug in IE7? In-Reply-To: <610592c90612130858s2d36c717s637f6ad55f64d5b0@mail.gmail.com> References: <457CF780.8060004@ryanrushton.com> <457DE052.9000809@bitstorm.org> <610592c90612120213i1037cf1ds2c056b52267edafa@mail.gmail.com> <610592c90612130858s2d36c717s637f6ad55f64d5b0@mail.gmail.com> Message-ID: On 12/13/06, Lee kowalkowski wrote: > > removing assignment to object property deletes DOM > > attribute node > > I didn't actually say that because you can't remove assignments to DOM > object properties without using removeAttribute (In IE). The > assignment operator is no way to go about removing objects or > properties. This was my point. > What I said was assigning a property to null or undefined doesn't > remove the property itself, just changes its value. It could even > create the property, if it doesn't exist, as stated in ECMA-262 > section 8.6.2.2. Of course. > So why shouldn't I expect it to create a DOM > attribute? See my domnode.length example. It was in the message you replied to, but just in case: http://lists.evolt.org/archive/Week-of-Mon-20061211/186779.html It should only be "transparent to the user" if it's the only way to manipulate attributes, or if all other methods are similarly transparent. Otherwise it's a dumb design decision that causes long-winded circular discussions like this one. -- Matt Warden Cleveland, OH, USA http://mattwarden.com This email proudly and graciously contributes to entropy. From john.destefano at gmail.com Wed Dec 13 15:39:15 2006 From: john.destefano at gmail.com (John DeStefano) Date: Wed, 13 Dec 2006 16:39:15 -0500 Subject: [thelist] rss feed pagination Message-ID: Anyone know of a quick & dirty way to clean up an RSS feed? There's a lot of data being sent from this particular feed, and it would be great to add pagination for easier navigation. I'm looking at the transforms being used currently, and I don't know XSLT well enough to be able to wade through the templates properly. Thanks, ~John From lee.kowalkowski at googlemail.com Wed Dec 13 15:58:28 2006 From: lee.kowalkowski at googlemail.com (Lee kowalkowski) Date: Wed, 13 Dec 2006 21:58:28 +0000 Subject: [thelist] Workaround for setAttribute bug in IE7? In-Reply-To: References: <457CF780.8060004@ryanrushton.com> <457DE052.9000809@bitstorm.org> <610592c90612120213i1037cf1ds2c056b52267edafa@mail.gmail.com> <610592c90612130858s2d36c717s637f6ad55f64d5b0@mail.gmail.com> Message-ID: <610592c90612131358h470ecbfesed97277ba3ac1e03@mail.gmail.com> On 13/12/06, Matt Warden wrote: > See my domnode.length example. It was in the message you replied to, > but just in case: > http://lists.evolt.org/archive/Week-of-Mon-20061211/186779.html :-) Your example doesn't illustrate how using setAttribute("length", 4) is any better than length=4 (which I think was your original assertion). Because on HTML DOM objects that actually have a length attribute, they are read-only, both methods will fail! If it isn't read only, both methods would work, and yes, both methods would result in the creation of an attribute. Hardly the fault of a design decision. > It should only be "transparent to the user" if it's the only way to > manipulate attributes, or if all other methods are similarly > transparent. What? Because there's more than one method neither should be transparent? Doesn't make any sense. One's explicit and one's implicit, I don't have a problem with that, except for the occasional browser bug. -- Lee From mwarden at gmail.com Wed Dec 13 16:23:37 2006 From: mwarden at gmail.com (Matt Warden) Date: Wed, 13 Dec 2006 17:23:37 -0500 Subject: [thelist] Workaround for setAttribute bug in IE7? In-Reply-To: <610592c90612131358h470ecbfesed97277ba3ac1e03@mail.gmail.com> References: <457CF780.8060004@ryanrushton.com> <457DE052.9000809@bitstorm.org> <610592c90612120213i1037cf1ds2c056b52267edafa@mail.gmail.com> <610592c90612130858s2d36c717s637f6ad55f64d5b0@mail.gmail.com> <610592c90612131358h470ecbfesed97277ba3ac1e03@mail.gmail.com> Message-ID: On 12/13/06, Lee kowalkowski wrote: > On 13/12/06, Matt Warden wrote: > > See my domnode.length example. It was in the message you replied to, > > but just in case: > > http://lists.evolt.org/archive/Week-of-Mon-20061211/186779.html > > :-) Your example doesn't illustrate how using setAttribute("length", > 4) is any better than length=4 (which I think was your original > assertion). Because on HTML DOM objects that actually have a length > attribute, they are read-only, both methods will fail! If it isn't > read only, both methods would work, and yes, both methods would result > in the creation of an attribute. Hardly the fault of a design > decision. Again, missing the entire point. The fact that you have to know this: > Because on HTML DOM objects that actually have a length > attribute, they are read-only, both methods will fail! is my point. And here is a real-world example: http://mwarden.f2o.org/sandbox/attributetest.htm -- Matt Warden Cleveland, OH, USA http://mattwarden.com This email proudly and graciously contributes to entropy. From tim at hyperlinkage.com Wed Dec 13 16:37:53 2006 From: tim at hyperlinkage.com (tim) Date: Wed, 13 Dec 2006 22:37:53 +0000 Subject: [thelist] rss feed pagination In-Reply-To: References: Message-ID: <458080C1.6090709@hyperlinkage.com> John DeStefano wrote: > Anyone know of a quick & dirty way to clean up an RSS feed? There's a > lot of data being sent from this particular feed, and it would be > great to add pagination for easier navigation. Could you give us some more information? Are you producing the feed yourself? Could you show us the XSLT you are using? Tim -- http://www.hyperlinkage.com/ From tim at hyperlinkage.com Wed Dec 13 16:48:20 2006 From: tim at hyperlinkage.com (tim) Date: Wed, 13 Dec 2006 22:48:20 +0000 Subject: [thelist] Proper commenting and code maintainance [was: Why does this JS code work the way it does???] In-Reply-To: References: Message-ID: <45808334.6040203@hyperlinkage.com> VOLKAN ?Z?EL?K wrote: > Then let us create a hybrid of both views: > > // if there keyCode is not one of the numbers in validkeys > // do not allow the keystroke > // otherwise allow it: > return (strValidKeys.indexOf(String.fromCharCode(event.keyCode)) == -1); Exactly. My point was not to suggest that comments should be removed, but that there were several lines of code which weren't needed. In my opinion it is much more readable to have the return values described on the line just before the method, rather than in the method body. Also, I would use an extra variable to store the pressed key, rather than calling multiple methods on one line. My version would be: // Returns true if the key stroke is allowed function CheckNumericKeyStroke() { var validKeys = "1234567890"; var pressedKey = String.fromCharCode(event.keyCode); return (validKeys.indexOf(pressedKey) != -1); } Rather than: function CheckNumericKeyStroke() { var strValidKeys = "1234567890"; // if there keyCode is not one of the numbers in validkeys... if (strValidKeys.indexOf(String.fromCharCode(event.keyCode)) == -1) { return false; // do not allow the keystroke } else { return true; // allow the keystroke } } Tim -- http://www.hyperlinkage.com/ From anthony at baratta.com Wed Dec 13 16:51:57 2006 From: anthony at baratta.com (Anthony Baratta) Date: Wed, 13 Dec 2006 14:51:57 -0800 Subject: [thelist] Proper commenting and code maintainance[was:Whydoesthis JS code work the way it does???] Message-ID: -----Original message----- From: "Luther, Ron" Ron.Luther at hp.com Date: Wed, 13 Dec 2006 10:24:24 -0800 To: "Anthony Baratta" anthony at baratta.com, thelist at lists.evolt.org Subject: RE: [thelist] Proper commenting and code maintainance [was:Whydoesthis JS code work the way it does???] > Hi Anthony, > > That sounds pretty neat - and much more comprehensive than what > I was thinking about! ;-) > > Actually, I was kinda thinking back to the old days when I seem > to recall you doing some neat things with ASP dictionary objects. > I was just kinda wondering if it might make some sense to leave > a helpful hint or two to future code maintainers on how to use > those boogers ... and whether it might be cool to leave that note > in a custom attribute on the object itself cuz paper and 'outfile' > documentation tends to disappear over time. Yeah - the nice thing about the assemblyembeded XML documentation is that it works with IntelliSense. Once you create a refernce to the DLL and Visual Studio gives you "code completion" with the object properties and methods - the XML comments show up as "hints". So at minimum you have that. Your style of embeded an object with help info is workable too, but if someone doesn't know the "Help" Property name - without some type of code completion / IntelliSense assiting you - that is as lost as any physical documentation. ;-P It all boils down to a basic framework that everyone buys in on, and as people come and go - is easy to maintain as part of the coding culture. But as you well know, when you switch environments / coding languages / platforms - all bets are off and everything changes anyway. Which is why I've been able to force my current work environment into the direction we are going because dotNet is all new here. No dotNet legacy and no current dotNet culture to over come. From lee.kowalkowski at googlemail.com Thu Dec 14 04:20:57 2006 From: lee.kowalkowski at googlemail.com (Lee kowalkowski) Date: Thu, 14 Dec 2006 10:20:57 +0000 Subject: [thelist] Workaround for setAttribute bug in IE7? In-Reply-To: References: <457CF780.8060004@ryanrushton.com> <457DE052.9000809@bitstorm.org> <610592c90612120213i1037cf1ds2c056b52267edafa@mail.gmail.com> <610592c90612130858s2d36c717s637f6ad55f64d5b0@mail.gmail.com> <610592c90612131358h470ecbfesed97277ba3ac1e03@mail.gmail.com> Message-ID: <610592c90612140220s4f0ee16pe071c542670dbdbc@mail.gmail.com> On 13/12/06, Matt Warden wrote: > Again, missing the entire point. I must admit, it's a good thing I miss your points so you can respond and make them again more clearly. > The fact that you have to know this: >> Because on HTML DOM objects that actually have a length >> attribute, they are read-only, both methods will fail! > is my point. It is? Well I'm glad I could help you make your point. I'm still not sure why you're making it. It's clear if you want to get by with anything, learning is the key to fewer problems. > And here is a real-world example: > http://mwarden.f2o.org/sandbox/attributetest.htm A cryptic example at best, there's no explanation to back up what you're trying to demonstrate. No wonder I'm having difficulty understanding you. I forgive you though, it's quite endearing actually, but only because you're a genius. -- Lee From mwarden at gmail.com Thu Dec 14 07:26:48 2006 From: mwarden at gmail.com (Matt Warden) Date: Thu, 14 Dec 2006 08:26:48 -0500 Subject: [thelist] Workaround for setAttribute bug in IE7? In-Reply-To: <610592c90612140220s4f0ee16pe071c542670dbdbc@mail.gmail.com> References: <457CF780.8060004@ryanrushton.com> <457DE052.9000809@bitstorm.org> <610592c90612120213i1037cf1ds2c056b52267edafa@mail.gmail.com> <610592c90612130858s2d36c717s637f6ad55f64d5b0@mail.gmail.com> <610592c90612131358h470ecbfesed97277ba3ac1e03@mail.gmail.com> <610592c90612140220s4f0ee16pe071c542670dbdbc@mail.gmail.com> Message-ID: On 12/14/06, Lee kowalkowski wrote: > > And here is a real-world example: > > http://mwarden.f2o.org/sandbox/attributetest.htm > > A cryptic example at best, there's no explanation to back up what > you're trying to demonstrate. No wonder I'm having difficulty > understanding you. Apologies. I unfortunately did not have a lot of time to explain what I thought was obvious since the example includes nothing other than the setAttribute issue I'm discussing. Writing up the example itself was pushing it, but you were obviously having trouble with th explanation and I thought an example would help. The alternative to the setAttribute lines is: input.type = 'text'; input.size = 5; These could easily be object properties with no side effect on the DOM, just like: input.foo = "bar"; input.size = 5; creates an attribute node in the DOM. input.foo = "bar"; does not. (As an aside, what if my doctype defines an attribute called "foo"?) Try to make this simple: Clearly an operation manipulating the DOM: input.setAttribute('size', 5); Clearly an assignment to an object property: input.size = 5; Aside from my entire argument, you can do a simple google search to see how often this issue trips up developers. (If you want me to respond to the name-calling portion of your email, perhaps we should take it offlist. Thanks.) -- Matt Warden Cleveland, OH, USA http://mattwarden.com This email proudly and graciously contributes to entropy. From codepo8 at gmail.com Thu Dec 14 07:30:19 2006 From: codepo8 at gmail.com (Christian Heilmann) Date: Thu, 14 Dec 2006 13:30:19 +0000 Subject: [thelist] Tip: using PHP to collate and cache several JavaScript includes. Message-ID: <30bd6ffd0612140530q3f03871ds4b5c72c4253b91e7@mail.gmail.com> Ed Elliot came up with a cool PHP script that allows you to create a single cached include automatically from all the JavaScripts or CSS files you want to include. http://www.ejeliot.com/blog/72 For those who wonder why the hey you'd want to do this: working on high traffic sites means you learn a lot about how to make pages perform. Every HTTP request sent from page results in another DNS lookup and generally slows down the perceived loading time of the page. Furthermore, scripts loaded in the head of the document get loaded first and add to the whole page loading time as browsers start rendering after all the dependencies in the head were loaded. Using Ed's script means you can have the maintenance benefit of different includes without having to worry about the performance problems. -- Chris Heilmann Book: http://www.beginningjavascript.com Blog: http://www.wait-till-i.com Writing: http://icant.co.uk/ From edwin at bitstorm.org Thu Dec 14 07:49:12 2006 From: edwin at bitstorm.org (Edwin Martin) Date: Thu, 14 Dec 2006 14:49:12 +0100 Subject: [thelist] Workaround for setAttribute bug in IE7? In-Reply-To: References: <457CF780.8060004@ryanrushton.com> <457DE052.9000809@bitstorm.org> <610592c90612120213i1037cf1ds2c056b52267edafa@mail.gmail.com> <610592c90612130858s2d36c717s637f6ad55f64d5b0@mail.gmail.com> <610592c90612131358h470ecbfesed97277ba3ac1e03@mail.gmail.com> <610592c90612140220s4f0ee16pe071c542670dbdbc@mail.gmail.com> Message-ID: <45815658.9080505@bitstorm.org> Matt Warden wrote: > Try to make this simple: > > Clearly an operation manipulating the DOM: > input.setAttribute('size', 5); > > Clearly an assignment to an object property: > input.size = 5; > But since all modern browsers bind the properties to the DOM, what is the difference? Here is a list of all bindings: http://www.w3.org/TR/REC-DOM-Level-1/ecma-script-language-binding.html > Aside from my entire argument, you can do a simple google search to > see how often this issue trips up developers. > I haven't noticed any problems since IE6. Edwin Martin From mwarden at gmail.com Thu Dec 14 08:22:57 2006 From: mwarden at gmail.com (Matt Warden) Date: Thu, 14 Dec 2006 09:22:57 -0500 Subject: [thelist] Workaround for setAttribute bug in IE7? In-Reply-To: <45815658.9080505@bitstorm.org> References: <457CF780.8060004@ryanrushton.com> <610592c90612120213i1037cf1ds2c056b52267edafa@mail.gmail.com> <610592c90612130858s2d36c717s637f6ad55f64d5b0@mail.gmail.com> <610592c90612131358h470ecbfesed97277ba3ac1e03@mail.gmail.com> <610592c90612140220s4f0ee16pe071c542670dbdbc@mail.gmail.com> <45815658.9080505@bitstorm.org> Message-ID: On 12/14/06, Edwin Martin wrote: > > Clearly an operation manipulating the DOM: > > input.setAttribute('size', 5); > > > > Clearly an assignment to an object property: > > input.size = 5; > > > But since all modern browsers bind the properties to the DOM, what is > the difference? I'm not really sure how to answer that question. There is no "difference" in the result. We're talking about a design decision that does not make it clear that there are magical side effects to javascript assignment if (1) the property name is one that maps to a set of possible attribute names and (2) the DOM is representing an HTML document. > Here is a list of all bindings: > http://www.w3.org/TR/REC-DOM-Level-1/ecma-script-language-binding.html Yes, and you must know this list in order to recognize that this is a DOM manipulation: input.size = 5; and this is not: input.foo = "bar"; > > Aside from my entire argument, you can do a simple google search to > > see how often this issue trips up developers. > > > I haven't noticed any problems since IE6. No, not browser bugs. We're discussing the design decision. This design decision has resulted in a lot of confusion. Why doesn't label.for work? Oh, namespace clash. Why doesn't div.class work? etc. Because of this and how browsers work around it, our very own Andrew Clover suggests to abandon setAttribute/getAttribute altogether: http://lists.evolt.org/archive/Week-of-Mon-20050221/169835.html There would be no issue and no confusion if set/getAttribute were the only way to manipulate DOM attributes. Attribute names would be strings and there would be no issue of reserved words. The same issue exists with modifying the style directly via JS (avoiding the question of whether anyone should do this). What happens? Libraries pop up everywhere allowing you to do .setStyle('background-color', 'red'); http://developer.yahoo.com/yui/docs/YAHOO.util.Dom.html#setStyle -- Matt Warden Cleveland, OH, USA http://mattwarden.com This email proudly and graciously contributes to entropy. From lee.kowalkowski at googlemail.com Thu Dec 14 09:32:13 2006 From: lee.kowalkowski at googlemail.com (Lee kowalkowski) Date: Thu, 14 Dec 2006 15:32:13 +0000 Subject: [thelist] Workaround for setAttribute bug in IE7? In-Reply-To: References: <457CF780.8060004@ryanrushton.com> <610592c90612120213i1037cf1ds2c056b52267edafa@mail.gmail.com> <610592c90612130858s2d36c717s637f6ad55f64d5b0@mail.gmail.com> <610592c90612131358h470ecbfesed97277ba3ac1e03@mail.gmail.com> <610592c90612140220s4f0ee16pe071c542670dbdbc@mail.gmail.com> Message-ID: <610592c90612140732r73d6dcafy98cfe77dfcdb996b@mail.gmail.com> On 14/12/06, Matt Warden wrote: > Try to make this simple: > > Clearly an operation manipulating the DOM: > input.setAttribute('size', 5); I'm not deliberately overlooking your point. It's more like: Who would expect location.href=url to load a new document when location.assign(url) is a better match for the behaviour? I just think setAttribute is a poor example of the same argument, because setAttribute is merely an encapsulated assignment. I also think you're focusing too much on distinguishment between an attribute and a property; they're essentially synonyms for "variable". With a good JavaScript binding to the HTML DOM, the assignment operator is sufficiently overloaded to ensure there isn't any actual difference (huh, in *most* cases). > (If you want me to respond to the name-calling portion of your email, > perhaps we should take it offlist. Thanks.) Eh? What name-calling? *Making a note to oneself to never try to pay Matt a compliment ever again!