From dave at members.evolt.org Thu Mar 1 00:52:40 2001 From: dave at members.evolt.org (dave mclean) Date: Thu Mar 1 00:52:40 2001 Subject: [thesite] Hey! Ho! Let's Go! Hey! Ho! AvantGo! In-Reply-To: <3A9C58CD.30311.6124433@localhost> Message-ID: On 2/27/01 11:47 PM you wrote: >> Oh yeah, you should see aardvark's article on the Palm. The screen >> caps come out nicely, but not exactly readable. Still cool. > > i wanna see !!! got a digital camera? I'll bring the Plam to austin... how's that sound? ___________________________________________________ http://evolt.org - We put the 'rrr' in 'werd' baby! Yeah baby, yeah!!! From roselli at earthlink.net Thu Mar 1 01:35:10 2001 From: roselli at earthlink.net (aardvark) Date: Thu Mar 1 01:35:10 2001 Subject: [thesite] Hey! Ho! Let's Go! Hey! Ho! AvantGo! In-Reply-To: References: <3A9C58CD.30311.6124433@localhost> Message-ID: <3A9DB542.11742.B63A27D@localhost> > From: dave mclean > > >> Oh yeah, you should see aardvark's article on the Palm. The screen > >> caps come out nicely, but not exactly readable. Still cool. > > > > i wanna see !!! got a digital camera? > > I'll bring the Plam to austin... how's that sound? sexy... tell Rosie to bring her five sisters, too... From dave at members.evolt.org Thu Mar 1 01:40:55 2001 From: dave at members.evolt.org (dave mclean) Date: Thu Mar 1 01:40:55 2001 Subject: [thesite] Hey! Ho! Let's Go! Hey! Ho! AvantGo! In-Reply-To: <3A9DB542.11742.B63A27D@localhost> Message-ID: On 3/1/01 12:34 AM you wrote: >>> i wanna see !!! got a digital camera? >> >> I'll bring the Plam to austin... how's that sound? > > sexy... tell Rosie to bring her five sisters, too... What list is this again? From sgd at ti3.com Thu Mar 1 11:20:18 2001 From: sgd at ti3.com (Scott Dexter) Date: Thu Mar 1 11:20:18 2001 Subject: [thesite] Hey! Ho! Let's Go! Hey! Ho! AvantGo! Message-ID: <8C8B8745C0FE7A43BFA6CABAB321117223FF64@gate.ti3.com> > > > sexy... tell Rosie to bring her five sisters, too... > > What list is this again? > I know. This one and the Dog and Pony show in another one has got me kinda worried about Austin.... "Hey man, I'm only here for the beer...." sgd From djc at starkmedia.com Thu Mar 1 18:30:30 2001 From: djc at starkmedia.com (Daniel J. Cody) Date: Thu Mar 1 18:30:30 2001 Subject: [thesite] Feb stats.. Message-ID: <3A9EE971.4030105@starkmedia.com> Thought I'd cut those rowdy chat'sters in on this too.. ;) Febuary stats are in and fresh from the oven. I decided to ditch Webtrends this month because its a piece of shit. Went back to good ole' command line interface'in, uses plus signs for graph'in, html 2.0 conformant -- Analog. I remember someone complaining a while back about the reports not being specific enough, so I turned on every piece of f*!@#in reporting I could find. As a result, you now have more data about our traffic than you could ever want. Added members.evolt.org logs files to the mix(they've been on since mid feb, so nota true hueristic but nice to have).. http://browsers.evolt.org/stats/stats.feb2001.html One BIG note: You *have* to add about another 40-60K page views to 'successful page requests' because every page on w.e.o is *not* a successful page view in the logs. each one is a 404. I figure the correct amount of page views for Feb to be right under 200K and the correct amount of 'hits' at about 750K - this is rough estimation but I dont feel like whipping out my abacus. And holy shit is Google our friend! general thoughts: adrians 'building our site' and martins 'cms' articles got the most page views for single aritcles peter vd's 'usability' article was in a close third with 4300 page views i'm not sure whats going on here: http://members.evolt.org/anarion/enterdream/enterdream.php but it pulled down almost 4k page views in under 5 days.. *shrug* interestingly, http://evolt.org/article/headlines/25/5281/index.html got almost 2000 views. interesting because the 'headlines' filler in the URL only gets posted to thelist from the 'articles of the day' email which 1.) hasn't been working for 2 months 2.) only gets sent out to 1600 people in the first place ;) From roselli at earthlink.net Thu Mar 1 20:38:04 2001 From: roselli at earthlink.net (aardvark) Date: Thu Mar 1 20:38:04 2001 Subject: [thesite] Re: Read the article, and a usability issue with the site In-Reply-To: References: <3A9E0F5E.15081.CC3AD66@localhost> Message-ID: <3A9EC119.31611.F7A090C@localhost> Ben, I hope you don't mind, but I'm bringing this over to thesite (the list devoted to discussions of, well, the site) so that some others can have some input. I think your points are good points, and I'd rather not make unilateral assertions without letting other people take a whack at my logic. > > Yeah, we use 'accesskey' attributes on the form elements to meet WAI > > and Bobby accessbility standards. > > After a few minutes' thought this occurred to me. It's fantastic > thing to do in principle, and I support it enthusiastically. Good. Dammit. > I just question the wisdom of assimilating a macro that's already set > aside for a function available to at least two-thirds of the site's > users, since the same alt-key-combo is used in IE4 as well. And those > with impaired mobiilty will definitely use the macro in question (as I > know firsthand, having discussed the subject some time back with a > quadriplegic who used to - and may still - maintain a Geocities page). > > I personally am not impaired - I just like to keep both hands on the > keyboard unless I'm just-browsing, and even then I often use the > keyboard to navigate. Agreed. While my own testing with impaired users hasn't shown this, that suggests I need to do more testing. One thing we might want to consider is gathering the alt-key combos from the major browsers and making those off-limits. But that begs the question, what if we run out of letters? Do we dump the accesskey attributes? And who can we blame for this, browsers or us? I need blame to made. > > Unfortunately, we can't account for the how the browsers handle it, > > and we also can't account for every letter in the alphabet > > considering all the browsers out there. > > I suppose not, but one would think on the grounds of common sense that > the Alt-key macros for NN & IE menu items would be used cautiously. Agreed, but that's the thing about common sense, it's an oxymoron. Either way, yes, I did consider that, but clearly my decision was based on my own perceptions. So, is it better to kill the accesskeys or force users to use the mouse? I don't know that answer... > > We discussed this one for a while, and figured we'd try it. You're > > the first to specifically comment. > > ...Which bothers me, but that's another thought for another time. That you're the first to comment on it? Anyway, verboten letters from the windows browsers i use regularly: IE5: f, e, v, a, t, h N6: f, e, v, s, g, b, t, h N3: f, e, v, g, b, o, d, w, h O5: f, e, v, n, b, m, g, s, w, n Any others? From r937 at interlog.com Thu Mar 1 21:15:41 2001 From: r937 at interlog.com (rudy) Date: Thu Mar 1 21:15:41 2001 Subject: [thesite] Feb stats.. Message-ID: <01c0a2c6$d89f0b60$ba51149a@rudy> >http://browsers.evolt.org/stats/stats.feb2001.html thanks dan as usual, more info than i can swallow in one sitting thanks for the Browser Report, i'm going to add those user agent strings to the ones i got offlist yesterday (thanks aardvark) some truly amazing nuggets buried in there, the non-browsers, crawlers, spiders and bots, for instance -- 7541: Mercator-Scrub-1.1 5916: Mozilla/3.01 (compatible;) 5264: ZyBorg/1.0 (ZyBorg at WISEnutbot.com; http://www.WISEnutbot.com) 3526: FAST-WebCrawler/2.2-pre41 2028: Teleport Pro/1.29.1590 2021: Come to papa v1.0 1932: ia_archiver 1732: Gulliver/1.3 1688: Googlebot/2.1 and get this -- not listed: 4,810 browsers!!! `:-o how the heck are you supposed to write a browser sniff routine if there's 5,000 browsers out there??!! i also liked the Search Word Report, Listing the top 20 query words by the number of requests, which sounds like a manic pseudo-haiku we all might chant during a nightmare... "javascript netscape asp in and to flash form css dreamweaver html submit php sql browser mac redirect download templates window" rudy From erika at seastorm.com Thu Mar 1 22:36:04 2001 From: erika at seastorm.com (Erika Meyer) Date: Thu Mar 1 22:36:04 2001 Subject: [thesite] Hey! Ho! Let's Go! Hey! Ho! AvantGo! In-Reply-To: <8C8B8745C0FE7A43BFA6CABAB321117223FF64@gate.ti3.com> References: <8C8B8745C0FE7A43BFA6CABAB321117223FF64@gate.ti3.com> Message-ID: >I know. This one and the Dog and Pony show in another one has got me kinda >worried about Austin.... what's a Dog and Pony show? yadda yadda hey! erika at seastorm.com From john at userfrenzy.com Fri Mar 2 05:19:49 2001 From: john at userfrenzy.com (John Handelaar) Date: Fri Mar 2 05:19:49 2001 Subject: [thesite] Feb stats.. In-Reply-To: <3A9EE971.4030105@starkmedia.com> Message-ID: Wow. Google really *likes* evolt. (It'd be easier to read, btw, if you included the images which the graphs use, but never mind for now) Some interesting points: 1 You did way more PIs per day in the last 7 days than over the whole month. 2 On the same note, your hits vs PIs ratio altered from 5:1 to around 3:1 during that last week. 3 Time-of-day graph suggests the Europeans use thesite more than anyone else (assuming the time zone is correct on the report). 4 Netscape does a lot better on evolt than it does on any sites I monitor :-) Oh, look: Macs too. 5 1500 favicons! In one month! If you'd like me to have a look at the analog.cfg file there is a way to make weo's 404s show up as PIs. I think... ------------------------------------------ John Handelaar T +44 20 7209 4117 M +44 7930 681789 F +44 870 169 7657 E john at userfrenzy.com ------------------------------------------ From persist1 at io.com Fri Mar 2 07:18:19 2001 From: persist1 at io.com (Ben Henick) Date: Fri Mar 2 07:18:19 2001 Subject: [thesite] Re: Read the article, and a usability issue with the site In-Reply-To: <3A9EC119.31611.F7A090C@localhost> Message-ID: On Thu, 1 Mar 2001, aardvark wrote: > > Ben, I hope you don't mind, but I'm bringing this over to thesite (the > list devoted to discussions of, well, the site) so that some others > can have some input. I think your points are good points, and I'd > rather not make unilateral assertions without letting other people > take a whack at my logic. Go for it. > Good. Dammit. *chuckle* > Agreed. While my own testing with impaired users hasn't shown > this, that suggests I need to do more testing. One thing we might > want to consider is gathering the alt-key combos from the major > browsers and making those off-limits. "Me too." > But that begs the question, what if we run out of letters? Do we > dump the accesskey attributes? And who can we blame for this, > browsers or us? I need blame to made. As the list at the bottom shows, there are still plenty of letters to play with once you get through the most-important stuff. when everything is said and done, you've got lots of keys to work with. With three keys to shift bit positions, and a long list of barely-used function keys, there are lots of possibilities. Or mahybe I'm just optimistic because I've seen kanji being typed into a computer, in person. > So, is it better to kill the accesskeys or force users to use the > mouse? I don't know that answer... If certain macros are ascribed to under unwritten rules... well, there's definitely a problem there. However, I think it'd be possible to have the cake and eat it too. What's discomfiting is finding a solid info architecture for documenting the macros employed on the site. > > ...Which bothers me, but that's another thought for another time. > > That you're the first to comment on it? Yep! Maybe I was a Luddite in a previous life. :-) > Anyway, verboten letters from the windows browsers i use regularly: > > IE5: f, e, v, a, t, h > N6: f, e, v, s, g, b, t, h > N3: f, e, v, g, b, o, d, w, h > O5: f, e, v, n, b, m, g, s, w, n > > Any others? Alt [sp] & Alt-F4, though Windows may itself put the latter off limits. Well, an investigation of Ctrl macros seems advisable as well. O, P, X, C, V, A, F, R, N, B, L... -- Ben Henick Web Author At-Large http://www.io.com/persist1/ persist1 at io.com "Are you pondering what I'm pondering, Pinky?" "I think so, Brain, but... (snort) no, no, it's too stupid." "We will disguise ourselves as a cow." "Oh!" (giggles) "That was it exactly!" From djc at starkmedia.com Fri Mar 2 09:17:05 2001 From: djc at starkmedia.com (Daniel J. Cody) Date: Fri Mar 2 09:17:05 2001 Subject: [thesite] Feb stats.. References: Message-ID: <3A9FB955.5000203@starkmedia.com> John Handelaar wrote: > 1 You did way more PIs per day in the last 7 days than > over the whole month. > 2 On the same note, your hits vs PIs ratio altered > from 5:1 to around 3:1 during that last week. we had a pretty big surge in traffic last week yes.. aardvarks 'redesign' article got a lot of links on other websites.. also, most of the people that signed up for m.e.o accounts came in last week > 3 Time-of-day graph suggests the Europeans use thesite > more than anyone else (assuming the time zone is correct > on the report). > 4 Netscape does a lot better on evolt than it does > on any sites I monitor :-) Oh, look: Macs too. > 5 1500 favicons! In one month! exactly.. thats what I like to see.. ;) > If you'd like me to have a look at the analog.cfg file > there is a way to make weo's 404s show up as PIs. I > think... Sure, find it below... LOGFILE /home/evolt/logs/old/*.log #APACHELOGFORMAT (%h %l %u %t \"%r\" %>s %b \"%{Referer}i\" \"%{User-Agent}i\) # # Your organisation's name and URL for the top line of the report HOSTNAME "[evolt.org]" HOSTURL none # # Where to write the report OUTFILE /home/evolt/archive_html/stats/stats.feb2001.html # # URL for the images directory #IMAGEDIR images/ # # Style sheet for the output page #STYLESHEET none # # /dir/ fetches /dir/index.html #DIRSUFFIX index.html # # Case sensitive or case insensitive file system: default depends on your OS. # If your web server and your analog run on different OS's, you may need to # change this according to the machine running your server. # CASE INSENSITIVE # USERCASE SENSITIVE # # How to uncompress compressed logfiles # UNCOMPRESS *.gz,*.Z "gzip -cd" # Cache files: see docs/cache.html CACHEFILE none CACHEOUTFILE noneBROWOUTPUTALIAS Mozilla Netscape BROWOUTPUTALIAS "Mozilla (compatible)" "Netscape (compatible)" BROWOUTPUTALIAS IWENG AOL TYPEOUTPUTALIAS .html ".html [Hypertext Markup Language]" TYPEOUTPUTALIAS .htm ".htm [Hypertext Markup Language]" TYPEOUTPUTALIAS .shtml ".shtml [Server-parsed HTML]" TYPEOUTPUTALIAS .ps ".ps [PostScript]" TYPEOUTPUTALIAS .gz ".gz [Gzip compressed files]" TYPEOUTPUTALIAS .tar.gz ".tar.gz [Compressed archives]" TYPEOUTPUTALIAS .jpg ".jpg [JPEG graphics]" TYPEOUTPUTALIAS .jpeg ".jpeg [JPEG graphics]" TYPEOUTPUTALIAS .gif ".gif [GIF graphics]" TYPEOUTPUTALIAS .png ".png [PNG graphics]" TYPEOUTPUTALIAS .txt ".txt [Plain text]" TYPEOUTPUTALIAS .cgi ".cgi [CGI scripts]" TYPEOUTPUTALIAS .pl ".pl [Perl