*. It was public praise, enjoy it while it's available! Seriously - I hold your contribution to this list in the highest regard. Only a few contributors demonstrate they really know their stuff. You do it with impressive consistency. I call it genius. Keep it up. I'd like everybody else to know that I still respect *their* contributions! LOL, if they're still reading, that is. -- Lee From lee.kowalkowski at googlemail.com Thu Dec 14 10:40:26 2006 From: lee.kowalkowski at googlemail.com (Lee kowalkowski) Date: Thu, 14 Dec 2006 16:40:26 +0000 Subject: [thelist] Workaround for setAttribute bug in IE7? In-Reply-To: References: <457CF780.8060004@ryanrushton.com> <610592c90612130858s2d36c717s637f6ad55f64d5b0@mail.gmail.com> <610592c90612131358h470ecbfesed97277ba3ac1e03@mail.gmail.com> <610592c90612140220s4f0ee16pe071c542670dbdbc@mail.gmail.com> <45815658.9080505@bitstorm.org> Message-ID: <610592c90612140840t62abcdf8q5bd73b8ebad3a3e7@mail.gmail.com> On 14/12/06, Matt Warden wrote: > Yes, and you must know this list in order to recognize that this is a > DOM manipulation: > > input.size = 5; > > and this is not: > > input.foo = "bar"; Correct. This is where you can get IE to absolutely rock (well, almost, let's not get too carried away). In IE, input.foo = "bar" *is* a DOM manipulation, unless document.expando == false, in which case IE actually throws an exception when you try to do input.foo = "bar" *OR* input.setAttribute("foo", "bar"). This can be really handy for catching typos, but unfortunately, setAttribute lovers are at a disadvantage because IE then expects the non-standard setAttribute("className") to set the class attribute. Infuriating indeed! If you're stashing all sorts of arbitrary properties into your objects for your own needs, this would cause problems, but you can still use it for a section of DOM manipulation you're having trouble with: document.expando = false; ... // some DOM manipulation that doesn't appear to work document.expando = true; If you get an exception, the penny will finally drop (hopefully). -- Lee From webdev at mountain.ch Thu Dec 14 10:55:06 2006 From: webdev at mountain.ch (Mark Howells) Date: Thu, 14 Dec 2006 17:55:06 +0100 Subject: [thelist] Apache - alternative paths Message-ID: <159BBE78-C03A-485F-8B91-B0D680EE73AA@mountain.ch> Hi all Is it possible to configure Apache so that it looks for a file in a certain location and, when it doesn't find it, source it from a default folder? e.g. http://test.com/myfile.gif Search in /root/images/myfile.gif (not found) Serve file /root2/images/myfile.gif ??? Many thanks Mark Howells www.permanenttourist.ch From anthony at baratta.com Thu Dec 14 11:02:09 2006 From: anthony at baratta.com (Anthony Baratta) Date: Thu, 14 Dec 2006 09:02:09 -0800 Subject: [thelist] Apache - alternative paths Message-ID: <1d28d7f916d56cd1d411d8a1826c5d62@baratta.com> Setup your Error Doc parameter for file not found, either in that directory or server wide to run a PHP page that will serve the default file for you. It's possible that maybe the image itself can be your error document. http://httpd.apache.org/docs/2.0/custom-error.html -----Original message----- From: Mark Howells webdev at mountain.ch Date: Thu, 14 Dec 2006 08:55:06 -0800 To: Evolt List thelist at lists.evolt.org Subject: [thelist] Apache - alternative paths > Hi all > > Is it possible to configure Apache so that it looks for a file in a > certain location and, when it doesn't find it, source it from a > default folder? > > e.g. > > http://test.com/myfile.gif > > Search in /root/images/myfile.gif (not found) > Serve file /root2/images/myfile.gif > > ??? > > Many thanks > Mark Howells > www.permanenttourist.ch > -- > > * * Please support the community that supports you. * * > http://evolt.org/help_support_evolt/ > > For unsubscribe and other options, including the Tip Harvester > and archives of thelist go to: http://lists.evolt.org > Workers of the Web, evolt ! > From lists at neptunewebworks.com Thu Dec 14 11:11:15 2006 From: lists at neptunewebworks.com (Max Schwanekamp) Date: Thu, 14 Dec 2006 09:11:15 -0800 Subject: [thelist] Apache - alternative paths In-Reply-To: <159BBE78-C03A-485F-8B91-B0D680EE73AA@mountain.ch> References: <159BBE78-C03A-485F-8B91-B0D680EE73AA@mountain.ch> Message-ID: <458185B3.8050802@neptunewebworks.com> Mark Howells wrote: > Is it possible to configure Apache so that it looks for a file in a > certain location and, when it doesn't find it, source it from a > default folder? > e.g. > http://test.com/myfile.gif > > Search in /root/images/myfile.gif (not found) > Serve file /root2/images/myfile.gif You could use a 404 handler to do that. Traditionally a custom 404 page is a static page, but there's no reason it cannot be a script in your language of choice. David Sklar has a decent intro: http://www.onlamp.com/pub/a/onlamp/2003/02/13/davidsklar.html In your case, maybe you would load an array from config or db containing the default fallback locations and if the request matches one of these serve that up, and if not give an actual 404 response. -- Max Schwanekamp NeptuneWebworks.com 541-517-9064 From joel at spinhead.com Thu Dec 14 11:13:29 2006 From: joel at spinhead.com (Joel D Canfield) Date: Thu, 14 Dec 2006 09:13:29 -0800 Subject: [thelist] CSS again: suspected IE funkiness on inline UL Message-ID: <72E9FAA171D63B48AAC707C72900E6B4A719A2@ireland.spinhead.com> This sample works in both FF and IE7 http://css.maxdesign.com.au/listamatic/horizontal05.htm but mine breaks in IE http://www.intranetintelligence.com/ii/ The selected tab has no top border, bottom border is blue when it should be white (making it invisible.) I really really tried to figger it out myself, but it's not coming to me. Any pointers on where to look? thanks joel From webdev at mountain.ch Thu Dec 14 11:25:49 2006 From: webdev at mountain.ch (Mark Howells) Date: Thu, 14 Dec 2006 18:25:49 +0100 Subject: [thelist] Apache - alternative paths In-Reply-To: <458185B3.8050802@neptunewebworks.com> References: <159BBE78-C03A-485F-8B91-B0D680EE73AA@mountain.ch> <458185B3.8050802@neptunewebworks.com> Message-ID: > In your case, maybe you would load an array from config or db > containing > the default fallback locations and if the request matches one of these > serve that up, and if not give an actual 404 response. Thanks, Anthony and Max. This is a good idea, but the implementation would require that this script would be called so many times, that it may not be feasible. Esssentially, we'd call this script for every image, SWF, Javascript and CSS file served by 50 projects. We're intending that only straight page requests would be served directly from the called path. An example: we have image paths - /i/icon/logo.gif - which the same across multiple projects, each with their own domain. If the project doesn't have it's own, bespoke image in the images folder associated directly with the project, then the default image should be served from a main master project folder. An average page has four Javascripts, five to ten images and four CSS files, so we're talking about possibly calling this script from between 15 and 20 times per page request. Bearing in mind that we're currently experiencing a throughput of 500,000 requests per day, that would be a pretty heavy load for such a script. What we were hoping for would be an Apache directive, rather than an external script. Mark Howells www.permanenttourist.ch From anthony at baratta.com Thu Dec 14 11:30:10 2006 From: anthony at baratta.com (Anthony Baratta) Date: Thu, 14 Dec 2006 09:30:10 -0800 Subject: [thelist] Apache - alternative paths Message-ID: Mark... Using the ErrorDocument parameter would mean that the script or whatever is called only when the requested file does not exist. Not on every request. -----Original message----- From: Mark Howells webdev at mountain.ch Date: Thu, 14 Dec 2006 09:25:49 -0800 To: "thelist at lists.evolt.org" thelist at lists.evolt.org Subject: Re: [thelist] Apache - alternative paths > > In your case, maybe you would load an array from config or db > > containing > > the default fallback locations and if the request matches one of these > > serve that up, and if not give an actual 404 response. > > Thanks, Anthony and Max. This is a good idea, but the implementation > would require that this script would be called so many times, that it > may not be feasible. Esssentially, we'd call this script for every > image, SWF, Javascript and CSS file served by 50 projects. We're > intending that only straight page requests would be served directly > from the called path. > > An example: we have image paths - /i/icon/logo.gif - which the same > across multiple projects, each with their own domain. If the project > doesn't have it's own, bespoke image in the images folder associated > directly with the project, then the default image should be served > from a main master project folder. An average page has four > Javascripts, five to ten images and four CSS files, so we're talking > about possibly calling this script from between 15 and 20 times per > page request. Bearing in mind that we're currently experiencing a > throughput of 500,000 requests per day, that would be a pretty heavy > load for such a script. > > What we were hoping for would be an Apache directive, rather than an > external script. > > Mark Howells > www.permanenttourist.ch > > > -- > > * * Please support the community that supports you. * * > http://evolt.org/help_support_evolt/ > > For unsubscribe and other options, including the Tip Harvester > and archives of thelist go to: http://lists.evolt.org > Workers of the Web, evolt ! > From webdev at mountain.ch Thu Dec 14 11:35:13 2006 From: webdev at mountain.ch (Mark Howells) Date: Thu, 14 Dec 2006 18:35:13 +0100 Subject: [thelist] Apache - alternative paths In-Reply-To: References: Message-ID: Yes, I'd realized that, but (for example) every request for the global javascript library global.js would be passed to the script, as it's never found in the individual project. Multiple that by the volume of requests I've mentioned and a Javascript library alone of 120 separate files, and you'll see the worry that we have with the script being too laborious... Mark Howells www.permanenttourist.ch On 14.12.2006, at 18:30, Anthony Baratta wrote: > Mark... > > Using the ErrorDocument parameter would mean that the script or > whatever is called only when the requested file does not exist. Not > on every request. From lists at neptunewebworks.com Thu Dec 14 11:45:03 2006 From: lists at neptunewebworks.com (Max Schwanekamp) Date: Thu, 14 Dec 2006 09:45:03 -0800 Subject: [thelist] Apache - alternative paths In-Reply-To: References: Message-ID: <45818D9F.3020304@neptunewebworks.com> Mark Howells wrote: > Yes, I'd realized that, but (for example) every request for the > global javascript library global.js would be passed to the script, as > it's never found in the individual project. Multiple that by the > volume of requests I've mentioned and a Javascript library alone of > 120 separate files, and you'll see the worry that we have with the > script being too laborious... If you already know that a requested will never be there, why not just use modrewrite to handle that, then use an errordoc handler to deal with files that you cannot be sure about? -- Max Schwanekamp NeptuneWebworks.com 541-517-9064 From evolt at markgroen.com Thu Dec 14 11:54:39 2006 From: evolt at markgroen.com (Mark Groen) Date: Thu, 14 Dec 2006 09:54:39 -0800 Subject: [thelist] Apache - alternative paths In-Reply-To: <159BBE78-C03A-485F-8B91-B0D680EE73AA@mountain.ch> References: <159BBE78-C03A-485F-8B91-B0D680EE73AA@mountain.ch> Message-ID: <200612140954.39874.evolt@markgroen.com> On Thursday 14 December 2006 08:55, Mark Howells wrote: > Hi all > > Is it possible to configure Apache so that it looks for a file in a > certain location and, when it doesn't find it, source it from a > default folder? > > e.g. > > http://test.com/myfile.gif > > Search in /root/images/myfile.gif (not found) > Serve file /root2/images/myfile.gif > .............. > What we were hoping for would be an Apache directive, rather than an ? > external script. Sounds like what you want to use then is the "Alias" directive in your httpd.conf or config.d directory/file then? http://httpd.apache.org/docs/2.0/misc/custom_errordocs.html To check the docs for your particular version of Apache, simply replace the version number in the URI cheers, Mark -- cheers, mark From hassan.schroeder at gmail.com Thu Dec 14 11:56:46 2006 From: hassan.schroeder at gmail.com (Hassan Schroeder) Date: Thu, 14 Dec 2006 09:56:46 -0800 Subject: [thelist] Apache - alternative paths In-Reply-To: References: Message-ID: <4eedb92a0612140956k7b89b1d9qddf98a22735ac6e7@mail.gmail.com> On 12/14/06, Mark Howells wrote: > Yes, I'd realized that, but (for example) every request for the > global javascript library global.js would be passed to the script, as > it's never found in the individual project. Multiple that by the > volume of requests I've mentioned and a Javascript library alone of > 120 separate files, and you'll see the worry that we have with the > script being too laborious... Regardless of whether it's handled by an external script or an internal Apache routine, it seems to me that the biggest performance impact is the disk access required to determine that the target file isn't there. If that's going to be the case more often than not, I'd suggest a rethink :-) FWIW, -- Hassan Schroeder ------------------------ hassan.schroeder at gmail.com From evolt at markgroen.com Thu Dec 14 12:57:41 2006 From: evolt at markgroen.com (Mark Groen) Date: Thu, 14 Dec 2006 10:57:41 -0800 Subject: [thelist] Apache - alternative paths In-Reply-To: References: Message-ID: <200612141057.41947.evolt@markgroen.com> On Thursday 14 December 2006 09:30, Anthony Baratta wrote: > Mark... > > Using the ErrorDocument parameter would mean that the script or whatever is > called only when the requested file does not exist. Not on every request. But I think that is what the original post was asking for though: "...Apache so that it looks for a file in a ?certain location and, when it doesn't find it, source it from a ?default folder..." > > What we were hoping for would be an Apache directive, rather than an > > external script. Thought that might be a starting point though as the error directive needs to be hit first, but perhaps mod_rewrite rules can only be used in an .htaccess file? Never occurred to me, haven't ever tried it myself, but thought this might work (change the rewrite rules to what you need, copy/paste example below): Alias /error/ "/var/www/error/" Order allow,deny Allow from all RewriteCond /your/docroot/dir1/%{REQUEST_FILENAME} -f RewriteRule ^(.+) /your/docroot/dir1/$1 [L] # second try to find it in pub/... # ...and if found stop and be happy: RewriteCond /your/docroot/dir2/%{REQUEST_FILENAME} -f RewriteRule ^(.+) /your/docroot/dir2/$1 [L] # else go on for other Alias or ScriptAlias directives, # etc. RewriteRule ^(.+) - [PT] I haven't tested it myself, but the docs say that mod_rewrite should work in the httpd.conf as well: http://httpd.apache.org/docs/2.0/mod/mod_rewrite.html -- cheers, mark From john.destefano at gmail.com Thu Dec 14 13:00:40 2006 From: john.destefano at gmail.com (John DeStefano) Date: Thu, 14 Dec 2006 14:00:40 -0500 Subject: [thelist] rss feed pagination Message-ID: tim wrote: > > Anyone know of a quick & dirty way to clean up an RSS feed? There's a > > lot of data being sent from this particular feed, and it would be > > great to add pagination for easier navigation. > > Could you give us some more information? Are you producing the feed > yourself? Could you show us the XSLT you are using? > Sorry... not much information to go on, huh? We're using a user ticket-tracking system called RT [1]. There is very little information available regarding the RSS feed capabilities of the application [2], but one of the downsides is that a user must be logged in in order to view this information (or a user's user/pass can be included in the URL query string). I suggested that an anonymous account be created, so that its creds could be passed in such a URL for feeds, but the folks here have concerns about allowing unauthorized users any sort of access to CGI scripts (even just read ones, like this one). So, their solution is to develop a local script that generates the feed data hourly using a system account, and then pushes this data to a static page that can be viewed without an account. I found a web article [3] that gives some perl code that seems to do this, but I haven't been able to figure out how to turn the RDF output from the application into something the perl script can parse. And, as always, I assume there's likely a better|easier way to do this that I know nothing about! Since you have to log in in order to generate a feed (and since this is work stuff and not my own, personal stuff), I can't post a link to an example page. But I would be glad to generate one and email it in a file if that would be of any benefit. Anyway ... I just tried viewing some of this feed data in Firefox 2, and it seems that the browser has a mind of its own when it comes to displaying RSS feeds: the display completely ignored the stylesheet or transformation provided by the feed. The only RSS feed configuration options I see in FF2 are: - Show me a preview and ask me which Feed Reader to use. Or, - Subscribe to the feed using: [application] What about another option, like: - Display the feed using the transform provided by the feed [if any] I actually like the default display that Firefox now shows... but I'm not sure I like the fact that the original transformation seems to be ignored. Thanks, ~John [1] http://bestpractical.com/rt [2] http://wiki.bestpractical.com/index.cgi?RssFeed [3] http://www.ddj.com/showArticle.jhtml?documentID=new1013637407&pgno=8 From mwarden at gmail.com Thu Dec 14 13:23:33 2006 From: mwarden at gmail.com (Matt Warden) Date: Thu, 14 Dec 2006 14:23:33 -0500 Subject: [thelist] Workaround for setAttribute bug in IE7? In-Reply-To: <610592c90612140732r73d6dcafy98cfe77dfcdb996b@mail.gmail.com> References: <457CF780.8060004@ryanrushton.com> <610592c90612120213i1037cf1ds2c056b52267edafa@mail.gmail.com> <610592c90612130858s2d36c717s637f6ad55f64d5b0@mail.gmail.com> <610592c90612131358h470ecbfesed97277ba3ac1e03@mail.gmail.com> <610592c90612140220s4f0ee16pe071c542670dbdbc@mail.gmail.com> <610592c90612140732r73d6dcafy98cfe77dfcdb996b@mail.gmail.com> Message-ID: On 12/14/06, Lee kowalkowski wrote: > I'm not deliberately overlooking your point. It's more like: Who > would expect location.href=url to load a new document when > location.assign(url) is a better match for the behaviour? I just > think setAttribute is a poor example of the same argument, because > setAttribute is merely an encapsulated assignment. Absolutely correct. The side effect of location.href changes is much