scripts]" TYPEOUTPUTALIAS .css ".css [Cascading Style Sheets]" TYPEOUTPUTALIAS .class ".class [Java class files]" TYPEOUTPUTALIAS .pdf ".pdf [Adobe Portable Document Format]" TYPEOUTPUTALIAS .zip ".zip [Zip archives]" TYPEOUTPUTALIAS .hqx ".hqx [Macintosh archives]" TYPEOUTPUTALIAS .exe ".exe [Executables]" TYPEOUTPUTALIAS .wav ".wav [WAV sound files]" TYPEOUTPUTALIAS .avi ".avi [AVI movies]" TYPEOUTPUTALIAS .arc ".arc [Compressed archives]" TYPEOUTPUTALIAS .mid ".mid [MIDI sound files]" TYPEOUTPUTALIAS .mp3 ".mp3 [MP3 sound files]" TYPEOUTPUTALIAS .doc ".doc [Microsoft Word document]" TYPEOUTPUTALIAS .rtf ".rtf [Rich Text Format]" TYPEOUTPUTALIAS .mov ".mov [Quick Time movie]" TYPEOUTPUTALIAS .mpg ".mpg [MPEG movie]" TYPEOUTPUTALIAS .mpeg ".mpeg [MPEG movie]" TYPEOUTPUTALIAS .asp ".asp [Active Server Pages]" TYPEOUTPUTALIAS .jsp ".jsp [Java Server Pages]" TYPEOUTPUTALIAS .cfm ".cfm [Cold Fusion]" TYPEOUTPUTALIAS .php ".php [PHP]" TYPEOUTPUTALIAS .js ".js [JavaScript code]" 304ISSUCCESS ON # Whether each report is wanted: see docs/output.html GENERAL ON MONTHLY ON DAILY ON FULLDAILY OFF WEEKLY OFF HOURLY ON FULLHOURLY OFF QUARTER OFF FIVE OFF DOMAIN ON ORGANISATION ON DIRECTORY ON FILETYPE ON REQUEST ON REDIR OFF FAILURE OFF SIZE ON PROCTIME OFF HOST OFF REFERRER ON REFSITE ON REDIRREF ON FAILREF OFF VHOST ON USER ON FAILUSER OFF SEARCHQUERY ON SEARCHWORD ON BROWSER ON FULLBROWSER ON OSREP ON STATUS ON # The columns in each report: see docs/timereps.html and docs/othreps.html TIMECOLS RP MONTHCOLS RP DAYCOLS RP FULLDAYCOLS RP WEEKCOLS RP HOURCOLS RP FULLHOURCOLS RP QUARTERCOLS RP FIVECOLS RP DOMCOLS Rb ORGCOLS Rb HOSTCOLS Rb DIRCOLS Rb TYPECOLS Rb REQCOLS RbD REDIRCOLS R FAILCOLS R REFCOLS R REFSITECOLS R REDIRREFCOLS R FAILREFCOLS R BROWCOLS R FULLBROWCOLS R OSCOLS NR SIZECOLS Rb PROCTIMECOLS R VHOSTCOLS Rb USERCOLS Rb FAILUSERCOLS R SEARCHQUERYCOLS R SEARCHWORDCOLS R STATUSCOLS R # How the graphs are plotted: see docs/timereps.html ALLGRAPH P MONTHGRAPH P DAYGRAPH P FULLDAYGRAPH P HOURGRAPH P FULLHOURGRAPH P WEEKGRAPH P QUARTERGRAPH P FIVEGRAPH P # Whether the time reports run forwards or backwards: see docs/timereps.html ALLBACK OFF MONTHBACK OFF FULLDAYBACK OFF FULLHOURBACK OFF WEEKBACK OFF QUARTERBACK OFF FIVEBACK OFF # The number of rows to display in each time report: see docs/timereps.html MONTHROWS 0 WEEKROWS 0 FULLDAYROWS 42 FULLHOURROWS 72 QUARTERROWS 96 FIVEROWS 288 SEARCHENGINE http://*altavista.*/* q SEARCHENGINE http://*yahoo.*/* p SEARCHENGINE http://*google.*/* q SEARCHENGINE http://*lycos.*/* query SEARCHENGINE http://*aol.*/* query SEARCHENGINE http://*excite.*/* search SEARCHENGINE http://*go2net.*/* general SEARCHENGINE http://*metacrawler.*/* general SEARCHENGINE http://*msn.*/* MT SEARCHENGINE http://*hotbot.com/* MT SEARCHENGINE http://*netscape.*/* search SEARCHENGINE http://*looksmart.*/* key SEARCHENGINE http://*infoseek.*/* qt SEARCHENGINE http://*webcrawler.*/* search,searchText SEARCHENGINE http://*goto.*/* Keywords SEARCHENGINE http://*snap.*/* keyword SEARCHENGINE http://*dogpile.*/* q SEARCHENGINE http://*askjeeves.*/* ask SEARCHENGINE http://*ask.*/* ask SEARCHENGINE http://*aj.*/* ask SEARCHENGINE http://*directhit.*/* qry SEARCHENGINE http://*alltheweb.*/* query SEARCHENGINE http://*northernlight.*/* qr SEARCHENGINE http://*nlsearch.*/* qr SEARCHENGINE http://*dmoz.*/* search SEARCHENGINE http://*newhoo.*/* search SEARCHENGINE http://*netfind.*/* query,search,s SEARCHENGINE http://*/netfind* query SEARCHENGINE http://*/pursuit query From djc at starkmedia.com Fri Mar 2 16:09:50 2001 From: djc at starkmedia.com (Daniel J. Cody) Date: Fri Mar 2 16:09:50 2001 Subject: [thesite] need another template.. References: Message-ID: <3AA01A06.3040302@starkmedia.com> ok, here's what I got.. for testting purposes, we're using SQL server here, not that it matters.. the datasource name in CF is 'thelist' shouldnt hdr_inreplyto be a number? anyways, whenever you get around to it, hve fun and lemm eknow if you have problems ;) thanks again seth CREATE TABLE [thelist] ( [listid] [int] NULL , [messageid] [int] NOT NULL , [hdr_messageid] [int] NOT NULL , [hdr_fromname] [varchar] (75) NOT NULL , [hdr_fromemail] [varchar] (100) NOT NULL , [hdr_toemail] [varchar] (100) NULL , [hdr_subject] [varchar] (100) NOT NULL , [hdr_date] [datetime] NOT NULL , [hdr_inreplyto] [varchar] (75) NULL , [bdy_content] [text] NOT NULL Seth Bienek wrote: > listid (fk to lookup table with list names) > messageid (pk) > hdr_message-id (text-75char) > hdr_fromemail (text-100char) > hdr_fromname (text-100char) > hdr_toemail (text-100char - optional may be used instead of listid?) > hdr_subject (text-100char) > hdr_date (date) > hdr_in-reply-to (text-75char) > bdy_content (long, memo, whatever) > From djc at starkmedia.com Sat Mar 3 14:08:25 2001 From: djc at starkmedia.com (Daniel J. Cody) Date: Sat Mar 3 14:08:25 2001 Subject: [thesite] didnt see this come through.. Message-ID: <3AA14D36.9090701@starkmedia.com> ok, here's what I got.. for testting purposes, we're using SQL server here, not that it matters.. the datasource name in CF is 'thelist' shouldnt hdr_inreplyto be a number? anyways, whenever you get around to it, hve fun and lemm eknow if you have problems ;) thanks again seth CREATE TABLE [thelist] ( [listid] [int] NULL , [messageid] [int] NOT NULL , [hdr_messageid] [int] NOT NULL , [hdr_fromname] [varchar] (75) NOT NULL , [hdr_fromemail] [varchar] (100) NOT NULL , [hdr_toemail] [varchar] (100) NULL , [hdr_subject] [varchar] (100) NOT NULL , [hdr_date] [datetime] NOT NULL , [hdr_inreplyto] [varchar] (75) NULL , [bdy_content] [text] NOT NULL Seth Bienek wrote: > listid (fk to lookup table with list names) > messageid (pk) > hdr_message-id (text-75char) > hdr_fromemail (text-100char) > hdr_fromname (text-100char) > hdr_toemail (text-100char - optional may be used instead of listid?) > hdr_subject (text-100char) > hdr_date (date) > hdr_in-reply-to (text-75char) > bdy_content (long, memo, whatever) > From seth at sethbienek.com Mon Mar 5 08:56:03 2001 From: seth at sethbienek.com (Seth Bienek) Date: Mon Mar 5 08:56:03 2001 Subject: [thesite] need another template.. In-Reply-To: <3AA01A06.3040302@starkmedia.com> Message-ID: Hey Dan, > anyways, whenever you get around to it, hve fun Mucho Gusto! > shouldn't hdr_inreplyto be a number? Nope, it's the hdr_messageID that this message is in reply to, which is a mixed string set by the outgoing mail server (as I'm sure you know. ;). I could cross-reference that with the messageID of the original message, but it would cause a problems if a reply were to be received before an original (is that possible)? And it would add a lookup query to the insert process. Anyhoo.. I'll probably have some time to play with it this afternoon, I'll report back my findings then. :) Seth ------------------------------ Seth Bienek Solutions Development Manager Stonebridge Technologies, Inc. 972.455.7294 tel 972.404.9754 fax ICQ #7673959 ------------------------------ > -----Original Message----- > Subject: Re: [thesite] need another template.. > > CREATE TABLE [thelist] ( > [listid] [int] NULL , > [messageid] [int] NOT NULL , > [hdr_messageid] [int] NOT NULL , > [hdr_fromname] [varchar] (75) NOT NULL , > [hdr_fromemail] [varchar] (100) NOT NULL , > [hdr_toemail] [varchar] (100) NULL , > [hdr_subject] [varchar] (100) NOT NULL , > [hdr_date] [datetime] NOT NULL , > [hdr_inreplyto] [varchar] (75) NULL , > [bdy_content] [text] NOT NULL From r937 at interlog.com Mon Mar 5 10:23:48 2001 From: r937 at interlog.com (rudy) Date: Mon Mar 5 10:23:48 2001 Subject: [thesite] query results caching Message-ID: <01c0a590$818c40a0$4451149a@rudy> [ in reply to a post on thelist ] > In ColdFusion you can do this by stuffing query results > into an structured application variable. hi seth excuse me, i hope i don't come across as whining, but nobody answered the last two times i asked this question... why would you not use query results caching? it's a cold fusion feature, so why would you go to all the trouble of saving stuff in structures? the query results are a structure already, right? disclaimer: i haven't had the chance to use query results caching myself rudy From seth at sethbienek.com Mon Mar 5 10:44:14 2001 From: seth at sethbienek.com (Seth Bienek) Date: Mon Mar 5 10:44:14 2001 Subject: [thesite] query results caching In-Reply-To: <01c0a590$818c40a0$4451149a@rudy> Message-ID: Hi Rudy, > excuse me, i hope i don't come across as whining, but nobody answered the > last two times i asked this question... Thelist traffic has grown so much that I don't get to read every message like I used to. Sorry `bout that.. The surefire way to get *my* attention is to put CF in the subject line somewhere. :) > why would you not use query results caching? Query results caching is good for queries that do not change, as CF will store a seperate copy of the query results for each *unique* query - not much use if your queries are dynamic, and/or you have alot of them. Also there is a limit of 100 cached queries, though this may be tweakable. Lastly, CF saves these stored query results as text files on the hard drive instead of in the server's memory. Not alot of performance overhead there, but certainly more than getting the data out of the server's RAM. For quick-and-dirty, or simple queries, query caching is probably the way to go. But if you are looking for a way to speed up access to large amounts of data, or optimize a heavy-hitting dynamic query, the structure method's good. BTW, it is kind of a pain in the butt to set up; I think that's something you have to weigh in the decision to implement it. I just re-read that and I think I may have overcomplicated the explaination. Lemme know if it's as clear as mud and I'll take another run at it.. :) Seth ------------------------------ Seth Bienek Solutions Development Manager Stonebridge Technologies, Inc. 972.455.7294 tel 972.404.9754 fax ICQ #7673959 ------------------------------ From seth at sethbienek.com Mon Mar 5 10:57:13 2001 From: seth at sethbienek.com (Seth Bienek) Date: Mon Mar 5 10:57:13 2001 Subject: [thesite] query results caching In-Reply-To: Message-ID: Hey Rudy, I wanted to verify what I told you, and as it turns out I missed a couple of points: *Cached queries are in fact stored in server memory, not on disk. I was confusing it with cfcache, which does use disk storage. *The limit to the number of cached queries can be set in CFAdministrator. Other than those two pretty significant differences, I was on point. Regards, Seth ------------------------------ Seth Bienek Solutions Development Manager Stonebridge Technologies, Inc. 972.455.7294 tel 972.404.9754 fax ICQ #7673959 ------------------------------ From bobd at members.evolt.org Mon Mar 5 11:05:08 2001 From: bobd at members.evolt.org (Bob Davis) Date: Mon Mar 5 11:05:08 2001 Subject: [thesite] query results caching In-Reply-To: Message-ID: on 3/5/01 11:43 AM, Seth Bienek at seth at sethbienek.com wrote: > For quick-and-dirty, or simple queries, query caching is probably the way to > go. But if you are looking for a way to speed up access to large amounts of > data, or optimize a heavy-hitting dynamic query, the structure method's > good. BTW, it is kind of a pain in the butt to set up; I think that's > something you have to weigh in the decision to implement it. > > I just re-read that and I think I may have overcomplicated the explaination. > Lemme know if it's as clear as mud and I'll take another run at it.. :) Content on the site changes irregularly. I had a similar situation with an intranet that I was maintaining. There were something like 6 queries on every page, most of which didn't change very often (and they were all static). When they did change, it was because one of the site administrators had logged in and done something. What I did was to create a cacheing time for the queries that was dynamic. If there was no administrator logged in, the time was 12 hours. If there was an administrator logged in, the cache time was 0. That made sure that any content that was added was seen immediately, and not x hours after the update was made. We could do something similar on evolt.org. The form action pages for article submissions, comments, etc. could kick out cache times of 0 for the queries they change - but only the queries they change. For instance, when an admin approves an article, the queries that grab the headlines, the pending approval list, etc. All of the stuff in the sidebars that doesn't change every time someone hits the site could be cached like this. What do you all think? bob -- bob davis bobd at members.evolt.org From r937 at interlog.com Mon Mar 5 11:05:13 2001 From: r937 at interlog.com (rudy) Date: Mon Mar 5 11:05:13 2001 Subject: [thesite] query results caching Message-ID: <01c0a596$3c58bda0$4451149a@rudy> thanks seth yeah, i knew about the uniqueness requirement for cached queries -- but it seems silly to want to cache a "dynamic" query anyway by "dynamic" i mean a query tailored to whoever's running it, e.g. select stuff, [number of unseen replies by this user] from content, etcetera where userid = [signed on user] everybody's going to get a different result, so why cache, right? on the other hand, getting the list of our categories for the dropdown list, that only changes when we add a new category.... ... but i wasn't aware that "CF saves these stored query results as text files on the hard drive instead of in the server's memory" -- which truly sucks, although i bet it'd still be faster than exiting from cf, going to the database server, and running the query again for everybody who visits the page i'm sure there must be "dynamic" queries that need to be saved, i just can't think of any that would need to be pulled from later... and saving them into a structure sounds okay, but would that be a client or session structure? just trying to catch up to you cf heavy-hitters... rudy From jeff at members.evolt.org Mon Mar 5 11:52:59 2001 From: jeff at members.evolt.org (jeff) Date: Mon Mar 5 11:52:59 2001 Subject: [thesite] query results caching In-Reply-To: <01c0a590$818c40a0$4451149a@rudy> Message-ID: rudy, :~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~ : From: rudy : : > In ColdFusion you can do this by stuffing : > query results into an structured application : > variable. : : why would you not use query results caching? :~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~ sometimes it's more appropriate to store the results of a query in a much long-term mechanism. query caching can do this, but it's not nearly as controllable as say an application scoped variable. rather than let the cold fusion server handle caching of the query (and possibly allowing the query to run again when i don't want it to), i can simply store the results of the query in an application scoped variable and know that query will *not* run again until the server is rebooted or i explicitly tell it to run again by resetting that application variable. there's also the likelihood that you may not have the choice of using query results caching because it's been