more severe and even more ridiculous. setAttribute was the topic, so that's what I was sticking with. But there are a lot of similar questionable decisions in JavaScript implementations, like using selectbox.selectedIndex = 3 to change the currently selected element. > I also think you're focusing too much on distinguishment between an > attribute and a property; they're essentially synonyms for "variable". > With a good JavaScript binding to the HTML DOM, the assignment > operator is sufficiently overloaded to ensure there isn't any actual > difference (huh, in *most* cases). If we were talking about a tree that existed wholly in the JavaScript world, I could perhaps agree. But we are talking about a tree that represents a document. Take a look at other OO DOM implementatons. Do they allow this? See: http://php.net/dom-domelement-setattribute An attribute and a property are the same thing. I totally agree. I was using the terms to make sure that I conveyed that I'm talking about a element attribute in the document rather than an object property in JavaScript. You're saying they are the same thing, and of course in this case they are. I'm just saying that's a bad idea, as (I would assume) the authors of other implementations of DOM agree. It is a mixing of semantics. If you operate on something (like, say, a document), it should be a method. (Let alone that good programming practice strongly suggests that even when getting and setting object properties one use accessors and mutators.) In your other email: > In IE, input.foo = "bar" *is* a DOM manipulation, unless > document.expando == false, in which case IE actually throws an > exception when you try to do input.foo = "bar" *OR* > input.setAttribute("foo", "bar"). ... > document.expando = false; > ... // some DOM manipulation that doesn't appear to work > document.expando = true; I must admit that I thought this was a joke. Expando? Sounds like half of a super hero's name (Captain Expando?). Thanks to a fruitful google search, it seems I'm just out of the loop: http://msdn.microsoft.com/workshop/author/dhtml/reference/properties/expando.asp Utterly weird. So, input.foo = "bar" literally creates an attribute *in the document* called foo? Then why doesn't selectbox.selectedIndex = 3 create an attribute in the document called selectedIndex? (rhetorical) There still seems to be a subset of object properties on node objects that don't map to an attribute in the DOM. I'm trying to get straight in my head when an object property wouldn't map, but I'm failing to find any kind of reasonable answer. Thanks for the tip on expando. Hadn't run across that before. -- Matt Warden Cleveland, OH, USA http://mattwarden.com This email proudly and graciously contributes to entropy. From beades at almonte.com Thu Dec 14 14:55:26 2006 From: beades at almonte.com (Brent Eades) Date: Thu, 14 Dec 2006 15:55:26 -0500 Subject: [thelist] rss feed pagination In-Reply-To: References: Message-ID: <4581BA3E.5010600@almonte.com> John DeStefano wrote: > Anyway ... I just tried viewing some of this feed data in > Firefox 2, and it seems that the browser has a mind of its own when it > comes to displaying RSS feeds: the display completely ignored the > stylesheet or transformation provided by the feed. The only RSS feed > configuration options I see in FF2 are: > - Show me a preview and ask me which Feed Reader to use. Or, > - Subscribe to the feed using: [application] That doesn't sound right. Here's a test RSS 1.0 feed with stylesheet reference I did up for a project I'm working on: http://centralbanks.org/press.xml FF2 should show you the XSL-transformed content of the feed, *plus* the options to subscribe to the feed as a Live Bookmark, Google Reader, etc. -- Brent Eades Almonte, Ontario http://almonte.com From lee.kowalkowski at googlemail.com Thu Dec 14 16:12:38 2006 From: lee.kowalkowski at googlemail.com (Lee kowalkowski) Date: Thu, 14 Dec 2006 22:12:38 +0000 Subject: [thelist] CSS again: suspected IE funkiness on inline UL In-Reply-To: <72E9FAA171D63B48AAC707C72900E6B4A719A2@ireland.spinhead.com> References: <72E9FAA171D63B48AAC707C72900E6B4A719A2@ireland.spinhead.com> Message-ID: <610592c90612141412l213ed494id8d6f3575d7be184@mail.gmail.com> On 14/12/06, Joel D Canfield wrote: > but mine breaks in IE > > http://www.intranetintelligence.com/ii/ > > The selected tab has no top border, bottom border is blue when it should > be white (making it invisible.) > > I really really tried to figger it out myself, but it's not coming to > me. > > Any pointers on where to look? Hmm, I'd just do this: No bottom border instead of a white one, relative position, shifted down a pixel to overlap the bottom border of the UL element, so in your .currentpage rules: border-bottom: 0; position:relative; top:1px; -- Lee From joel at spinhead.com Thu Dec 14 16:21:34 2006 From: joel at spinhead.com (Joel D Canfield) Date: Thu, 14 Dec 2006 14:21:34 -0800 Subject: [thelist] CSS again: suspected IE funkiness on inline UL Message-ID: <72E9FAA171D63B48AAC707C72900E6B4A719A9@ireland.spinhead.com> > No bottom border instead of a white one, relative position, shifted > down a pixel to overlap the bottom border of the UL element, so in > your .currentpage rules: > > border-bottom: 0; > position:relative; > top:1px; then, go look at it in FireFox, which gets the existing rules right :) yeah, tried that, hoping somehow FireFox would read my mind. It didn't. If I can only have it right in one browser, I'm stubborn enough to make it FireFox (at least this afternoon I am) and I'd really really prefer not to resort to IE hacks, but I also realize that, in the real world, sometimes that's what you do. I just don't know what that would be, and I don't even wanna know unless there's not a 'unhacky' fix. joel From rob at sanchothefat.com Thu Dec 14 17:17:52 2006 From: rob at sanchothefat.com (Rob O'Rourke) Date: Thu, 14 Dec 2006 23:17:52 +0000 Subject: [thelist] CSS again: suspected IE funkiness on inline UL In-Reply-To: <72E9FAA171D63B48AAC707C72900E6B4A719A9@ireland.spinhead.com> References: <72E9FAA171D63B48AAC707C72900E6B4A719A9@ireland.spinhead.com> Message-ID: <4581DBA0.1090904@sanchothefat.com> Joel D Canfield wrote: >> No bottom border instead of a white one, relative position, shifted >> down a pixel to overlap the bottom border of the UL element, so in >> your .currentpage rules: >> >> border-bottom: 0; >> position:relative; >> top:1px; >> > > then, go look at it in FireFox, which gets the existing rules right :) > > yeah, tried that, hoping somehow FireFox would read my mind. It didn't. > If I can only have it right in one browser, I'm stubborn enough to make > it FireFox (at least this afternoon I am) and I'd really really prefer > not to resort to IE hacks, but I also realize that, in the real world, > sometimes that's what you do. I just don't know what that would be, and > I don't even wanna know unless there's not a 'unhacky' fix. > > joel > Hi Joel, Unless you want that navlist centred I'd reccommend floating those list items left instead of displaying them as inline elements. That way you can treat the anchor tags as block level elements and things should be more consistent in IE. Take a look at some of the examples on listamatic for guidance: http://css.maxdesign.com.au/listamatic/ You will probably need to float the

    to contain the
  • s and then use a clear: both; rule on your main content area. Also watch out for IEs doubled float margin bug, use display: inline; on the
  • s themselves then display: block; on the links within. Good luck, Rob From lead at offlead.com Thu Dec 14 18:40:56 2006 From: lead at offlead.com (J.C. Johnson) Date: Thu, 14 Dec 2006 18:40:56 -0600 Subject: [thelist] subdomain or dns hijacking problem Message-ID: <012b01c71fe1$b13fa770$4f20400a@OffLead.local> Hello knowledgeable folks. I need some help pinpointing a problem. I've got a dedicated (managed) server, with a standard LAMP setup with cpanel, hosting 50 or 60 sites. All the sites are under the same IP address. We've just discovered a problem where someone is hijacking subdomains off of one of our domains. These subdomains aren't set up through cpanel, and don't show up there. We can find no signs of them on the server, and no files that don't belong are present. In fact, if you trace the subdomains they come up under a different IP address altogether. So where everything on our server is under ww.xx.yyy.zz, these rogue subdomains are coming up under aa.bbb.cc.ddd. The addresses are all with a format of www.sub.domain.com. If you try to pull up sub.domain.com, it does not resolve to anything. If you pull up www.sub.domain.com it goes off to a server in Korea. domain.com or www.domain.com resolve to our server. If someone can help me determine how this is being accomplished, and how to stop it, I would be very grateful. TIA! Jeniffer From beades at almonte.com Thu Dec 14 19:35:25 2006 From: beades at almonte.com (Brent Eades) Date: Thu, 14 Dec 2006 20:35:25 -0500 Subject: [thelist] rss feed pagination Message-ID: <4581FBDD.5080200@almonte.com> John DeStefano wrote: > Anyway ... I just tried viewing some of this feed data in > Firefox 2, and it seems that the browser has a mind of its own when it > comes to displaying RSS feeds: the display completely ignored the > stylesheet or transformation provided by the feed. The only RSS feed > configuration options I see in FF2 are: > - Show me a preview and ask me which Feed Reader to use. Or, > - Subscribe to the feed using: [application] That doesn't sound right. Here's