disabled. :~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~ : it's a cold fusion feature, so why would you : go to all the trouble of saving stuff in : structures? :~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~ it's not really that much trouble. take these two as an example. the first one is using query results caching. the second is storing the query in an application scoped variable. example #1 SELECT id FROM movies WHERE dialogue LIKE '%fubar%' example #2 SELECT id FROM movies WHERE dialogue LIKE '%fubar%' you'll notice that there are a few more lines to the second example. however, for the most part these lines (, with a minor change to the name the query will be stored as, and with a minor change to the name the query will be searched for by) can be copy-n-pasted for future uses. :~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~ : the query results are a structure already, : right? :~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~ well, yes sort of. they're a structure the same way a normal query is. whether they're cached or scoped to an application or session variable is irrelevant -- the results are accessed in the same fashion. does this make more sense? .jeff name://jeff.howden game://web.development http://www.evolt.org/ mailto:jeff at members.evolt.org From r937 at interlog.com Mon Mar 5 12:01:09 2001 From: r937 at interlog.com (rudy) Date: Mon Mar 5 12:01:09 2001 Subject: [thesite] query results caching Message-ID: <01c0a59e$34c1b620$4451149a@rudy> >does this make more sense? ayup and the examples help say, you're not (also) writing a cf book, are you? rudemeister From joshua at alphashop.net Mon Mar 5 12:05:09 2001 From: joshua at alphashop.net (Joshua OIson) Date: Mon Mar 5 12:05:09 2001 Subject: [thesite] query results caching References: Message-ID: <013601c0a5b9$0a66b590$38360141@mrtnz1.ga.home.com> Additionally, For Cold Fusion to use cached queries, the SQL statement, db, and everything must be exactly identical. IOW, CF Server would have to pattern match the entire query against cached queries on record. That does, while however minimal, take time. When compiling a template to P-Code, the pattern match would still have to take place against the database of cached queries. However, if you explicitly store the query, you may lose some time in storing the query, but in P-Code the recall of the query runs in a flat time, not dependent on the number of queries cached and the length of the SQL query string. Additionally, if you hand craft a query by using and hand stuffing the rows, then you can use the same mechanisms you use to store the query. So, now you have one repository for saved queries instead of two. -joshua From seth at sethbienek.com Mon Mar 5 14:13:32 2001 From: seth at sethbienek.com (Seth Bienek) Date: Mon Mar 5 14:13:32 2001 Subject: [thesite] didnt see this come through.. In-Reply-To: <3AA14D36.9090701@starkmedia.com> Message-ID: Hey Dan, > [hdr_messageid] [int] NOT NULL Should be text instead of int.. It'll hold data like this: 3AA14D36.9090701 at starkmedia.com Same thing with: > [hdr_inreplyto] [varchar] (75) NULL Seth ------------------------------ Seth Bienek Solutions Development Manager Stonebridge Technologies, Inc. 972.455.7294 tel 972.404.9754 fax ICQ #7673959 ------------------------------ From djc at starkmedia.com Mon Mar 5 14:34:22 2001 From: djc at starkmedia.com (Daniel J. Cody) Date: Mon Mar 5 14:34:22 2001 Subject: [thesite] didnt see this come through.. References: Message-ID: <3AA3F8B4.70804@starkmedia.com> ok, good to go :) Seth Bienek wrote: > Hey Dan, > > >> [hdr_messageid] [int] NOT NULL > > > Should be text instead of int.. It'll hold data like this: > > 3AA14D36.9090701 at starkmedia.com > > Same thing with: > >> [hdr_inreplyto] [varchar] (75) NULL > > > > Seth > > ------------------------------ > Seth Bienek > Solutions Development Manager > Stonebridge Technologies, Inc. > 972.455.7294 tel > 972.404.9754 fax > ICQ #7673959 > ------------------------------ > > > _______________________________________________ > http://lists.evolt.org/thesitearchive/ > and new & improved kentucky fried old archives: > http://lists.evolt.org/thesitearchive/old/ From djc at starkmedia.com Mon Mar 5 17:08:38 2001 From: djc at starkmedia.com (Daniel J. Cody) Date: Mon Mar 5 17:08:38 2001 Subject: [thesite] test.evolt.org Message-ID: <3AA41CDD.9010402@starkmedia.com> is in transition to here. the site is over and the DB came over fine, but I need to install oracle client software on the webserver so CF can see it.. anyways, just in case anyone sees an error or something looks like its not working.. Thanks.. .djc. From joxn at vernum.com Wed Mar 7 13:43:23 2001 From: joxn at vernum.com (Joxn) Date: Wed Mar 7 13:43:23 2001 Subject: [thesite] integrating third party scripts Message-ID: <3AA68FA8.23E3ACF5@vernum.com> Hi everybody, I've got the UBB messageboard and simply embeded it in my site by using frames. Well, I'd like to ditch the old frame-based navigation in favour of including it. (The site is PHP driven.) What's the easiest way to integrate UBB now? The problem that I see is that every message would need the site navigation included. Is there a way that I can do this by SSI eg. say "every file that is called from my ubb directory get this file included"? Any other ideas? How do you usually integrate third party scripts? TIA, Joxn -- || //\\ \\// |\\|| :: joxn at vernum.com :: \\|| \\// //\\ ||\\| :: 8053703 :: From joxn at vernum.com Wed Mar 7 13:44:08 2001 From: joxn at vernum.com (Joxn) Date: Wed Mar 7 13:44:08 2001 Subject: [thesite] integrating third party scripts References: <3AA68FA8.23E3ACF5@vernum.com> Message-ID: <3AA68FD8.98DDE737@vernum.com> Damn, wrong list. I'm sorry. bye Joxn -- || //\\ \\// |\\|| :: joxn at vernum.com :: \\|| \\// //\\ ||\\| :: 8053703 :: From djc at members.evolt.org Wed Mar 7 13:55:01 2001 From: djc at members.evolt.org (Daniel J. Cody) Date: Wed Mar 7 13:55:01 2001 Subject: [thesite] integrating third party scripts References: <3AA68FA8.23E3ACF5@vernum.com> <3AA68FD8.98DDE737@vernum.com> Message-ID: <3AA69279.F56B071A@members.evolt.org> dont worry, happens to me all the time :) .djc. Joxn wrote: > > Damn, wrong list. I'm sorry. > bye Joxn > -- > || //\\ \\// |\\|| :: joxn at vernum.com :: > \\|| \\// //\\ ||\\| :: 8053703 :: > > _______________________________________________ > http://lists.evolt.org/thesitearchive/ > and new & improved kentucky fried old archives: > http://lists.evolt.org/thesitearchive/old/ From martin at members.evolt.org Wed Mar 7 14:13:57 2001 From: martin at members.evolt.org (Martin) Date: Wed Mar 7 14:13:57 2001 Subject: [thesite] thelist archives Message-ID: I was trying to check out an old post as the basis for an article, and got this error: Error in Program Execution Error Information Date: Wed Mar 7 14:10:45 2001 The file that has the error: /home/evolt/leo_html/thelist/readMessage.cfm The query string: The IP of the user with the error: 138.88.88.104 The user's browser: Mozilla/4.0 (compatible; MSIE 5.0; Mac_PowerPC) The referring Script: http://www.evolt.org/article/view/22/3889/index.html The information on the error: ODBC Error Code = S1000 (General error) [MERANT][ODBC MySql driver][MySql]Access denied for user: '@localhost' to database 'evolt' SQL = "select subject, id, author, body, popdate from posts where id=48000" Data Source = "EVOLT" The error occurred while processing an element with a general identifier of (CFQUERY), occupying document position (1:1) to (1:65). Send this error report to dcody at oracular.com _______________________________ I was following an old bookmark, but I'm fairly sure that it *used* to work. Oh, Dan, better change that email address too ;-) Cheers Martin From djc at starkmedia.com Wed Mar 7 14:17:34 2001 From: djc at starkmedia.com (Daniel J. Cody) Date: Wed Mar 7 14:17:34 2001 Subject: [thesite] thelist archives References: Message-ID: <3AA697BF.B2E31D5F@starkmedia.com> ya, dont use those anymore.. you'll have to use the plain jane archives for now :( Martin wrote: > > I was trying to check out an old post as the basis for > an article, and got this error: > > Error in Program Execution > > Error Information > Date: Wed Mar 7 14:10:45 2001 > The file that has the error: /home/evolt/leo_html/thelist/readMessage.cfm > The query string: > The IP of the user with the error: 138.88.88.104 > The user's browser: Mozilla/4.0 (compatible; MSIE 5.0; Mac_PowerPC) > The referring Script: http://www.evolt.org/article/view/22/3889/index.html > The information on the error: > ODBC Error Code = S1000 (General error) > > [MERANT][ODBC MySql driver][MySql]Access denied for user: > '@localhost' to database 'evolt' > > SQL = "select subject, id, author, body, popdate from posts where id=48000" > > Data Source = "EVOLT" > > The error occurred while processing an element with a general > identifier of (CFQUERY), occupying document position (1:1) to (1:65). > > Send this error report to dcody at oracular.com > > _______________________________ > > I was following an old bookmark, but I'm fairly sure > that it *used* to work. > > Oh, Dan, better change that email address too ;-) > > Cheers > Martin > > _______________________________________________ > http://lists.evolt.org/thesitearchive/ > and new & improved kentucky fried old archives: > http://lists.evolt.org/thesitearchive/old/ From martin at members.evolt.org Wed Mar 7 14:23:41 2001 From: martin at members.evolt.org (Martin) Date: Wed Mar 7 14:23:41 2001 Subject: [thesite] thelist archives In-Reply-To: <3AA697BF.B2E31D5F@starkmedia.com> References: <3AA697BF.B2E31D5F@starkmedia.com> Message-ID: Damn. Means I have to remember when I wrote that rant :-( Cheers Martin >ya, dont use those anymore.. you'll have to use the plain jane archives >for now :( > >Martin wrote: >> >> I was trying to check out an old post as the basis for >> an article, and got this error: >> > > Error in Program Execution From john at userfrenzy.com Mon Mar 12 10:31:13 2001 From: john at userfrenzy.com (John Handelaar) Date: Mon Mar 12 10:31:13 2001 Subject: [thesite] RDF/RSS again In-Reply-To: Message-ID: Hope you're all having a nice time in Austin... To revisit something I asked a while ago - There must be more than just one or two of us who'd like to syndicate thesite headlines using a Slash (or Slashalike) engine, or even just put it on the left hand side of Mozilla. I said RDF before - more specifically, what I'm after is RSS/My Netscape format. Is it already there, or is it a new feature? Free code from Morbus: http://evolt.org/article/Headlines_in_RSS_and_Perl/17/5535/index.html ------------------------------------------ John Handelaar T +44 20 7209 4117 M +44 7930 681789 F +44 870 169 7657 E john at userfrenzy.com ------------------------------------------ From djc at starkmedia.com Tue Mar 13 17:15:46 2001 From: djc at starkmedia.com (Daniel J. Cody) Date: Tue Mar 13 17:15:46 2001 Subject: [thesite] RDF/RSS again References: Message-ID: <3AAEAA63.C62AC770@starkmedia.com> Hey John - Had something like this for a long time, but it stopped working cus the new url scheme, and no one noticed(including me).. http://evolt.org/evolt/xml/rdf.cfm should be all cleaned up and reflect out new url scheme.. what do you wanna do with it? :) .djc. John Handelaar wrote: > > Hope you're all having a nice time in Austin... > > To revisit something I asked a while ago - > > There must be more than just one or two of us > who'd like to syndicate thesite headlines using > a Slash (or Slashalike) engine, or even just > put it on the left hand side of Mozilla. > > I said RDF before - more specifically, what > I'm after is RSS/My Netscape format. Is it > already there, or is it a new feature? > > Free code from Morbus: > > http://evolt.org/article/Headlines_in_RSS_and_Perl/17/5535/index.html From jeff at members.evolt.org Tue Mar 13 17:48:26 2001 From: jeff at members.evolt.org (jeff) Date: Tue Mar 13 17:48:26 2001 Subject: [thesite] RDF/RSS again In-Reply-To: Message-ID: john, :~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~ : From: John Handelaar : : There must be more than just one or : two of us who'd like to syndicate : thesite headlines using a Slash (or : Slashalike) engine, or even just put : it on the left hand side of Mozilla. : : I said RDF before - more specifically, : what I'm after is RSS/My Netscape format. : Is it already there, or is it a new feature? :~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~ i already put together something like this in response to your query back in late february. http://www.evolt.org/xml/articles/index.html is that not what you're looking for? if it isn't the right format, let me know and i can make a version that's the correct format. thanks, .jeff name://jeff.howden game://web.development http://www.evolt.org/ mailto:jeff at members.evolt.org From djc at starkmedia.com Tue Mar 13 17:58:32 2001 From: djc at starkmedia.com (Daniel J. Cody) Date: Tue Mar 13 17:58:32 2001 Subject: [thesite] RDF/RSS again References: Message-ID: <3AAEB469.647A7AD3@starkmedia.com> oh.. i dont know what the difference between rdf and rss is, but it looks like we now have one of both(although yours willprobably work better jeff).. looks like we need dave winer to sort this out for us ;) jeff wrote: > > john, > > :~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~ > : From: John Handelaar > : > : There must be more than just one or > : two of us who'd like to syndicate > : thesite headlines using a Slash (or > : Slashalike) engine, or even just put > : it on the left hand side of Mozilla. > : > : I said RDF before - more specifically, > : what I'm after is RSS/My Netscape format. > : Is it already there, or is it a new feature? > :~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~ > > i already put together something like this in response to your query back in > late february. > > http://www.evolt.org/xml/articles/index.html > > is that not what you're looking for? if it isn't the right format, let me > know and i can make a version that's the correct format. > > thanks, > > .jeff > > name://jeff.howden > game://web.development > http://www.evolt.org/ > mailto:jeff at members.evolt.org > > _______________________________________________ > http://lists.evolt.org/thesitearchive/ > and new & improved kentucky fried old archives: > http://lists.evolt.org/thesitearchive/old/ From jeff at members.evolt.org Tue Mar 13 18:56:33 2001 From: jeff at members.evolt.org (jeff) Date: Tue Mar 13 18:56:33 2001 Subject: [thesite] RDF/RSS again In-Reply-To: <3AAEB469.647A7AD3@starkmedia.com> Message-ID: djc, :~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~ : From: Daniel J. Cody : : oh.. i dont know what the difference : between rdf and rss is, but it looks : like we now have one of both (although : yours will probably work better jeff).. :~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~ i don't know what the difference is either really. from a precursory reading of the article on thesite about it by morbus it looks like it's just some basic structural markup differences. i