a test RSS 1.0 feed with stylesheet reference I did up for a project I'm working on: http://centralbanks.org/press.xml FF2 should show you the XSL-transformed content of the feed, *plus* the options to subscribe to the feed as a Live Bookmark, Google Reader, etc. -- Brent Eades Almonte, Ontario http://almonte.com -- Brent Eades Almonte, Ontario http://almonte.com From lists at neptunewebworks.com Thu Dec 14 19:42:36 2006 From: lists at neptunewebworks.com (Max Schwanekamp) Date: Thu, 14 Dec 2006 17:42:36 -0800 Subject: [thelist] subdomain or dns hijacking problem In-Reply-To: <012b01c71fe1$b13fa770$4f20400a@OffLead.local> References: <012b01c71fe1$b13fa770$4f20400a@OffLead.local> Message-ID: <4581FD8C.60305@neptunewebworks.com> J.C. Johnson wrote: > Hello knowledgeable folks. I need some help pinpointing a problem. I've got > a dedicated (managed) server, with a standard LAMP setup with cpanel, > hosting 50 or 60 sites. All the sites are under the same IP address. We've > just discovered a problem where someone is hijacking subdomains off of one > of our domains. These subdomains aren't set up through cpanel, and don't > show up there. We can find no signs of them on the server, and no files that > don't belong are present. In fact, if you trace the subdomains they come up > under a different IP address altogether. So where everything on our server > is under ww.xx.yyy.zz, these rogue subdomains are coming up under > aa.bbb.cc.ddd. The addresses are all with a format of www.sub.domain.com. If > you try to pull up sub.domain.com, it does not resolve to anything. If you > pull up www.sub.domain.com it goes off to a server in Korea. domain.com or > www.domain.com resolve to our server. Is this the case for other users as well, or just for your local machine? I'm guessing you've tested on other computers, so it sounds like a compromised DNS. Is the DNS server for the domain(s) entirely under your control? When you create subdomains in cPanel, it auto-creates the www.sub.domain.tld entry as well as the regular sub.domain.tld. So, perhaps your zone files have been modified? The cPanel WHM "DNS Functions > Edit DNS Zone > [mydomain.com]" page AFAIK just parses the zone file and thus should be accurate, but if you want to be sure take a look at the zone files directly. On redhat, you'll find them in /var/named with names like 'mydomain.com.db' -- on another OS you may find them elsewhere. Of course, if you find that your zone files are compromised, you likely have a big ol' security problem that needs to be addressed immediately. But really, if this is a *managed* dedicated server, you ought to just contact the your server management provider for assistance. Hope this is at least slightly useful. -- Max Schwanekamp NeptuneWebworks.com From lead at offlead.com Thu Dec 14 20:44:33 2006 From: lead at offlead.com (J.C. Johnson) Date: Thu, 14 Dec 2006 20:44:33 -0600 Subject: [thelist] subdomain or dns hijacking problem In-Reply-To: <4581FD8C.60305@neptunewebworks.com> Message-ID: <015801c71ff2$f7720560$4f20400a@OffLead.local> Thanks Max. I wasn't entirely sure where to start on this, having run through the extent of my existing knowledge earlier this afternoon. Yes, one would think I could just turn this issue over to my hosting company. That's a discussion for a whole different thread, and we were already planning on moving to another company at the first of the year. Looks like our timeline is being moved up. Yes, this issue is occurring from any machine. In fact, I learned of the problem after being contacted by a law enforcement officer in another state yesterday. It was only when I went looking for the rogue subdomains that were supposedly on my machine that I discovered that they are not, in fact, on my machine. I've now looked at the zone file for this and other domains on my server. As far as I can tell, it's all the way it is supposed to be. I see only entries for mail, www and ftp, and then one entry for dev and one for www.dev, dev being the only subdomain I've actually set up myself on that domain. The IP address listed is correct for these entries. Jeniffer > -----Original Message----- > From: Max Schwanekamp [mailto:lists at neptunewebworks.com] > Sent: Thursday, December 14, 2006 7:43 PM > To: lead at offlead.com; thelist at lists.evolt.org > Subject: Re: [thelist] subdomain or dns hijacking problem > > J.C. Johnson wrote: > > Hello knowledgeable folks. I need some help pinpointing a problem. I've > got > > a dedicated (managed) server, with a standard LAMP setup with cpanel, > > hosting 50 or 60 sites. All the sites are under the same IP address. > We've > > just discovered a problem where someone is hijacking subdomains off of > one > > of our domains. These subdomains aren't set up through cpanel, and don't > > show up there. We can find no signs of them on the server, and no files > that > > don't belong are present. In fact, if you trace the subdomains they come > up > > under a different IP address altogether. So where everything on our > server > > is under ww.xx.yyy.zz, these rogue subdomains are coming up under > > aa.bbb.cc.ddd. The addresses are all with a format of > www.sub.domain.com. If > > you try to pull up sub.domain.com, it does not resolve to anything. If > you > > pull up www.sub.domain.com it goes off to a server in Korea. domain.com > or > > www.domain.com resolve to our server. > > Is this the case for other users as well, or just for your local > machine? I'm guessing you've tested on other computers, so it sounds > like a compromised DNS. Is the DNS server for the domain(s) entirely > under your control? When you create subdomains in cPanel, it > auto-creates the www.sub.domain.tld entry as well as the regular > sub.domain.tld. So, perhaps your zone files have been modified? The > cPanel WHM "DNS Functions > Edit DNS Zone > [mydomain.com]" page AFAIK > just parses the zone file and thus should be accurate, but if you want > to be sure take a look at the zone files directly. On redhat, you'll > find them in /var/named with names like 'mydomain.com.db' -- on another > OS you may find them elsewhere. > > Of course, if you find that your zone files are compromised, you likely > have a big ol' security problem that needs to be addressed immediately. > > But really, if this is a *managed* dedicated server, you ought to just > contact the your server management provider for assistance. > > Hope this is at least slightly useful. > > -- > Max Schwanekamp > NeptuneWebworks.com From lee.kowalkowski at googlemail.com Fri Dec 15 03:07:58 2006 From: lee.kowalkowski at googlemail.com (Lee kowalkowski) Date: Fri, 15 Dec 2006 09:07:58 +0000 Subject: [thelist] CSS again: suspected IE funkiness on inline UL In-Reply-To: <72E9FAA171D63B48AAC707C72900E6B4A719A9@ireland.spinhead.com> References: <72E9FAA171D63B48AAC707C72900E6B4A719A9@ireland.spinhead.com> Message-ID: <610592c90612150107x53be3362i9f17681145a750be@mail.gmail.com> On 14/12/06, Joel D Canfield wrote: > then, go look at it in FireFox, which gets the existing rules right :) Hmm, I did, and Opera and Netscape - they all looked the same, but I haven't got round to Firefox 2 yet, what versions are you using? -- Lee From webdev at mountain.ch Fri Dec 15 03:19:19 2006 From: webdev at mountain.ch (Mark Howells) Date: Fri, 15 Dec 2006 10:19:19 +0100 Subject: [thelist] Apache - alternative paths In-Reply-To: <4eedb92a0612140956k7b89b1d9qddf98a22735ac6e7@mail.gmail.com> References: <4eedb92a0612140956k7b89b1d9qddf98a22735ac6e7@mail.gmail.com> Message-ID: We implemented the functionality a couple of years ago on a 4th Dimension server, using a script, and it works well enough. We're just investigating the options open to us with an Apache set-up. I think we need to re-think the whole logic anyway; whilst it's very convenient to drop an alternative image into a folder and have it just work, we may be better off sacrificing flexibility for performance, as requests increase. Thanks to all for their suggestions. If there are any other good idea, I'd love to hear of them! Mark Howells www.permanenttourist.ch On 14.12.2006, at 18:56, Hassan Schroeder wrote: > Regardless of whether it's handled by an external script or an > internal > Apache routine, it seems to me that the biggest performance impact > is the disk access required to determine that the target file isn't > there. > > If that's going to be the case more often than not, I'd suggest a > rethink :-) From eduardok at gmail.com Fri Dec 15 03:54:50 2006 From: eduardok at gmail.com (Eduardo Kienetz) Date: Fri, 15 Dec 2006 07:54:50 -0200 Subject: [thelist] subdomain or dns hijacking problem In-Reply-To: <015801c71ff2$f7720560$4f20400a@OffLead.local> References: <4581FD8C.60305@neptunewebworks.com> <015801c71ff2$f7720560$4f20400a@OffLead.local> Message-ID: <3e7d042a0612150154m45c1879fv5650cba5dea9f734@mail.gmail.com> On 12/15/06, J.C. Johnson wrote: > Thanks Max. I wasn't entirely sure where to start on this, having run > through the extent of my existing knowledge earlier this afternoon. Yes, one > would think I could just turn this issue over to my hosting company. That's > a discussion for a whole different thread, and we were already planning on > moving to another company at the first of the year. Looks like our timeline > is being moved up. > > Yes, this issue is occurring from any machine. In fact, I learned of the > problem after being contacted by a law enforcement officer in another state > yesterday. It was only when I went looking for the rogue subdomains that > were supposedly