couldn't tell what the output of yours looked like though. it wouldn't bring up anything when i tried it. odd really. i envision reporting all kinds of information via xml in the near future. the trick will be figuring out what makes sense to syndicate and what formats. :~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~ : looks like we need dave winer to sort : this out for us ;) :~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~ oooooooooh, celebrity death match. muahahahahahaha thanks, .jeff name://jeff.howden game://web.development http://www.evolt.org/ mailto:jeff at members.evolt.org From peter at vardus.com Wed Mar 14 02:36:36 2001 From: peter at vardus.com (Peter Van Dijck) Date: Wed Mar 14 02:36:36 2001 Subject: [thesite] submit articles back button In-Reply-To: References: Message-ID: <4.3.2.20010314083410.02162d00@mail.vardus.net> Hi, I just submitted an article and the back button (after you've clicked preview) isn't working, so that means there was an error in the article I didn't correct. The article is "lost in the matrix" The error is at the bottom: a link to "my experimental manial site" should be "my experimental manila site". :) (or maybe it shouldn't? oh well...) btw: thanks for all the great work on evolt to everyone who's worked on it! P ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~ http://liga1.com: building multiple language/culture websites From john at userfrenzy.com Wed Mar 14 06:37:47 2001 From: john at userfrenzy.com (John Handelaar) Date: Wed Mar 14 06:37:47 2001 Subject: [thesite] RDF/RSS again In-Reply-To: <3AAEAA63.C62AC770@starkmedia.com> Message-ID: > -----Original Message----- > From: thesite-admin at lists.evolt.org > [mailto:thesite-admin at lists.evolt.org]On Behalf Of Daniel J. Cody > Sent: 13 March 2001 23:17 > To: thesite at lists.evolt.org > Subject: Re: [thesite] RDF/RSS again > > > Hey John - > > Had something like this for a long time, but it stopped working cus the > new url scheme, and no one noticed(including me).. > > http://evolt.org/evolt/xml/rdf.cfm Doesn't seem to be there... The reasons I'm suggesting an RSS feed in particular are: 1 It works with NS6 sidebars 2 It works with Weblog apps including Slash, Geeklog, PHPwebsite, PHPnuke (ugh), and (cough) Userland Jeff's file certainly is XML (with ?CF syntax throughout) but it won't do either of these very-useful-for-promoting-evolt things. Well, not without an external server and a DTD, anyway. Another link: http://dmoz.org/Reference/Libraries/Library_and_Information_Science/Technica l_Services/Cataloguing/Metadata/Resource_Description_Framework_-_RDF/Applica tions/RSS/ jh (Suddenly beginning to feel like a PITA) ------------------------------------------ John Handelaar T +44 20 7209 4117 M +44 7930 681789 F +44 870 169 7657 E john at userfrenzy.com ------------------------------------------ From djc at starkmedia.com Wed Mar 14 06:57:57 2001 From: djc at starkmedia.com (Daniel J. Cody) Date: Wed Mar 14 06:57:57 2001 Subject: [thesite] RDF/RSS again References: Message-ID: <3AAF6B17.90BEE89F@starkmedia.com> John Handelaar wrote: > The reasons I'm suggesting an RSS feed in particular are: > > 1 It works with NS6 sidebars > 2 It works with Weblog apps including Slash, Geeklog, > PHPwebsite, PHPnuke (ugh), and (cough) Userland cool.. those sound like good reasons 100% > Jeff's file certainly is XML (with ?CF syntax throughout) > but it won't do either of these very-useful-for-promoting-evolt > things. Well, not without an external server and a DTD, > anyway. ok even though http://evolt.org/evolt/xml/rdf.rdf says, rdf.rdf isn't that an RSS file? from what i read, that file should validate as an rss file, correct? Maybe if you could show an example of what we need, i or jeff can code it > (Suddenly beginning to feel like a PITA) seriously, dont worry about it.. its helpful when people actually tell us what they want rather than just expecting us to do it :) .djc. From john at userfrenzy.com Wed Mar 14 07:08:03 2001 From: john at userfrenzy.com (John Handelaar) Date: Wed Mar 14 07:08:03 2001 Subject: [thesite] RDF/RSS again In-Reply-To: <3AAF6B17.90BEE89F@starkmedia.com> Message-ID: > -----Original Message----- > From: thesite-admin at lists.evolt.org > [mailto:thesite-admin at lists.evolt.org]On Behalf Of Daniel J. Cody > Sent: 14 March 2001 12:59 > To: thesite at lists.evolt.org > Subject: Re: [thesite] RDF/RSS again > > ok even though http://evolt.org/evolt/xml/rdf.rdf says, rdf.rdf isn't > that an RSS file? Can't tell - it doesn't seem to be there. ------------------------------------------ John Handelaar T +44 20 7209 4117 M +44 7930 681789 F +44 870 169 7657 E john at userfrenzy.com ------------------------------------------ From djc at starkmedia.com Wed Mar 14 07:09:53 2001 From: djc at starkmedia.com (Daniel J. Cody) Date: Wed Mar 14 07:09:53 2001 Subject: [thesite] RDF/RSS again References: Message-ID: <3AAF6DE3.26D4923F@starkmedia.com> cool, you're up :) did you try 'view source'? John Handelaar wrote: > > > -----Original Message----- > > From: thesite-admin at lists.evolt.org > > [mailto:thesite-admin at lists.evolt.org]On Behalf Of Daniel J. Cody > > Sent: 14 March 2001 12:59 > > To: thesite at lists.evolt.org > > Subject: Re: [thesite] RDF/RSS again > > > > ok even though http://evolt.org/evolt/xml/rdf.rdf says, rdf.rdf isn't > > that an RSS file? > > Can't tell - it doesn't seem to be there. From john at userfrenzy.com Wed Mar 14 07:14:18 2001 From: john at userfrenzy.com (John Handelaar) Date: Wed Mar 14 07:14:18 2001 Subject: [thesite] RDF/RSS again In-Reply-To: <3AAF6DE3.26D4923F@starkmedia.com> Message-ID: > -----Original Message----- > From: thesite-admin at lists.evolt.org > [mailto:thesite-admin at lists.evolt.org]On Behalf Of Daniel J. Cody > Sent: 14 March 2001 13:11 > To: thesite at lists.evolt.org > Subject: Re: [thesite] RDF/RSS again > > > cool, you're up :) It's 1.15 in the afternoon in London :-) > did you try 'view source'? Here you go... ------------------------------------------ John Handelaar T +44 20 7209 4117 M +44 7930 681789 F +44 870 169 7657 E john at userfrenzy.com ------------------------------------------ From djc at starkmedia.com Wed Mar 14 07:19:23 2001 From: djc at starkmedia.com (Daniel J. Cody) Date: Wed Mar 14 07:19:23 2001 Subject: [thesite] RDF/RSS again References: Message-ID: <3AAF701B.C22FD83@starkmedia.com> hehe.. allrighty then.. well me up before 8, much less being at work at 6:15 is a minor miracle.. trying to get my evolt stuff done in mornings now too.. http://evolt.org/evolt/xml/rss.txt John Handelaar wrote: > > > -----Original Message----- > > From: thesite-admin at lists.evolt.org > > [mailto:thesite-admin at lists.evolt.org]On Behalf Of Daniel J. Cody > > Sent: 14 March 2001 13:11 > > To: thesite at lists.evolt.org > > Subject: Re: [thesite] RDF/RSS again > > > > > > cool, you're up :) > > It's 1.15 in the afternoon in London :-) From john at userfrenzy.com Wed Mar 14 07:20:03 2001 From: john at userfrenzy.com (John Handelaar) Date: Wed Mar 14 07:20:03 2001 Subject: [thesite] RDF/RSS again In-Reply-To: <3AAF6DE3.26D4923F@starkmedia.com> Message-ID: > -----Original Message----- > From: thesite-admin at lists.evolt.org > [mailto:thesite-admin at lists.evolt.org]On Behalf Of Daniel J. Cody > Sent: 14 March 2001 13:11 > To: thesite at lists.evolt.org > Subject: Re: [thesite] RDF/RSS again > > > cool, you're up :) > > did you try 'view source'? Ah - it's a browser thing. M'colleague just WGET-ed it from a Linux box and it retrieved. MIME type problem? Neither of us can get the file to load into a browser at all. ------------------------------------------ John Handelaar T +44 20 7209 4117 M +44 7930 681789 F +44 870 169 7657 E john at userfrenzy.com ------------------------------------------ From djc at starkmedia.com Wed Mar 14 07:23:23 2001 From: djc at starkmedia.com (Daniel J. Cody) Date: Wed Mar 14 07:23:23 2001 Subject: [thesite] rudy! need some answers :) Message-ID: <3AAF710E.5B7C675A@starkmedia.com> ok, so i got a fire under my ass again and am working on some of our features.. rudy, can you explain how the members, membattr, and users tables are interconnected? why didnt we just add another column to the users table that labeled them as a member, user, god, or admin instead of creating new tables? Just needing a memory refresh so i can get working on these member pages thanks :) .djc. From john at userfrenzy.com Wed Mar 14 07:34:34 2001 From: john at userfrenzy.com (John Handelaar) Date: Wed Mar 14 07:34:34 2001 Subject: [thesite] RDF/RSS again In-Reply-To: <3AAF701B.C22FD83@starkmedia.com> Message-ID: > -----Original Message----- > From: thesite-admin at lists.evolt.org > [mailto:thesite-admin at lists.evolt.org]On Behalf Of Daniel J. Cody > Sent: 14 March 2001 13:20 > To: thesite at lists.evolt.org > Subject: Re: [thesite] RDF/RSS again > > > hehe.. allrighty then.. well me up before 8, much less being at work at > 6:15 is a minor miracle.. trying to get my evolt stuff done in mornings > now too.. > > http://evolt.org/evolt/xml/rss.txt OK - here's what it is. 1 The header is 'wrong', by which I mean that the output at http://slashdot.org/slashdot.rdf a) shows a different one and b) actually displays as a collapsible tree of XML data in IE5+, which ours doesn't. Although this might or might not be as relevant as... 2 The lines should have only the link URL in plaintext, not an entire tag 3 Then all we need is the new URL format :-) ------------------------------------------ John Handelaar T +44 20 7209 4117 M +44 7930 681789 F +44 870 169 7657 E john at userfrenzy.com ------------------------------------------ From djc at starkmedia.com Wed Mar 14 07:46:54 2001 From: djc at starkmedia.com (Daniel J. Cody) Date: Wed Mar 14 07:46:54 2001 Subject: [thesite] RDF/RSS again References: Message-ID: <3AAF768E.7FFE85CE@starkmedia.com> alrighty, check this out: http://evolt.org/evolt/xml/rdf.cfm that validated through the my.netscape shit, so i would assume its good. does it need to have a .rss extension for others to use it? John Handelaar wrote: > > > -----Original Message----- > > From: thesite-admin at lists.evolt.org > > [mailto:thesite-admin at lists.evolt.org]On Behalf Of Daniel J. Cody > > Sent: 14 March 2001 13:20 > > To: thesite at lists.evolt.org > > Subject: Re: [thesite] RDF/RSS again > > > > > > hehe.. allrighty then.. well me up before 8, much less being at work at > > 6:15 is a minor miracle.. trying to get my evolt stuff done in mornings > > now too.. > > > > http://evolt.org/evolt/xml/rss.txt > > OK - here's what it is. > > 1 The header is 'wrong', by which I mean that the output > at http://slashdot.org/slashdot.rdf a) shows a different > one and b) actually displays as a collapsible tree of XML > data in IE5+, which ours doesn't. Although this might > or might not be as relevant as... > > 2 The lines should have only the link URL in plaintext, > not an entire tag > > 3 Then all we need is the new URL format :-) From john at userfrenzy.com Wed Mar 14 08:01:34 2001 From: john at userfrenzy.com (John Handelaar) Date: Wed Mar 14 08:01:34 2001 Subject: [thesite] RDF/RSS again In-Reply-To: <3AAF768E.7FFE85CE@starkmedia.com> Message-ID: > -----Original Message----- > From: thesite-admin at lists.evolt.org > [mailto:thesite-admin at lists.evolt.org]On Behalf Of Daniel J. Cody > Sent: 14 March 2001 13:48 > To: thesite at lists.evolt.org > Subject: Re: [thesite] RDF/RSS again > > > alrighty, check this out: > http://evolt.org/evolt/xml/rdf.cfm > > that validated through the my.netscape shit, so i would assume its good. > does it need to have a .rss extension for others to use it? Still doesn't display in my browser, but it did OK for GeekLog at http://www.newmediawhore.com. I suspect the browser thing will choke NS6, though. ------------------------------------------ John Handelaar T +44 20 7209 4117 M +44 7930 681789 F +44 870 169 7657 E john at userfrenzy.com ------------------------------------------ From djc at starkmedia.com Wed Mar 14 08:04:29 2001 From: djc at starkmedia.com (Daniel J. Cody) Date: Wed Mar 14 08:04:29 2001 Subject: [thesite] RDF/RSS again References: Message-ID: <3AAF7AB0.312C74A8@starkmedia.com> hmm.. thats odd, but i really dont know enough about this stuff to comment yet.. are there any other mainstream places we could submit it to see if it works? John Handelaar wrote: > Still doesn't display in my browser, but it did OK for GeekLog > at http://www.newmediawhore.com. I suspect the browser thing > will choke NS6, though. From r937 at interlog.com Wed Mar 14 08:12:59 2001 From: r937 at interlog.com (rudy) Date: Wed Mar 14 08:12:59 2001 Subject: [thesite] rudy! need some answers :) Message-ID: <01c0ac90$f8e37c20$8751149a@rudy> > ok, so i got a fire under my ass again and am > working on some of our features.. We get signal. > rudy, can you explain how the members, membattr, and > users tables are interconnected? Main screen turn on. > why didnt we just add another column to the users table > that labeled them as a member, user, god, or admin > instead of creating new tables? For great justice. > Just needing a memory refresh so i can get working > on these member pages You are on the way to destruction. ;o) give me a couple hours, please, and i'll write something up rud? From djc at starkmedia.com Wed Mar 14 08:15:24 2001 From: djc at starkmedia.com (Daniel J. Cody) Date: Wed Mar 14 08:15:24 2001 Subject: [thesite] rudy! need some answers :) References: <01c0ac90$f8e37c20$8751149a@rudy> Message-ID: <3AAF7D3C.CFB43A40@starkmedia.com> coolio :) > give me a couple hours, please, and i'll write something up From djc at starkmedia.com Wed Mar 14 14:27:34 2001 From: djc at starkmedia.com (Daniel J. Cody) Date: Wed Mar 14 14:27:34 2001 Subject: [thesite] wee! Message-ID: <3AAFD476.2D3CEA71@starkmedia.com> thanks to john for getting my going on something while i wait for rudy ;) add us to your netscape 6/mozilla sidebar: http://members.evolt.org/djc/ fun fun :) From bobd at members.evolt.org Wed Mar 14 16:25:51 2001 From: bobd at members.evolt.org (Bob Davis) Date: Wed Mar 14 16:25:51 2001 Subject: [thesite] wee! In-Reply-To: <3AAFD476.2D3CEA71@starkmedia.com> Message-ID: on 3/14/01 3:28 PM, Daniel J. Cody at djc at starkmedia.com wrote: > add us to your netscape 6/mozilla sidebar: > > http://members.evolt.org/djc/ > > fun fun :) That's so cool Dan! thanks! bob -- bob davis bobd at members.evolt.org From roselli at earthlink.net Sun Mar 18 14:24:22 2001 From: roselli at earthlink.net (aardvark) Date: Sun Mar 18 14:24:22 2001 Subject: [thesite] FAQ category note Message-ID: <3AB4C520.30835.EC33350@localhost> just wanted to throw this back out there... at SXSW i had proposed that we create a new category (we have code, news, etc.) called 'FAQ'... that category would have the following properties: - would not appear in the drop-down... - would not appear as a category on the article submission page... - would hold all 'about evolt' type stuff in convenient article formats... now, links at the top of the page (like FAQ) or our submission style guide, or other related things, would live in that category... whenever we would need to reference them, we'd link to them like any other article, but via the links we have in place... some advantages: - they fall into the new article process... - they can be edited by all, approved by all, etc... - no managing of static pages to worry about... - slides right into our templates... - searches can hit them... i'm sure y'all can think of other benefits... anyway, there were enough other folks in the hotel room who seemed to agree that i thought i'd bring it up... From seth at sethbienek.com Fri Mar 23 15:52:19 2001 From: seth at sethbienek.com (Seth Bienek) Date: Fri Mar 23 15:52:19 2001 Subject: [thesite] RE: Marriage (was: Re: [thechat] Announcement) In-Reply-To: <3ABBC262.83CA77BC@starkmedia.com> Message-ID: Doh! Don't you hate it when you set something aside for "a few minutes" and never get back to it? I'll play with it some this afternoon. Is it ok if I spank the hell out of LEO? Er, maybe I should continue this on thesite list... Seth ------------------------------ Seth Bienek Solutions Development Manager Stonebridge Technologies, Inc. 972.455.7294 tel 972.404.9754 fax ICQ #7673959 ------------------------------ > -----Original Message----- > From: thechat-admin at lists.evolt.org > [mailto:thechat-admin at lists.evolt.org]On Behalf Of Daniel J. Cody > Sent: Friday, March 23, 2001 3:39 PM > To: thechat at lists.evolt.org > Subject: Re: Marriage (was: Re: [thechat] Announcement) > > > sure, as soon as you're done with the archives and 's ;) > > Seth Bienek wrote: > > > > Does this mean that MG is nearing some kind of release?! :) > From djc at starkmedia.com Wed Mar 28 15:47:22 2001 From: djc at starkmedia.com (Daniel J. Cody) Date: Wed Mar 28 15:47:22 2001 Subject: [thesite] permalinks for comments.. Message-ID: <3AC25BA3.7F6FDD7D@starkmedia.com> I know i mentioned them a while back cus of a suggestion javi had, but i just added them in there.. I used the '#' sign for the actual link, but that can easily be changed... if someone thinks it should go elsewhere or be different, lemme know :) .djc. From matt at haughey.com Wed Mar 28 15:55:13 2001 From: matt at haughey.com (Matthew Haughey) Date: Wed Mar 28 15:55:13 2001 Subject: [thesite] permalinks for comments.. In-Reply-To: <3AC25BA3.7F6FDD7D@starkmedia.com> Message-ID: <5.0.2.1.2.20010328135243.00ae7998@lepen.pair.com> On 03:46 PM 3/28/2001 -0600, Daniel J. Cody wrote: >I know i mentioned them a while back cus of a suggestion javi had, but i >just added them in there.. I used the '#' sign for the actual link, but >that can easily be changed... if someone thinks it should go elsewhere >or be different, lemme know :) I get 404's when I click on them (there's a comments folder being appended to the path) but I could hack the URL to get at the proper comment, so they do work if the URL was correct. On metafilter, I used to have "permalink" href'd in the footer of every comment, but later on I streamlined the interface, by stashing the href around the time something was posted. For evolt's interface, I'd suggest just replacing the "#" link with maybe "link to this" to make it obvious, to help people along with getting used to seeing it and using it to reply to specific comments. Matt From djc at starkmedia.com Wed Mar 28 15:58:48 2001 From: djc at starkmedia.com (Daniel J. Cody) Date: Wed Mar 28 15:58:48 2001 Subject: [thesite] permalinks for comments.. References: <5.0.2.1.2.20010328135243.00ae7998@lepen.pair.com> Message-ID: <3AC25E50.D3E4C97B@starkmedia.com> Matthew Haughey wrote: > > On 03:46 PM 3/28/2001 -0600, Daniel J. Cody wrote: > >I know i mentioned them a while back cus of a suggestion javi had, but i > >just added them in there.. I used the '#' sign for the actual link, but > >that can easily be changed... if someone thinks it should go elsewhere > >or be different, lemme know :) > > I get 404's when I click on them (there's a comments folder being appended > to the path) but I could hack the URL to get at the proper comment, so they > do work if the URL was correct. ya, noticed that right after i sent out the email.. fixed now > On metafilter, I used to have "permalink" href'd in the footer of every > comment, but later on I streamlined the interface, by stashing the href > around the time something was posted. > > For evolt's interface, I'd suggest just replacing the "#" link with maybe > "link to this" to make it obvious, to help people along with getting used > to seeing it and using it to reply to specific comments. what do you think about the 'clutter' factor then? where would you recommend putting it? i like what you did with the mefi ones, but it may be a nice idea to put it in the wide open to begin.. thanks matt .djc. From matt at haughey.com Wed Mar 28 16:04:43 2001 From: matt at haughey.com (Matthew Haughey) Date: Wed Mar 28 16:04:43 2001 Subject: [thesite] permalinks for comments.. In-Reply-To: <3AC25E50.D3E4C97B@starkmedia.com> References: <5.0.2.1.2.20010328135243.00ae7998@lepen.pair.com> Message-ID: <5.0.2.1.2.20010328140022.02719008@lepen.pair.com> On 03:57 PM 3/28/2001 -0600, Daniel J. Cody wrote: > > For evolt's interface, I'd suggest just replacing the "#" link with maybe > > "link to this" to make it obvious, to help people along with getting used > > to seeing it and using it to reply to specific comments. > >what do you think about the 'clutter' factor then? where would you >recommend putting it? i like what you did with the mefi ones, but it may >be a nice idea to put it in the wide open to begin.. It's a clutter yeah, you could maybe reduce "link to this" to just "link." I just fear that if it's a pound sign or hidden link in the timestamp, people might not notice this "new" thing. At metafilter, I started out making it as obvious as possible, then when people were fully using them to reply to things (since the comments aren't threaded), it made it easier to reduce clutter, while still getting people to use them (since they were still seeing others using them). so the least clutter would just be linking the timestamp, since that's already in the interface, but there's a risk people won't know what it links to or why they would use that link. it's a tossup. Matt From djc at starkmedia.com Thu Mar 29 11:07:40 2001 From: djc at starkmedia.com (Daniel J. Cody) Date: Thu Mar 29 11:07:40 2001 Subject: [thesite] new sidebars.. Message-ID: <3AC36B84.916A0904@starkmedia.com> i just added a 'most commented on articles' sidebar, that displays the 5 articles with the most comments and hw many comments each has.. something i've been thinking about fora while as people tend to gravitate to the articles that have the most feedback on them.. another one i'm thinking about is the 'Top 5 Rated Articles' sidebar.. basically grab the highest average rating(minimum 5 votes).. opinions? thoughts? its pretty obvious when i'm bored at work i think, when i start adding new features to the evolt site(m.e.o was born cus i had a whole week free.. ;) thankie .djc. From roselli at earthlink.net Thu Mar 29 11:17:13 2001 From: roselli at earthlink.net (aardvark) Date: Thu Mar 29 11:17:13 2001 Subject: [thesite] new sidebars.. In-Reply-To: <3AC36B84.916A0904@starkmedia.com> Message-ID: <200103291717.f2THH8515623@leo.evolt.org> > From: "Daniel J. Cody" > > i just added a 'most commented on articles' sidebar, that displays the > 5 articles with the most comments and hw many comments each has.. > something i've been thinking about fora while as people tend to > gravitate to the articles that have the most feedback on them.. cool... is there a master page a person could go to in order to see the next 5, or 10, etc? > another one i'm thinking about is the 'Top 5 Rated Articles' sidebar.. > basically grab the highest average rating(minimum 5 votes).. coolio... From djc at starkmedia.com Thu Mar 29 11:22:53 2001 From: djc at starkmedia.com (Daniel J. Cody) Date: Thu Mar 29 11:22:53 2001 Subject: [thesite] new sidebars.. References: <200103291717.f2THH8515623@leo.evolt.org> Message-ID: <3AC36F11.B8FB0180@starkmedia.com> aardvark wrote: > > > From: "Daniel J. Cody" > > > > i just added a 'most commented on articles' sidebar, that displays the > > 5 articles with the most comments and hw many comments each has.. > > something i've been thinking about fora while as people tend to > > gravitate to the articles that have the most feedback on them.. > > cool... is there a master page a person could go to in order to see > the next 5, or 10, etc? not yet, but thats a good idea. i'll put it in the queue > > another one i'm thinking about is the 'Top 5 Rated Articles' sidebar.. > > basically grab the highest average rating(minimum 5 votes).. > > coolio... got any better way to make them show up? right now its TITLE : rating / number_of_ratings dunno if thats how you guys want it to look, its on the front page now.. .djc. From roselli at earthlink.net Thu Mar 29 11:49:17 2001 From: roselli at earthlink.net (aardvark) Date: Thu Mar 29 11:49:17 2001 Subject: [thesite] new sidebars.. In-Reply-To: <3AC36F11.B8FB0180@starkmedia.com> Message-ID: <200103291749.f2THn8517863@leo.evolt.org> > From: "Daniel J. Cody" > > got any better way to make them show up? right now its TITLE : rating > / number_of_ratings > > dunno if thats how you guys want it to look, its on the front page > now.. dunno if we need to know how many people rated it right there... maybe just show the rating? and how big is the top 10 list? other than 10 entries? or rather, i'd love to surf the site based on ratings... From djc at starkmedia.com Thu Mar 29 12:34:00 2001 From: djc at starkmedia.com (Daniel J. Cody) Date: Thu Mar 29 12:34:00 2001 Subject: [thesite] new sidebars.. References: <200103291749.f2THn8517863@leo.evolt.org> Message-ID: <3AC37FBE.9CF576D2@starkmedia.com> just hacked this together: http://evolt.org/rating/index.html top 50 with at least 5 ratings.. aardvark wrote: > and how big is the top 10 list? other than 10 entries? or rather, i'd > love to surf the site based on ratings... From roselli at earthlink.net Thu Mar 29 12:46:00 2001 From: roselli at earthlink.net (aardvark) Date: Thu Mar 29 12:46:00 2001 Subject: [thesite] new sidebars.. In-Reply-To: <3AC37FBE.9CF576D2@starkmedia.com> Message-ID: <200103291845.f2TIjx520643@leo.evolt.org> > From: "Daniel J. Cody" > just hacked this together: > > http://evolt.org/rating/index.html > > top 50 with at least 5 ratings.. you da man... on all 27 of 'em... From martin at easyweb.co.uk Thu Mar 29 12:48:48 2001 From: martin at easyweb.co.uk (martin burns) Date: Thu Mar 29 12:48:48 2001 Subject: [thesite] Re: [Admin] permalinks for comments.. In-Reply-To: <3AC25BA3.7F6FDD7D@starkmedia.com> Message-ID: <5.0.2.1.0.20010329195110.00a61990@pop.easyweb.co.uk> At 15:46 28/03/01 -0600, Daniel J. Cody wrote: >I know i mentioned them a while back cus of a suggestion javi had, but i >just added them in there.. I used the '#' sign for the actual link, but >that can easily be changed... if someone thinks it should go elsewhere >or be different, lemme know :) And they look *fantastic* - thanks so much. Cheers Martin From McCreath_David at xmail.asd.k12.ak.us Fri Mar 30 11:10:17 2001 From: McCreath_David at xmail.asd.k12.ak.us (McCreath_David) Date: Fri Mar 30 11:10:17 2001 Subject: [thesite] Tip Harvester question (was: [***] Formatting tips ) Message-ID: <9627DBCB3063D311BC4B00902785C55F02A12521@EXPO1> Seth's point brings up a question that I've had. If somebody (like me earlier today) replies to a with another , but leaves the original intact, does that original get picked up twice? I don't recall ever seeing duplicate tips in the harvester, but I've never really though to look. David -----Original Message----- From: Seth Bienek > > Make sure to format those tip tags correctly, or they won't make it into the tips database! From seth at sethbienek.com Fri Mar 30 13:29:27 2001 From: seth at sethbienek.com (Seth Bienek) Date: Fri Mar 30 13:29:27 2001 Subject: [thesite] Tip Harvester question (was: [***] Formatting tips ) In-Reply-To: <9627DBCB3063D311BC4B00902785C55F02A12521@EXPO1> Message-ID: Hi David, Good question.. The tip parser that I wrote looks for a line break followed by and includes everything between the closing bracket and the first bracket of . If it doesn't find the closing tag, it throws the whole thing out. The drawback to this approach is that if someone replies and their email software doesn't support "quoting", the tip would go in twice. I have put some thought into this, but have not been able to come up with a real solution other than a review process, or doing exact comparisons of every tip (which seems like it would inhibit performance, and I'm not sure if you can even compare two whole text fields..) I'm open to suggestions.. BTW, the tip parser doesn't currently run, it hasn't been running in 8 months or so.. The code is written, but we are working on putting the archives into a database, and then we'll set up the tip parser to check against the database. Seth ------------------------------ Seth Bienek Solutions Development Manager Stonebridge Technologies, Inc. 972.455.7294 tel 972.404.9754 fax ICQ #7673959 ------------------------------ > -----Original Message----- > From: thesite-admin at lists.evolt.org > [mailto:thesite-admin at lists.evolt.org]On Behalf Of McCreath_David > Sent: Friday, March 30, 2001 11:11 AM > To: 'thesite at lists.evolt.org' > Subject: [thesite] Tip Harvester question (was: [***] Formatting tips ) > > > Seth's point brings up a question that I've had. > > If somebody (like me earlier today) replies to a with another , > but leaves the original intact, does that original > get picked up > twice? > > I don't recall ever seeing duplicate tips in the harvester, but I've never > really though to look. > > David > > -----Original Message----- > From: Seth Bienek > > > > > > > Make sure to format those tip tags correctly, or they won't make > it into the > tips database! > > > _______________________________________________ > http://lists.evolt.org/thesitearchive/ > and new & improved kentucky fried old archives: > http://lists.evolt.org/thesitearchive/old/ > > From McCreath_David at xmail.asd.k12.ak.us Fri Mar 30 13:42:32 2001 From: McCreath_David at xmail.asd.k12.ak.us (McCreath_David) Date: Fri Mar 30 13:42:32 2001 Subject: [thesite] Tip Harvester question (was: [***] Formatting tips ) Message-ID: <9627DBCB3063D311BC4B00902785C55F02A1252B@EXPO1> Hi, Seth -- I just want to make sure I understand this. >The drawback to this approach is that if someone replies and >their email software doesn't support "quoting", the tip would >go in twice. Okay, so my message looked like this: ------------------------ > From: Daniel J. Cody > > > [ ... yadda yadda yadda ... ] > [ ... yadda yadda yadda ... ] David ------------------------ My client (Outlook Exchange client, set to send text-only) quotes replies with a ">" on every line, which is what it did to Dan's . So are you saying that the quoting ">"s on every line would have prevented his tip from being snagged twice? What about somebody like .jeff who quotes all his messages