on my machine that I discovered that they are not, in fact, > on my machine. > > I've now looked at the zone file for this and other domains on my server. As > far as I can tell, it's all the way it is supposed to be. I see only entries > for mail, www and ftp, and then one entry for dev and one for www.dev, dev > being the only subdomain I've actually set up myself on that domain. The IP > address listed is correct for these entries. > > Jeniffer Please read: http://en.wikipedia.org/wiki/DNS_cache_poisoning -- Eduardo Bacchi Kienetz LPI Certified - Level 2 http://www.noticiaslinux.com.br/eduardo/ From morrison.ben at gmail.com Fri Dec 15 04:47:13 2006 From: morrison.ben at gmail.com (ben morrison) Date: Fri, 15 Dec 2006 10:47:13 +0000 Subject: [thelist] CSS again: suspected IE funkiness on inline UL In-Reply-To: <4581DBA0.1090904@sanchothefat.com> References: <72E9FAA171D63B48AAC707C72900E6B4A719A9@ireland.spinhead.com> <4581DBA0.1090904@sanchothefat.com> Message-ID: <6073aef90612150247n698265dfy21cbfed4cda4eb83@mail.gmail.com> On 12/14/06, Rob O'Rourke wrote: > Joel D Canfield wrote: > >> No bottom border instead of a white one, relative position, shifted > >> down a pixel to overlap the bottom border of the UL element, so in > >> your .currentpage rules: > >> > >> border-bottom: 0; > >> position:relative; > >> top:1px; > >> > > > > then, go look at it in FireFox, which gets the existing rules right :) > > > > yeah, tried that, hoping somehow FireFox would read my mind. It didn't. > > If I can only have it right in one browser, I'm stubborn enough to make > > it FireFox (at least this afternoon I am) and I'd really really prefer > > not to resort to IE hacks, but I also realize that, in the real world, > > sometimes that's what you do. I just don't know what that would be, and > > I don't even wanna know unless there's not a 'unhacky' fix. > > > > joel > > > > Hi Joel, > > Unless you want that navlist centred I'd reccommend floating those list > items left instead of displaying them as inline elements. That way you > can treat the anchor tags as block level elements and things should be > more consistent in IE. > > Take a look at some of the examples on listamatic for guidance: > http://css.maxdesign.com.au/listamatic/ > > You will probably need to float the
      to contain the
    • s and then > use a clear: both; rule on your main content area. > > Also watch out for IEs doubled float margin bug, use display: inline; on > the
    • s themselves then display: block; on the links within. I'd agree with this approach as well, floating them gives you much more control over the layout. To achieve the bottom border you may find it easier to use a background image. I've knocked up a quick example here: http://html.eval.poptech.coop/ben/navFix.htm ben -- Ben Morrison From hassan.schroeder at gmail.com Fri Dec 15 06:44:39 2006 From: hassan.schroeder at gmail.com (Hassan Schroeder) Date: Fri, 15 Dec 2006 04:44:39 -0800 Subject: [thelist] Apache - alternative paths In-Reply-To: References: <4eedb92a0612140956k7b89b1d9qddf98a22735ac6e7@mail.gmail.com> Message-ID: <4eedb92a0612150444l2407a3fp5d4be83256987f1a@mail.gmail.com> On 12/15/06, Mark Howells wrote: > just investigating the options open to us with an Apache set-up. You probably want to look at mod_filter and mod_ext_filter, then. It's also possible mod_negotiation/MultiViews could be applied to this. FWIW! -- Hassan Schroeder ------------------------ hassan.schroeder at gmail.com From John.Brooking at sappi.com Fri Dec 15 08:21:31 2006 From: John.Brooking at sappi.com (Brooking, John) Date: Fri, 15 Dec 2006 09:21:31 -0500 Subject: [thelist] Proper commenting and code maintainance Message-ID: <7DFF91610114AB4C8C0B068CFC34734F08DEC065@CDFNEXC3.NA.Sappi.com> tim: > My version would be: > > // Returns true if the key stroke is allowed > function CheckNumericKeyStroke() > { > var validKeys = "1234567890"; > var pressedKey = String.fromCharCode(event.keyCode); > > return (validKeys.indexOf(pressedKey) != -1); > } > > Rather than: > > function CheckNumericKeyStroke() > { > var strValidKeys = "1234567890"; > > // if there keyCode is not one of the numbers in validkeys... > if (strValidKeys.indexOf(String.fromCharCode(event.keyCode)) == -1) { > return false; // do not allow the keystroke > } > else { > return true; // allow the keystroke > } > } That reminds me of another best practice I learned so long ago it's almost subconscious, that a function should only have one return statement. So for instance if your function could return an error number at a variety of points, it's clearer to set a variable to the error number, use conditionals if necessary to prevent the rest of the function from executing if that variable is set, then return the variable once at the end. - John -- This message may contain information which is private, privileged or confidential and is intended solely for the use of the individual or entity named in the message. If you are not the intended recipient of this message, please notify the sender thereof and destroy / delete the message. Neither the sender nor Sappi Limited (including its subsidiaries and associated companies) shall incur any liability resulting directly or indirectly from accessing any of the attached files which may contain a virus or the like. From mwarden at gmail.com Fri Dec 15 09:09:38 2006 From: mwarden at gmail.com (Matt Warden) Date: Fri, 15 Dec 2006 10:09:38 -0500 Subject: [thelist] Proper commenting and code maintainance In-Reply-To: <7DFF91610114AB4C8C0B068CFC34734F08DEC065@CDFNEXC3.NA.Sappi.com> References: <7DFF91610114AB4C8C0B068CFC34734F08DEC065@CDFNEXC3.NA.Sappi.com> Message-ID: On 12/15/06, Brooking, John wrote: > That reminds me of another best practice I learned so long ago it's > almost subconscious, that a function should only have one return > statement. So for instance if your function could return an error number > at a variety of points, it's clearer to set a variable to the error > number, use conditionals if necessary to prevent the rest of the > function from executing if that variable is set, then return the > variable once at the end. Agreed. Perhaps an exception (no pun intended): In JavaScript it's common to see this: foo: function(arg1, arg2) { if (!document.getElementById) return false; if (!arg1) return false; if (!arg2) return false; // ... real code here ... } which is an almost self-documenting way of coding preconditions for the method. (Not to say documentation isn't necessary, just that the code is highly readable.) -- Matt Warden Cleveland, OH, USA http://mattwarden.com This email proudly and graciously contributes to entropy. From ontheroad at frii.com Fri Dec 15 20:10:08 2006 From: ontheroad at frii.com (Bob Meetin) Date: Fri, 15 Dec 2006 19:10:08 -0700 Subject: [thelist] web-calendar Message-ID: <45835580.2060302@frii.com> I'm trying to get web-calendar installed. I've done this before on several different hosts, but not successfully today. I have tried both versions 1.0.3 and 1.0.4. Both fail with the same errors. I have verified that the host/login/password are correct. The error I am getting is: Error finding WebCalendar tables in database '$database' using db login '$whatever' on db server 'sql'. Then it asks if you have created the required tables which I have. Replace $database and $whatever with the actual names. This is MySQL. The only difference from other sites where I have installed the application is here they use 'sql' as the host rather than 'localhost'. When I search this error on google I find a number of installs displaying the same error message. Any ideas? http://sourceforge.net/projects/webcalendar/ -Bob From ontheroad at frii.com Fri Dec 15 20:45:19 2006 From: ontheroad at frii.com (Bob Meetin) Date: Fri, 15 Dec 2006 19:45:19 -0700 Subject: [thelist] web-calendar In-Reply-To: <45835580.2060302@frii.com> References: <45835580.2060302@frii.com> Message-ID: <45835DBF.1090104@frii.com> Never mind - got it - my mistake. Not all the tables had gotten created correctly - I verified by looking at a previous install and identifying the missing culprits. -Bob > I'm trying to get web-calendar installed. I've done this before on > several different hosts, but not successfully today. I have tried both > versions 1.0.3 and 1.0.4. Both fail with the same errors. I have > verified that the host/login/password are correct. The error I am > getting is: > > Error finding WebCalendar tables in database '$database' using db login > '$whatever' on db server 'sql'. Then it asks if you have created the > required tables which I have. Replace $database and $whatever with the > actual names. > > This is MySQL. The only difference from other sites where I have > installed the application is here they use 'sql' as the host rather than > 'localhost'. > > When I search this error on google I find a number of installs > displaying the same error message. Any ideas? > > http://sourceforge.net/projects/webcalendar/ > > -Bob > > From meredith at pintsize.com Sat Dec 16 08:51:38 2006 From: meredith at pintsize.com (Meredith Tupper) Date: Sat, 16 Dec 2006 08:51:38 -0600 Subject: [thelist] ftp client for Mac Message-ID: <20061216085138.03zkg7em9lg8ccco@66.135.39.9> I'm using Cyberduck for now and while it's not bad, I'm looking for something more robust. What's the best Mac ftp client for heavy use? -- PintSize Graphics & Web Hosting, Inc. http://www.pintsize.com 888-875-1082 From toby at creativefibre.co.uk Sat Dec 16 09:26:54 2006 From: toby at creativefibre.co.uk (Toby) Date: Sat, 16 Dec 2006 15:26:54 -0000 Subject: [thelist] ftp client for Mac In-Reply-To: <20061216085138.03zkg7em9lg8ccco@66.135.39.9> Message-ID: <200612161526.kBGFQVjK082769@mail.authsmtp.com> Transmit might be okay, but I have never used it for heavy transfers. Its got a cool icon, that normal does it for mac users ;] http://www.panic.com/transmit/ > -----Original Message----- > From: thelist-bounces at lists.evolt.org [mailto:thelist- > bounces at lists.evolt.org] On Behalf Of Meredith Tupper > Sent: 16 December 2006 14:52 > To: thelist at lists.evolt.org > Subject: [thelist] ftp client for Mac > > I'm using Cyberduck for now and while it's not bad, I'm looking for > something more robust. What's the best Mac ftp client for heavy use? > > -- > PintSize Graphics & Web Hosting, Inc. > http://www.pintsize.com > 888-875-1082 > -- > > * * Please support the community that supports you. * * > http://evolt.org/help_support_evolt/ > > For unsubscribe and other options, including the Tip Harvester > and archives of thelist go to: http://lists.evolt.org > Workers of the Web, evolt ! From steven.pierce at gmail.com Sat Dec 16 09:48:43 2006 From: steven.pierce at gmail.com (Steven Pierce) Date: Sat, 16 Dec 2006 07:48:43 -0800 Subject: [thelist] ftp client for Mac In-Reply-To: <20061216085138.03zkg7em9lg8ccco@66.135.39.9> References: <20061216085138.03zkg7em9lg8ccco@66.135.39.9> Message-ID: <6e0cf1cc0612160748w1888eaa7g8243cbbfac842aef@mail.gmail.com> Depends on what you mean by heavy use. I have not seen a FTP Client that could not be used for almost anything you want to. Heavy use it only what the user does, and since it almost all user base, I think it should be good. I currently use Transmit for my work, and I really like it. On 12/16/06, Meredith Tupper wrote: > > I'm using Cyberduck for now and while it's not bad, I'm looking for > something more robust. What's the best Mac ftp client for heavy use? > > -- > PintSize Graphics & Web Hosting, Inc. > http://www.pintsize.com > 888-875-1082 > -- > > * * Please support the community that supports you. * * > http://evolt.org/help_support_evolt/ > > For unsubscribe and other options, including the Tip Harvester > and archives of thelist go to: http://lists.evolt.org > Workers of the Web, evolt ! > From warren0728 at gmail.com Sat Dec 16 10:00:20 2006 From: warren0728 at gmail.com (ww) Date: Sat, 16 Dec 2006 11:00:20 -0500 Subject: [thelist] ftp client for Mac In-Reply-To: <6e0cf1cc0612160748w1888eaa7g8243cbbfac842aef@mail.gmail.com> References: <20061216085138.03zkg7em9lg8ccco@66.135.39.9> <6e0cf1cc0612160748w1888eaa7g8243cbbfac842aef@mail.gmail.com> Message-ID: <29C93777-3BFC-4479-BBE1-5633CAEF73B4@gmail.com> i have used fetch for years.... ww > > On 12/16/06, Meredith Tupper wrote: >> >> I'm using Cyberduck for now and while it's not bad, I'm looking for >> something more robust. What's the best Mac ftp client for heavy use? >> From bglassman at gmail.com Sat Dec 16 16:35:58 2006 From: bglassman at gmail.com (Ben Glassman) Date: Sat, 16 Dec 2006 17:35:58 -0500 Subject: [thelist] ftp client for Mac In-Reply-To: <20061216085138.03zkg7em9lg8ccco@66.135.39.9> References: <20061216085138.03zkg7em9lg8ccco@66.135.39.9> Message-ID: <51a158b10612161435i60fe3141r3c23d498c5f610ec@mail.gmail.com> I am a big fan of Transmit. It has a Docksend option where if you have defined a ftp site with a corresponding local folder you can just drag files from finder inside of that folder to the Transmit icon and it will upload them automatically to the correct directory on the ftp server. I use this in conjunction with BBEdit all the time because you can drag files from the drawer/sidebar of BBEdit to the Transmit Icon and upload them. Ben On 12/16/06, Meredith Tupper wrote: > > I'm using Cyberduck for now and while it's not bad, I'm looking for > something more robust. What's the best Mac ftp client for heavy use? > > -- > PintSize Graphics & Web Hosting, Inc. > http://www.pintsize.com > 888-875-1082 > -- > > * * Please support the community that supports you. * * > http://evolt.org/help_support_evolt/ > > For unsubscribe and other options, including the Tip Harvester > and archives of thelist go to: http://lists.evolt.org > Workers of the Web, evolt ! > From evolt at markgroen.com Sat Dec 16 20:24:10 2006 From: evolt at markgroen.com (Mark Groen) Date: Sat, 16 Dec 2006 18:24:10 -0800 Subject: [thelist] Apache - alternative paths In-Reply-To: References: <4eedb92a0612140956k7b89b1d9qddf98a22735ac6e7@mail.gmail.com> Message-ID: <200612161824.10576.evolt@markgroen.com> On Friday 15 December 2006 01:19, Mark Howells wrote: > We implemented the functionality a couple of years ago on a 4th > Dimension server, using a script, and it works well enough. We're > just investigating the options open to us with an Apache set-up. I > think we need to re-think the whole logic anyway; whilst it's very > convenient to drop an alternative image into a folder and have it > just work, we may be better off sacrificing flexibility for > performance, as requests increase. > > Thanks to all for their suggestions. If there are any other good > idea, I'd love to hear of them! It's rather an old article, but the info is still fairly relevant since when I bookmarked it: http://www.serverwatch.com/tutorials/article.php/3436911 -- cheers, mark From volkan.ozcelik at gmail.com Sat Dec 16 23:40:27 2006 From: volkan.ozcelik at gmail.com (=?ISO-8859-9?Q?VOLKAN_=D6Z=C7EL=DDK?=) Date: Sun, 17 Dec 2006 07:40:27 +0200 Subject: [thelist] Slightly [OT] - Think Twice When Migrating to Blogger Beta Message-ID: Especially - if you have localized content - and you have a large blog Or be prepared to share my suffering [1]. [1] http://www.volkanozcelik.com/cre8/blog/2006/12/this-is-test-post-please-ignore.html And here are two tips the OT: two quick php tips: In php, if your rendered (X)HTML output includes uncessarry whitespace at the beginning of the buffer (which makes it an invalid XML) make sure that you trimmed all whitespace from the files imported with - include - require - require_once etc. Extra whitespace can be a real pain in php. It can cause errors if you have a strict XML (RSS) feed, and it can cause render anomailes in some browsers (like MSIE). Actually most of text editors and IDEs has some kind of a "trim trailing space" option. If your php page occasionally sends "header information already sent" warning try including ob_start at the beginning of your file(s) and ob_end_flush() at the end. Here's an example: < ?php function callback($buffer) { // replace all the apples with oranges return (str_replace("apples", "oranges", $buffer)); } ob_start("callback"); ? > < html> < body> < p>It's like comparing apples to oranges.< /p> < / body> < / html> < ? php ob_end_flush(); ? > Cheers, -- Volkan Ozcelik +>Yep! I'm blogging! : http://www.volkanozcelik.com/cre8/blog/ +> Going solo in Turkish: http://www.volkanozcelik.com/donkisot/ +> My projects/studies/trials/errors : http://www.sarmal.com/ +> Sardalya JavaScript Library: http://www.sarmal.com/sardalya/ From joel at spinhead.com Sun Dec 17 18:06:04 2006 From: joel at spinhead.com (Joel D Canfield) Date: Sun, 17 Dec 2006 16:06:04 -0800 Subject: [thelist] CSS again: suspected IE funkiness on inline UL Message-ID: <72E9FAA171D63B48AAC707C72900E6B4A719CC@ireland.spinhead.com> > Unless you want that navlist centred I'd reccommend floating > those list > items left instead of displaying them as inline elements. > That way you > can treat the anchor tags as block level elements and things > should be > more consistent in IE. Theoretically interesting, but none of this fixed it :( > Take a look at some of the examples on listamatic for guidance: > http://css.maxdesign.com.au/listamatic/ That's where my code came from, here: http://css.maxdesign.com.au/listamatic/horizontal05.htm Which works exactly as I expect in both FF and IE, so I strongly suspect the issue is somewhere else in the CSS of the header; I just can't find it. I do appreciate the continuing education - every one of y'all's posts has had some useful tidbit that's gonna come in handy down the road. joel From ontheroad at frii.com Sun Dec 17 21:22:58 2006 From: ontheroad at frii.com (Bob Meetin) Date: Sun, 17 Dec 2006 20:22:58 -0700 Subject: [thelist] test environment for zen-cart Message-ID: <45860992.4060105@frii.com> To create a test environment for a site I just inherited I: * create a staging subdirectory, * copy the package into that subdirectory * update the configure.php file to reference the new location * create a backup copy of the database, but with a different prefix (stage_, etc) * update configure.php to use the backup database/tables This seems to work fine on a demo version (not client's) I have set up for playing around. However with the client's install if I visit the staging subdirectory home page (the shop home page), I get the following message in Firefox (IE just hangs): ---------------------------------------------------------------------------------------- The page isn't redirecting properly Firefox has detected that the server is redirecting the request for this address in a way that will never complete. * This problem can sometimes be caused by disabling or refusing to accept cookies. ---------------------------------------------------------------------------------------- Suggestions? -Bob
Item List
id="10490" value="10490" />15410490 href="itemDescription.do?action=getItems&display=details&items=10490" > onclick="toggleSummary('10490'); return false;">Item One12/8/06ActiveJohn Doeedit