thusly? :~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~ : :[ ... yadda yadda yadda ... ] : :~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~ [ ... yadda yadda yadda ... ] >I have put some thought into this, but have not been able to >come up with a real solution other than a review process, or >doing exact comparisons of every tip (which seems like it would >inhibit performance, and I'm not sure if you can even compare >two whole text fields..) It doesn't seem like it happens often enough to really worry about it a lot, but maybe some monthly maintenance routine? Or maybe even a way for volunteers to delete duplicates if they spot them... hmmm... Thanks David From seth at sethbienek.com Fri Mar 30 13:56:32 2001 From: seth at sethbienek.com (Seth Bienek) Date: Fri Mar 30 13:56:32 2001 Subject: [thesite] Tip Harvester question (was: [***] Formatting tips ) In-Reply-To: <9627DBCB3063D311BC4B00902785C55F02A1252B@EXPO1> Message-ID: So are you > saying that the quoting ">"s on every line would have prevented > his tip from > being snagged twice? Yes. The parser looks for a clean line break followed by "This is a fish This is cow And these would not: : > > > by the way, here comes a > It doesn't seem like it happens often enough to really worry > about it a lot, It doesn't happen too often, but I've been paying close attention and they do come once in awhile.. I hate to even consider the idea of human intervention, but it looks like that may be the only fasible solution to this problem. While we're at it, I'm not really sure what to do with the type="" attribute. I mean, stick it in a database, sure, but then what? It would be cool if people could search by type, but there is not really a standard for the type attribute.. For instance, the same tip might get the type "coldfusion", "Cold Fusion" (<- with a space), "Optimizing ColdFusion Apps", "FuseBox", etc... The only solution I really see for that is either using the type attribute for display purposes only, or again, using human eyeballs to make adjustments, which means people-time, which is something we're short on.. I have considered using a look-up table, but it seems like it could get messy.. These are the things that keep me awake at night.. ;) The perils of being a geek... Seth ------------------------------ Seth Bienek Solutions Development Manager Stonebridge Technologies, Inc. 972.455.7294 tel 972.404.9754 fax ICQ #7673959 ------------------------------ From dsmah at home.com Fri Mar 30 14:05:02 2001 From: dsmah at home.com (Dean Mah) Date: Fri Mar 30 14:05:02 2001 Subject: [thesite] Tip Harvester question (was: [***] Formatting tips ) In-Reply-To: from "Seth Bienek" at Mar 30, 2001 01:29:34 PM Message-ID: <200103302009.NAA01759@alice.monkeyland.ca> When I was working on the Perl version I grabbed tips where the tag appeared as the first non-whitespace characters. I would also skip over the part of a message that followed the "--- Original Message ---" text that Outlook tends to put in for quoted messages. I would also handle tips within tips, i.e., if you had something like: it would pull it out as two separate tips, which would happen from time to time. Dean Seth Bienek writes: > The tip parser that I wrote looks for a line break followed by and > includes everything between the closing bracket and the first bracket of > . If it doesn't find the closing tag, it throws the whole thing out. > > The drawback to this approach is that if someone replies and their email > software doesn't support "quoting", the tip would go in twice. > > I have put some thought into this, but have not been able to come up with a > real solution other than a review process, or doing exact comparisons of > every tip (which seems like it would inhibit performance, and I'm not sure > if you can even compare two whole text fields..) > > I'm open to suggestions.. From mwarden at odyssey-design.com Fri Mar 30 14:16:12 2001 From: mwarden at odyssey-design.com (Warden, Matt) Date: Fri Mar 30 14:16:12 2001 Subject: [thesite] Tip Harvester question (was: [***] Formatting tips ) References: Message-ID: <018501c0b956$0368cde0$beae1b18@cinci.rr.com> > Yes. The parser looks for a clean line break followed by " would be valid: > > This is a fish > > This is cow > > > And these would not: > > : > > > > > by the way, here comes a this is good info. I had always changed the < to [ and the > to ]. So: remove every zig would become (in my reply): [tip] remove every zig [/tip] glad I don't have to do that anymore. *cheer* -- mattwarden mattwarden.com From seth at sethbienek.com Fri Mar 30 14:24:42 2001 From: seth at sethbienek.com (Seth Bienek) Date: Fri Mar 30 14:24:42 2001 Subject: [thesite] Tip Harvester question (was: [***] Formatting tips ) In-Reply-To: <200103302009.NAA01759@alice.monkeyland.ca> Message-ID: Hey Dean, The flexibility of being able to parse tips within tips is useful, but are we willing to assume the performance hit of recursive parsing? Also, tip tags should not be nested IMO because each tip is it's own simple entity, and keping track of nested tips' relationships seems like it would be a nightmare, unless the idea is to store them independent of their parent(s) or child(ren), in which case, is the parent tag incomplete without the tip that was contained inside it? Or is the 'child' tip left in the original tip and stored redundantly as it's own tip as well? I am open to discussion on this because it's a possibility that I'm overlooking some of the usefulness of this feature. Parsing only text before "-----Original Message-----" could keep good tips from being stored, depending on the method people use to reply. Also, I think that Outlook is the mail prog of choice for only a small percentage of our members (unfortunately), so I would question whether this feature is worth the processing overhead for the additional messages. Again, I could be talked out of these assumptions. Sometimes thinking "outside the box" is a challenge for me. Do I get points for using the obnoxious cliche? Seth ------------------------------ Seth Bienek Solutions Development Manager Stonebridge Technologies, Inc. 972.455.7294 tel 972.404.9754 fax ICQ #7673959 ------------------------------ > -----Original Message----- > From: thesite-admin at lists.evolt.org > [mailto:thesite-admin at lists.evolt.org]On Behalf Of Dean Mah > Sent: Friday, March 30, 2001 2:09 PM > To: thesite at lists.evolt.org > Subject: Re: [thesite] Tip Harvester question (was: [***] Formatting > tips ) > > > When I was working on the Perl version I grabbed tips where the tag > appeared as the first non-whitespace characters. I would also skip > over the part of a message that followed the "--- Original Message > ---" text that Outlook tends to put in for quoted messages. I would > also handle tips within tips, i.e., if you had something like: > > > > > > > it would pull it out as two separate tips, which would happen from time > to time. > > Dean From roselli at earthlink.net Fri Mar 30 14:38:02 2001 From: roselli at earthlink.net (aardvark) Date: Fri Mar 30 14:38:02 2001 Subject: [thesite] Tip Harvester question (was: [***] Formatting tips ) In-Reply-To: References: <200103302009.NAA01759@alice.monkeyland.ca> Message-ID: <200103302037.f2UKbx501410@leo.evolt.org> > From: "Seth Bienek" > > The flexibility of being able to parse tips within tips is useful, but > are we willing to assume the performance hit of recursive parsing? ugh... as long as you aren't storing them nested in any way... > Also, tip tags should not be nested IMO because each tip is it's own > simple entity, and keping track of nested tips' relationships seems > like it would be a nightmare, unless the idea is to store them > independent of their parent(s) or child(ren), in which case, is the > parent tag incomplete without the tip that was contained inside it? Or > is the 'child' tip left in the original tip and stored redundantly as > it's own tip as well? I am open to discussion on this because it's a > possibility that I'm overlooking some of the usefulness of this > feature. no nesting... that's just... well... absurd... > Parsing only text before "-----Original Message-----" could keep good > tips from being stored, depending on the method people use to reply. > Also, I think that Outlook is the mail prog of choice for only a small > percentage of our members (unfortunately), so I would question whether > this feature is worth the processing overhead for the additional > messages. Again, I could be talked out of these assumptions. > Sometimes thinking "outside the box" is a challenge for me. even when i used outlook for list mail, i *always* trimmed the 'original message' crap... that's what '>' are for... too bad not everybody uses those... we need a primer on email again, and we need to kill apps like Groupwise... anyway, no, there is no generic standard we could use... > Do I get points for using the obnoxious cliche? 15? From r937 at interlog.com Fri Mar 30 14:42:27 2001 From: r937 at interlog.com (rudy) Date: Fri Mar 30 14:42:27 2001 Subject: [thesite] Tip Harvester question (was: [***] Formatting tips ) Message-ID: <01c0b95a$04e6be80$df4d149a@rudy> > Sometimes thinking "outside the box" is a challenge for me. i dunno, i think you've been doing fine so far as far as the harvesting logic is concerned, i dunno if scanning for "\n" and parse everything inside it as name/value pairs then scan ahead and find "" and everything in between is the tip body the name/value pairs get stored in the database under multiple categories or whatever(*) yes, this might result in duplications when tips are (re)posted in replies, etc. but so what? look at how matt and michele have gone through tons of old articles cleaning them up besides, we are not taking advantage of a huge untapped labour force -- the authors themselves (*) special note -- please let me know when you want to start testing harvest code and i'll be sure to have a document ready within a day explaining the evolt tables and relationships oh, and i just remembered, you must extract the date and author's email id from the post (headers?) rudy From dsmah at home.com Fri Mar 30 14:57:47 2001 From: dsmah at home.com (Dean Mah) Date: Fri Mar 30 14:57:47 2001 Subject: [thesite] Tip Harvester question (was: [***] Formatting tips ) In-Reply-To: from "Seth Bienek" at Mar 30, 2001 02:24:53 PM Message-ID: <200103302102.OAA01987@alice.monkeyland.ca> Seth Bienek writes: > The flexibility of being able to parse tips within tips is useful, > but are we willing to assume the performance hit of recursive > parsing? I don't think that I implemented it in Perl recursively. The Perl performance was fine. I'm not knowledgable in Cold Fusion and so do not know the performace impact that it has there. > Also, tip tags should not be nested IMO because each tip is it's own > simple entity, I would agree. I'm just relating some of the experiences that I've had while running the Tip Harvester for a couple of years. I also allowed parsing of tags that didn't appear like tags. Things like |tip|...|/tip| and (tip)...(/tip). > Parsing only text before "-----Original Message-----" could keep > good tips from being stored, depending on the method people use to > reply. I had no problems with this. The only time that I can seen it being a large problem is when someone replies inline to a message but still uses the --Original Message-- delimiter. Again in Perl these things weren't a real performance issue but I can't speak for CF. Dean From djc at starkmedia.com Fri Mar 30 15:03:47 2001 From: djc at starkmedia.com (Daniel J. Cody) Date: Fri Mar 30 15:03:47 2001 Subject: [thesite] if anyones interested.. Message-ID: <3AC4F435.977EADEC@starkmedia.com> here is a list of terms that were used in google searches that pointed to the browser archive.. i know i'll probably have this info when i run the logs tomorrow, but its nice to have it seperated kinda.. oh, btw 50,000+ referers from google this month *on the lists.evolt.org site alone*(doesnt have anything to do with the data below..) they really like us :) -------- Original Message -------- Date: Fri, 30 Mar 2001 14:59:18 -0600 From: root To: djc at starkmedia.com Subject: arabic user interface DR-Webspyder Dr WebSpyder Dr-webspyder Dynamic font earch%3Asourc IE Solaris Mobile Browsing Sprynet Mail acroread.exe address bar icon address bar icon af arabic module download autosurf & browser archiv download browsers installing OmniWeb long filename manual for minuet microphone download net sentry netpositive netpositive netscape 6 microsoft proxy browser netscape 6 socks proxy o old browser archive old web browsers tn older browser versions perl filter stdin product evaluation agreement productivity works tn refresh all open windows u url blocker web browsers windows 2000 keyboard unicode input cyrillic %2B netscape-i686-pc-linux-gnu-sea.tar.gz %2B1600x1200 %2Btouch %2Bscreen %2BCello %2Bbrowser %2BHandWeb %2Bfull %2Bdownload %2BNetscape %2Bprefs.js %2Bsize %2Bzero %2BPSM %2BBeonex %2Bamiga %2Bvoyager %2Bkeyfile tn %2Bbobby %2Bdownload %2Bblind %2Bbrowsers %2Bdownload %2Barchive %2Bbrowsers %2Bdownload %2Bold %2Bbrowsers %2Bold %2Bbrowser tn %2Bunzip.exe %2Bdownload %2Bdos %2Fusr%2Fbin%2Frun earch%3Asourc %2b activex is an open technology %2b roger gross & %2bBig5 %2bcharset %2bxml %2bparse %2bISP %2b static IP %2b905& %2bKOI8 %2bNetscape %2bSun %2buser_pref %2bNetscape %2bsmtp %2bbug %2bprefs.js %2bhotjava%2bbrowser%2bconfiguration& %2bmicroviet %2bbrowser %2bnetscape %2bcache %2bfat.db %2bformat %2bold %2bbrowsers %2bDownload %a3%f2%aedZ %d4f K%e2%f3%a7 %a3%f8%a3%f8%a3%f8pic %a5_%a8%ca%a5j%c2%dd%ae%c8%b9C%c2I %ac%fc%b0%ea%c1p%a6X%c4U%b6%d2 %b9%fc%c7%c7 %b0%d4%c0%d3 %be%c6%c0%cc%b5%f0%26%be%cf%c8%a3 %b5%b5%bf%eb %c9%ab%c7%e9%d3%d7%c5%ae%cd%f8%d5%be %d6%d0%b9%fa%bb%c6%cd%f8%d7%dc%d5%be& %df%d3 %cd%e1%e6 %f2%e5%e5%f8 %fdcq%2b%fdp -> .mpe video& 00311148772807112856& 00325139192700359572& 00342876862834015745& 1.44 diskette 100 12345%2f%2fdata 16ivw238.exe tn 1xsetup.exe 20 30 40bit browsers 50 ACROREAD.EXE Acroread.exe tn Adobe Acroread.exe tn Amiga Voyager keyfile Arabic Display Interface Arachne Internet Browser Archive Web Browser BROWSER BROWSER ARCHIVE BeOS Browser Archive Browser archive Browser key 128 RC4 export Browser links Browsers Browsers %2Bfree Browsers Code Browsers ourcei Browsers tn Buddy the internet helper Robot CC32Demo.exe tn CC32demo.exe tn CSS align table Caldera WebSpyder download Caldera%20Webspyder tn Chimera web browser Circle Group Internet CyberPassage Cyberbit 2.0 Cyberbit 2.0 and Bold Cyberdog 2 browser DEF.MAK DOWNLOAD%20OPERA%204%20BROWSER tar DRWEBDEM.EXE DRWEBDEM.EXE tn DeleGate MacOSX DiskCopy download Download SOCKS 5 Proxy Windows 2000 free Downloadable Spanish Translaters Dutch language translaters EINET browser ourcei First Graphical Web browser First graphical web browser Free ISP in the Phillippines& HandWeb Handweb browser& Hexabit%2Bweb%2Bbrowser tn HotJava Browser 3.0 source code tn HotJava HTML Component tn HotJava linux IE Solaris IE browser archive IE for Solaris IE on solaris h IME reconversion Internet Browser download mosaic netscape amaya Internet Browsers Internet Browsers tn Internet Explorer Browser Archive Internet Explorer browser archive Internet browsers Internet browsers Internet browsers tn Internet web browsers IpixViewer JRE 1.1.5 Windows& Java%201.1.5 tar JavaOSTM 1.1 Javascript 1.5 LIST OF WEB BROWSERS tn MDI browser MacCroatian Fonts MacNS6FullInstaller.sea.bin tn MacNetscape6Installer.sea.bin Malformed hex mac address Minuet%27s FTP Mosaic dialer Mozilla%27s DOM body text Mozilla%27s DOM body text. NCSA Mosiac NCSetup NETSCAPE frame size bug pixel marginwidth ourcei NatDat Corporation NetPositive NetPositive 3 NetPositive 3.0d3 NetPositive 3.0d3 tn NetPositive update Netpositive Netpositive Development tn Netscape %26 Old Browsers O351e32 Old Browsers explorer tn Old Browsers tn Old Internet Explorer archive Old Web Browsers Old browsers Old web browsers Old%2Bweb%2Bbrowsers tn OmniWeb OpenUrl Visual Basic Password Protected directory PS browser PWwebspeak Palm Handweb download Quarterdeck Mosaic Quicktime Problems On WinNT4 SSMTP a ScriptActive Simulbrowse SiteKiosk SlipKnot%20 %20browser ourcei Spry dialer& Swinst4.exe TALKBACK.EXE Talkit& Talva U tar URL Blocker UTF-8 earc V25_B117.EXE& V25_B117.exe VC%2b%2b jpeg& WIN31 BROWSER Web Browser Reviews Web Browsers Web Browsers Web browsers Web browsers tn Web%2Bbrowsers WebSpyder WebSpyder tn WebTV Viewer Webprowler tn Webspyder What is a browser%3f WinWeb browser Windows CE Release notes %26 IMAP %26 Netscape XKeysymDB solaris x86 abook mab abook.mab accent software& accessible disabilities rate bobby acroread .exe acroread%2Bexe tn acroread.exe acroread.exe acroread.exe tn activex browser activex browser add trailing spaces to String adobe acroread.exe alert dialog in JavaScript using UTF-8 tn all about browsers alte browser evolt amaya aix amiga voyager hack animated atom animated baloon gif antialias visual basic source arabic module windows arachne browser archive browsers archived browsers archived browsers for download internet explorer netscape archived browsers internet explorer archived browsers tn archived browsers& ariadna browser ariadna browser ourcei ariadna browser tn ariadna-13-x86.exe atom net atomnet%2Bbrowser atomnetpe.exe autoexec.bat file temporary directory Netscape automatic AOL logon automatic browser refresh backward and rowspan barber pole jpeg beonex beos NetPositive download best browsers bg af bobby certification tn bohemian%20link tar brownie browser brownie browser browser browser amaya netscape opera browser archiv browser archiv tn browser archive browser archive browser archive IE browser archive tn browser archives browser design tn browser dos frames browser downloads browser emulator browser evolt browser evolt browser evolt ourcei browser evolution& browser list browser list browser multiple sites tab tn browser old browser. .org browser. .org browsers browsers browsers a browsers archive browsers h browsers old download ourcei browsers ource-i browsers ourcei browsers tn browsers u browsers.com bugtrack& bugzilla delete profile cab archive cache:browsers.evolt.org/ cache:browsers.evolt.org/ Browsers cache:browsers.evolt.org/ escape browsers cache:browsers.evolt.org/ minuet internet browser cache:browsers.evolt.org/1x/0.12a/1XFAQ.htm remove ie3 95 cache:browsers.evolt.org/dr-webspyder/20/download.html webspyder cache:browsers.evolt.org/mozilla/m13/m13-detail.html netscape https document.write print bug OR problem OR fix cache:browsers.evolt.org/netpositive/3.0d3_Intel_R4/r4_netpositive_3.htm javascript for NetPositive cache:browsers.evolt.org/omniweb/OPENSTEP_Mach4.2/3.0/releasenotes.html dateWithNaturalLanguageString cache:browsers.evolt.org/omniweb/OPENSTEP_Mach4.2/3.0/releasenotes.html ie won%27t load qt movies cache:browsers.evolt.org/webexplorer/OS2/1.1h/adobe.htm cache:browsers.evolt.org/webexplorer/OS2/1.1h/adobe.htm acroread.exe cache:jcintern.uoregon.edu/macbrowsers/ mac browsers cache:www.evolt.org/article/Your_clients_need_a_Content_Management_System/20/5127/ content management cache:www.guppi.de/links.shtml browserarchiv download cache:www.hoary.org/browse/ webspyder javascript download caldera dos download& caldera download caldera download can netscape view txt files carboncopy32 cartoon pictures of a monkey party cd writer compression softwares ce korean ime cello browser cello browser cello browser tn cello web browser download charlie.zip check if an instance of Internet Explorer is running before open a new window with javascript chinese simplified text support citrix full screen mode clearing browsers history object while using session tracking code browser& code browsers comprehensive browser list comprehensive collection of browsers tn computer dealer in phillippines configure cyberdog 2 apop configure pc speaker on sound card content.rdf parse cookies netscape clark release cross frame custom browser design cybergate browser de de af deleting abook.mab file from netscape 6 address file tn dialup usernames and passwords hacking disk image .144& diskcopy .img on a pc diskcopy download diskcopy software& download 4.01 english download SPRY Dialer download big5 pcf fonts download browsers netscape old collection download diskcopy download free slipknot browser download old browser download old browser u download old browsers download old browsers download old browsers o download old browsers tn download old web browser download pw1118.exe download pwwebspeak tn download webspyder downloading party invitations downloads browser dr webspyder dr-webspyder drag manager 68k edit.res embedded browser en en af enreach download es af escape browsers tn escape web browser every web browser evolt evolt evolt browser evolt browser archive evolt browser ourcei evolt mozilla download evolt tn evolt%2C browser evolt%2C browser archive evolt.org-browser Archive evolution of internet browsers explorer 4.5 mac download explorer bar folders explorer bars development explorer browser archive explorer right pane fi af finger.pl first graphical web browser first graphical web browser tn first internet browser it first web browser flash menu that you can resize and scroll tn fr af free acroread program free old browser free web browsers free web browsers& free%2BFtp%2Bcomponent%2Bwith%2Bunicode%2Bsupport freeware mpeg video player fullsoft.dll gifs of set top box u googlet& grasshopper browser tn handle the mousewheel event in javascript handweb 2.0 -demo heirarchical naming implemented in yahoo hexadecimal calculator& hotjava browser hotjava error hotjava html component hotjava html component hotjava linux hotjava linux tn hp 10.20 communicator installation html component& htmlview redraw iPixViewer tn ibrowse %2B amiga ibrowse 2 news ibrowse amiga ibrowse amiga tn ibrowse%2Bamiga icab download icab hilfe icab web browser ie for solaris ie win31 ie3 browser archive tn ie4 browser archive tn ie4 mac download ie5setup solaris ie5setup.zip imagemap outline border internet explorer inline viewer newsreader install plug-ins xpi netscape install z-modem unix installing bitstream cyberbit font in java interactive%2Binternet%2Bbrowsers international version of the JRE1.3 internet browser archive internet browsers internet browsers internet browsers h internet browsers ourcei internet browsers tn internet chameleon 4.5 download& internet explorer 4.5 macintosh download internet explorer archive internet explorer browser archive internet monkey helper internet%2Bbrowsers tn iod4 it it af java plugin 1.3 switch jre control panel java static methods recursion quiz java thin browser java%2bbrowsers& java, design of web browser javascript background image fixed content scroll works in netscape code javascript UTF-8 charset converting javascript activeX multi interface javascript alert box display chinese characters javascript alert display chinese javascript alert i18n javascript changing cell background netscape javascript dialog multi-byte javascript editable combo box& javascript position of highlight text in textfield javascript%3AOnClick textfield jdk %2b trouble %2b nt& jpg images in adobe acrobat tn jre1.3 plugins k%ebng keeping%20message%20content%20pane tar kidnet explorer korean text display support laetitia.jpg lahman internet tn latest browsers Mozilla list of browsers logitech mousewheel freeze logitech mousewheel windows 2000 u lynx_w32.zip m13 mozilla mac %26 68k %26 browsers mac 68k& mac browser archive mac browser archive mac os translaters macosx custom view drawing macosx hack focus cursor macweb macweb downloads mailto.pl mdi web browser mediaplayer%2b16bit meta charset japanese browser& minuarc minuet browser source mosaic source code mosaic%2bnetscape& mosiac in c%2b%2b code mouseover and submenu software& mousewheel event in javascript mozilla %26 mdi& mozilla beos mozilla for win mozilla plugins problems mozilla solaris mozilla-win32-installer-M18.exe ms proxy server 2.0 indonesia mspie11e.exe mspie11p.exe tn msword to text& navclient& navigator 16bit nc 5.0 download ncsetup neozilla neozilla neozilla ourcei netmanfull netomat netomatic netomatic files ourcei netpositive netpositive 3 ourcei netpositive 3 tn netpositive 3.0 netpositive 3.0 tn netpositive 3.0d netpositive 3.0d3 netpositive browser netpositive download netpositive problems netpositive%2Bbeos%2Bdownload netscape old browser archive tn netscape percentage widths image netscape 0.9 netscape 6 mozilla gecko faq history& netscape Gecko plugin Browser Helper Object netscape crashes layers maximize workaround netscape https document.write print u netscape initialization file netscape mdi& netscape navigation toolbar focus& netscape%2Bexplorer%2Bbrowsers netscape-i686-pc-linux-gnu-sea.tar.gz netsentry netshark netshow20 news browsers nl af nntp.pl npjava13.dll nsiwidsrc earc nt and support and french and dutch and cv& nt explorer right pane is not displayed o351e32.exe off-line web browsing mail offline folder 2000 windows old Browser& old alte Browser download tn old archived browsers old browser old browser old browser archive old browser archive tn old browser download old browser download tn old browser downloads old browser downloads old browser downloads tn old browser software archive tn old browser tn old browser& old browsers old browsers old browsers Lynx download old browsers archive download old browsers download old browsers download ourcei old browsers download tn old browsers download archive old browsers download tn old browsers downloads old browsers ource-i old browsers tn old browsers u old browsers.com old internet browsers c old mosaic download old netscape browser u old web Browser collection download tn old web browser archive old web browser download old web browsers old web browsers old web browsers download old web browsers for download old web browsers museum omni form viewer omni mess omni page hacks omni web broswer omniweb version 3.0 opera 16bit opera 3.62 software& opera browsers tn opera web browsers oracle%2Bbrowser p s_ palmIII browser paste XFree86 wheel mouse patching instructions pc-dos downoad pictures of inside of rocketship ppp users cannot browse microsoft network pragma-nocache processor equirement productivity works ourcei proxiweb 3.0 proxy omniweb problem preferences pwWebspeak quarterdeck browser quarterdeck browser tn quarterdeck corporation qwinsock read jpeg dimensions javascript resize dynamic reconversion IME redrawing cache images in netscape 6 related checkboxes javascript checking group remove ie3 95 resizing textareas in netscape reviews offline browser ripterm tn running aol on se%2F30 sa2800e.exe sample code for javascript Confirmation Dialogs scripts for quote follow-up search search for doc xls all& secure sites broswer hang u secure surfing secure surfing& see web old browser cello sendmail perl replyto sherlock imap sherlocksearch sizing submit buttons for table columns slipknot %234 slipknot web browser softvoice.com tar solaris a solaris ie browser solaris javascript multi-byte sprynet mosaic sprynet.exe spyglass browser spyglass mosiac spyglass web browser srpadlgoo earc srpwidsrc earc ssmtp 465& stop submit refreshing form javascript sv af tabs browser explorer& talkback.exe talkback.exe tango browser ourcei tar 1.12 solaris 2.6 download templates gnome password ftp desktop url default preferences application thai font solaris netscape& the first web browser themes for win32 toolbarwallpaper toolbarwallpaper tn toolbarwallpaper& touch screen browser client unicode converter windows-1252 utf-8 macroman v25_b117.exe view favorites offline organize virtual browser virtual web browser tn visio document viewer voyager browser vt340 lpr w s_ w32s.386 download w3c web browser web browsers web browser web browser archive web browser archive web browser java code web browser technology web browsers web browsers web browsers a web browsers ourcei web browsers tn web%2Bbrowsers tn webexplorer webexplorer%201%2E1h%20download tn webprowler webscripter 4 keygen webspyder webspyder webspyder and ppp webspyder browser tn webspyder download webspyder ourcei webspyder tn webtv browser downloads webtv for windows upgrades& webtv online viewer webtvinstall.exe webwhacker 2000& webwhacker mac download webwhacker v3 what is a browsers tn what is browser%3f& where to find old browsers wildcat browser win-cgi search engine windows 2000 java netscape repaint& windows and display and gif and explorer winnt cookies viewer& winweb browser winword.exe error troubleshooting& wordia tn worldwideweb browser tim ourcei wp510 tn wp510.zip tn www.browsers.com www.browsers.com zh-CN af From seth at sethbienek.com Fri Mar 30 15:07:02 2001 From: seth at sethbienek.com (Seth Bienek) Date: Fri Mar 30 15:07:02 2001 Subject: [thesite] Tip Harvester question (was: [***] Formatting tips ) In-Reply-To: <01c0b95a$04e6be80$df4d149a@rudy> Message-ID: > as far as the harvesting logic is concerned, i dunno if scanning for "\n > why not just scan for " -----Original Message----- > From: thesite-admin at lists.evolt.org > [mailto:thesite-admin at lists.evolt.org]On Behalf Of rudy > Sent: Friday, March 30, 2001 2:43 PM > To: thesite at lists.evolt.org > Subject: Re: [thesite] Tip Harvester question (was: [***] Formatting > tips ) > > > > Sometimes thinking "outside the box" is a challenge for me. > > i dunno, i think you've been doing fine so far > > as far as the harvesting logic is concerned, i dunno if scanning for > "\n > why not just scan for " > then look for the closing ">" and parse everything inside it as name/value > pairs > > then scan ahead and find "" and everything in between is > the tip body > > the name/value pairs get stored in the database under multiple categories > or whatever(*) > > yes, this might result in duplications when tips are (re)posted > in replies, > etc. > > but so what? look at how matt and michele have gone through tons of old > articles cleaning them up > > besides, we are not taking advantage of a huge untapped labour > force -- the > authors themselves > > > (*) special note -- please let me know when you want to start testing > harvest > code and i'll be sure to have a document ready within a day explaining the > evolt tables and relationships > > oh, and i just remembered, you must extract the date and author's email id > from the post (headers?) > > rudy > > > > > > _______________________________________________ > http://lists.evolt.org/thesitearchive/ > and new & improved kentucky fried old archives: > http://lists.evolt.org/thesitearchive/old/ > > From djc at starkmedia.com Fri Mar 30 15:09:47 2001 From: djc at starkmedia.com (Daniel J. Cody) Date: Fri Mar 30 15:09:47 2001 Subject: [thesite] Tip Harvester question (was: [***] Formatting tips ) References: <200103302102.OAA01987@alice.monkeyland.ca> Message-ID: <3AC4F59C.7BDA055E@starkmedia.com> Dean Mah wrote: > > Seth Bienek writes: > > > The flexibility of being able to parse tips within tips is useful, > > but are we willing to assume the performance hit of recursive > > parsing? > > I don't think that I implemented it in Perl recursively. The Perl > performance was fine. I'm not knowledgable in Cold Fusion and so do > not know the performace impact that it has there. the performance hit is going to come from inserting them into the DB as seth and i have seen.. i don't know if it would be any faster in another language or what.. but if we were really going after speed we'd be going compiled or pl/sql *shrug* From r937 at interlog.com Fri Mar 30 15:12:32 2001 From: r937 at interlog.com (rudy) Date: Fri Mar 30 15:12:32 2001 Subject: [thesite] Tip Harvester question (was: [***] Formatting tips ) Message-ID: <01c0b95e$0105de00$df4d149a@rudy> >> why not just scan for " >'cause this would snag all tips from quoted replies where the cheap bastards >did not trim their emails properly, or where they want to include the tip as >part of their reply.. Right? right, but my question is, so what? no offence i don't think it happens often enough to worry about, and if you use something restrictive like "\n1 rudy From seth at sethbienek.com Fri Mar 30 16:40:22 2001 From: seth at sethbienek.com (Seth Bienek) Date: Fri Mar 30 16:40:22 2001 Subject: [thesite] Tip Harvester question (was: [***] Formatting tips ) In-Reply-To: <01c0b95e$0105de00$df4d149a@rudy> Message-ID: > right, but my question is, so what? no offence None taken. My concern was that it would be an effort to de-dupe.. But... > dupes are easy to find, they lend themselves to strategies like Well, that sounds easy enough. :) I will make it be so. Seth ------------------------------ Seth Bienek Solutions Development Manager Stonebridge Technologies, Inc. 972.455.7294 tel 972.404.9754 fax ICQ #7673959 ------------------------------ From isaac at members.evolt.org Sat Mar 31 06:18:43 2001 From: isaac at members.evolt.org (isaac) Date: Sat Mar 31 06:18:43 2001 Subject: [thesite] new sidebars.. In-Reply-To: <3AC37FBE.9CF576D2@starkmedia.com> Message-ID: > just hacked this together: > > http://evolt.org/rating/index.html > > top 50 with at least 5 ratings.. another sidebar idea - maybe another 5 high ranking articles, but those which don't have the minimum five rankings. ie, i have some articles that are ranked "5" from 1-3 rankings. they really shouldn't be 5's. so, my idea is to have a box which encourages people to rank these to give a better indication of their true value. it'd be hard to describe this concept in a simple title, but maybe a title attrib onmouseover could help explain it. or, we could call it 'fresh and hot?' or 'up and coming'. who knows... another possibility is to re-raise half-decent articles to be featured/re-promoed on the frontdoor. and maybe a paragraph encouraging people to rank/comment if they haven't previously. i From dsmah at home.com Sat Mar 31 12:24:13 2001 From: dsmah at home.com (Dean Mah) Date: Sat Mar 31 12:24:13 2001 Subject: [thesite] Tip Harvester question (was: [***] Formatting tips ) In-Reply-To: from "Seth Bienek" at Mar 30, 2001 02:24:53 PM Message-ID: <200103311828.LAA00953@alice.monkeyland.ca> Without resorting to PL/SQL you may want to investigate some of the following to see if performance can be improved. I know that in some languages, e.g., Perl and C, you can use bind variables to speed up queries. Oracle can cache the SQL in its SGA and the do a variable binding/substitution just before execution. In Perl I've seen 50 times speed up on frequently executed queries. Also, connects/disconnects in Oracle are extremely expensive performance-wise. Ensure that you are only connecting once to Oracle and then doing all of the queries and then disconnect. You may also want to turn off auto-commits. If you are commiting after each query, Oracle needs to write the transaction to the redologs each time. Of course, this means that your rollback segments must be large enough to contain all of the uncommitted data. Dean From joshua at alphashop.net Sat Mar 31 13:42:23 2001 From: joshua at alphashop.net (Joshua OIson) Date: Sat Mar 31 13:42:23 2001 Subject: [thesite] Tip Harvester question (was: [***] Formatting tips ) References: <01c0b95e$0105de00$df4d149a@rudy> Message-ID: <017601c0ba34$01873dc0$38360141@mrtnz1.ga.home.com> Rudy, You wrote: > dupes are easy to find, they lend themselves to strategies like > > GROUP BY HAVING COUNT(*)>1 I'm trying to visualize how such code may be applied to the current situation. How exactly will this query find duplicate tips, especially when quotes in emails tend to be lead by any number of line-begin tokens? If this actually can work, I'd love to use it in a number of applications, so believe me, I'm not trying to be argumentative. -joshua From joshua at alphashop.net Sat Mar 31 13:45:53 2001 From: joshua at alphashop.net (Joshua OIson) Date: Sat Mar 31 13:45:53 2001 Subject: [thesite] new sidebars.. References: Message-ID: <018c01c0ba34$807a5360$38360141@mrtnz1.ga.home.com> Perhaps the side bar listing of the highest rated articles should be in context of the last 3 months so that fresh articles can still recycle through the sidebar. -joshua ----- Original Message ----- Subject: RE: [thesite] new sidebars.. > > > just hacked this together: > > > > http://evolt.org/rating/index.html > > > > top 50 with at least 5 ratings.. From roselli at earthlink.net Sat Mar 31 13:49:33 2001 From: roselli at earthlink.net (aardvark) Date: Sat Mar 31 13:49:33 2001 Subject: [thesite] new sidebars.. In-Reply-To: <018c01c0ba34$807a5360$38360141@mrtnz1.ga.home.com> Message-ID: <3AC5EDC1.22719.1E3A29D1@localhost> > From: "Joshua OIson" > > Perhaps the side bar listing of the highest rated articles should be > in context of the last 3 months so that fresh articles can still > recycle through the sidebar. i don't really agree with that... i think it gives the older articles the opportunity to come out into the light again... and since the older articles were pre-ratings, they're already at a disadvantage since so many are unrated... besides, i wanna see highest articles ever, not just the highest articles in some period of time... From bobd at members.evolt.org Sat Mar 31 13:53:38 2001 From: bobd at members.evolt.org (Bob Davis) Date: Sat Mar 31 13:53:38 2001 Subject: [thesite] new sidebars.. In-Reply-To: <3AC5EDC1.22719.1E3A29D1@localhost> Message-ID: <200103311954.f2VJsMU01021@chmls06.mediaone.net> On Saturday, March 31, 2001, at 02:46 PM, aardvark wrote: >> From: "Joshua OIson" >> >> Perhaps the side bar listing of the highest rated articles should be >> in context of the last 3 months so that fresh articles can still >> recycle through the sidebar. > > i don't really agree with that... i think it gives the older articles > the > opportunity to come out into the light again... and since the older > articles were pre-ratings, they're already at a disadvantage since so > many are unrated... > > besides, i wanna see highest articles ever, not just the highest > articles in some period of time... I'm with Aardvark on this one. I like seeing the highest rated articles there. You can look at the highest rated recent articles by just scanning the index. bob -- bob davis bobd at members.evolt.org From joshua at alphashop.net Sat Mar 31 14:07:33 2001 From: joshua at alphashop.net (Joshua OIson) Date: Sat Mar 31 14:07:33 2001 Subject: [thesite] new sidebars.. References: <200103311954.f2VJsMU01021@chmls06.mediaone.net> Message-ID: <019701c0ba37$85f27180$38360141@mrtnz1.ga.home.com> I suppose my idea would be a lot more useful if there were a lot more articles being posted where if you missed a month, you might've missed a lot. -joshua From: "Bob Davis" Subject: Re: [thesite] new sidebars.. > I'm with Aardvark on this one. > > I like seeing the highest rated articles there. You can look at the > highest rated recent articles by just scanning the index. > > bob From r937 at interlog.com Sat Mar 31 14:41:33 2001 From: r937 at interlog.com (rudy) Date: Sat Mar 31 14:41:33 2001 Subject: [thesite] Tip Harvester question (was: [***] Formatting tips ) Message-ID: <01c0ba22$fe459000$014f149a@rudy> > How exactly will this query find duplicate tips, especially when > quotes in emails tend to be lead by any number of line-begin > tokens? If this actually can work, I'd love to use it in a number > of applications, so believe me, I'm not trying to be argumentative. hi joshua you are right, there would be a difference between the tip harverster's processing of this tip -- Epsum factorial non deposit quid pro quo hic escorol. Olypian quarrels et gorilla congolium sic ad nauseum. and this one -- > >Epsum factorial non deposit quid pro quo hic escorol. >Olypian quarrels et gorilla congolium sic ad nauseum. > if the harvester just picks up the text, line feeds and quote tokens and all, then these two tips will end up having different bodies this is an area where the front end harvester logic must be very careful note that in order to attribute a tip to the appropriate author, it is important to detect when the tip is being quoted, and not include those instances my only point was that expecting "" and "\n:" and then try the GROUP BY again another strategy for detecting duplicates is to search for a particular known phrase e.g. select contentid, contentname, body where body like '%user%control%fonts%' which gets around the problem that line breaks and quote tokens might be between those words note that this method doesn't work in our case because you can't use LIKE on an oracle LONG datatype come to think of it, the GROUP BY might not work either :o( rudy From djc at starkmedia.com Sat Mar 31 17:22:00 2001 From: djc at starkmedia.com (Daniel J. Cody) Date: Sat Mar 31 17:22:00 2001 Subject: [thesite] March Stats for *.e.o Message-ID: <3AC66606.14EFF916@starkmedia.com> Hot & tasty: http://browsers.evolt.org/stats/stats.mar2001.html Couple quickie notes: 220,000 Page views 730,000 'Hits'(almost to that 1mil mark!) 60,000 Distinct IP address' 45.275 Gigabytes of data transferred.. 100,000+ referers from google(we should send them a couple t-shirts) If you have a m.e.o account, you should be able to grep your stats out of the report. Our most popular m.e.o member seems to be anarion's weblog: http://members.evolt.org/anarion/ About 1000 people from thelist are checking out the sites in that days "Headlines from evolt.org" email that gets sent daily..(grep the headlines directory) which is about 50% of people on thelist, very nice.. Search engines are finally starting to return a correct URL string in their search results, instead of the /index.cfm?menu=8&aid=123 like we used to have.. The 301 status code along with a 'location' header pointing to the correct URL should get those bots to update their engines.. Anyways, peruse at your own will :) .djc. From mwarden at odyssey-design.com Sat Mar 31 18:57:33 2001 From: mwarden at odyssey-design.com (Warden, Matt) Date: Sat Mar 31 18:57:33 2001 Subject: [thesite] Tip Harvester question (was: [***] Formatting tips ) References: <200103311828.LAA00953@alice.monkeyland.ca> Message-ID: <002201c0ba46$561d1f00$beae1b18@cinci.rr.com> > I know that in some languages, e.g., Perl and C, you can use bind > variables to speed up queries. Oracle can cache the SQL in its SGA > and the do a variable binding/substitution just before execution. In > Perl I've seen 50 times speed up on frequently executed queries. You mean like prepared/parameterized statements? select blah from thetips where evolt='sexy' and insertdate=? and whatever=? and then the above is sent to the database and, through the language, you can set those two parameters. Is that what you're talking about? Or are you talking about something different. If you are, then is that better than the prepared statement option? The query is run through the optimizer prior to batch loading/selecting/other so the database path is already figured out. Then, it doesn't have to go through the optimizer every time. > You may also want to turn off auto-commits. If you are commiting > after each query, Oracle needs to write the transaction to the > redologs each time. Of course, this means that your rollback segments > must be large enough to contain all of the uncommitted data. That's something I've never thought of. Can it be done through the language (not sure if that would be considered SQL). IOW, could you turn off autocommit for the duration of the loading and then turn them back on after you have committed everything and completed loading? -- mattwarden mattwarden.com From dsmah at home.com Sat Mar 31 21:51:44 2001 From: dsmah at home.com (Dean Mah) Date: Sat Mar 31 21:51:44 2001 Subject: [thesite] Tip Harvester question (was: [***] Formatting tips ) In-Reply-To: <002201c0ba46$561d1f00$beae1b18@cinci.rr.com> from "Warden, Matt" at Mar 31, 2001 07:54:40 PM Message-ID: <200104010356.UAA00884@alice.monkeyland.ca> Warden, Matt writes: > You mean like prepared/parameterized statements? > > select blah from thetips where evolt='sexy' and insertdate=? and whatever=? > > and then the above is sent to the database and, through the language, you can > set those two parameters. Is that what you're talking about? Or are you > talking about something different. If you are, then is that better than the > prepared statement option? The query is run through the optimizer prior to > batch loading/selecting/other so the database path is already figured out. > Then, it doesn't have to go through the optimizer every time. Yes, this is what I am talking about. You still need to prepare the query, however. Then bind and then execute. Oracle caches queries so that they can be used between sessions. So a Perl script that gets called repeatedly, e.g., a login script that queries the database for account information, can reap performance benefits because the queries are cached in Oracle's System Global Area (SGA). > That's something I've never thought of. Can it be done through the > language (not sure if that would be considered SQL). IOW, could you > turn off autocommit for the duration of the loading and then turn > them back on after you have committed everything and completed > loading? AutoCommit is usually set at the application level through your interface library and is set for the duration of execution. I don't think that the database sets this at all. Dean