[theforum] weo signoff

Jeff Howden jeff at jeffhowden.com
Thu Mar 3 21:19:33 CST 2005


Martin,

><><><><><><><><><><><><><><><><><><><><><><><><><><><><><
> From: Martin Burns
> 
> Cherry-picking somewhat here, for the rest, please do as
> repeatedly requested and file the bugs on the tracker at
> http://islands.easyweb.co.uk:8080/evolt_bugs/bugs/
> with enough detail (test cases helpful) to enable us to
> diagnose and fix any actual bugs.
><><><><><><><><><><><><><><><><><><><><><><><><><><><><><

I don't remember voting to use this really annoying bug tracker you've
chosen.

:p

><><><><><><><><><><><><><><><><><><><><><><><><><><><><><
> >  What about the highest rated articles feature?
> 
> Don't think we committed to doing that by launch.
><><><><><><><><><><><><><><><><><><><><><><><><><><><><><

Honestly, you didn't commit to much by launch except "getting off the
bespoke" solution.  Sorry, but some of the things *you're* willing to cut
are important to other members of the community.

><><><><><><><><><><><><><><><><><><><><><><><><><><><><><
> > What about validating to the specified DTD?
> 
> I see someone's changed something - other than
> user-submitted content (which we're not touching as a
> surprisingly large amount doesn't *currently* validate),
> there are only a couple of tweaks. Document it on the
> tracker please
><><><><><><><><><><><><><><><><><><><><><><><><><><><><><

You seem to like that thing so much, *you* document it.

><><><><><><><><><><><><><><><><><><><><><><><><><><><><><
> > Why is the article submission part of the site *still*
> > so *very* confusing?
> 
> I say po-tay-to.  You say po-tah-to.
><><><><><><><><><><><><><><><><><><><><><><><><><><><><><

Please provide documentation for the article submission process then so that
we have some material to work from when existing users come to us with
article submission questions.

><><><><><><><><><><><><><><><><><><><><><><><><><><><><><
> > Why does every article page crash IE6?
> 
> Doesn't on my laptop. Submit it with info about your
> version of XP and exact browser build.
><><><><><><><><><><><><><><><><><><><><><><><><><><><><><

Done.

><><><><><><><><><><><><><><><><><><><><><><><><><><><><><
> > Why do we have to show a pic for the author if one
> > doesn't exist?  Why not just *not* show a pic?
> 
> I say po-tay-to.  You say po-tah-to.
> Decisions made by the development team. I could have
> make exactly the same comment about a number of
> decisions made by those who worked on the existing
> site.  But hey, I didn't join in any of the work on
> that, so didn't.
><><><><><><><><><><><><><><><><><><><><><><><><><><><><><

It's surprising that existing site behavior wasn't taken into consideration.

><><><><><><><><><><><><><><><><><><><><><><><><><><><><><
> > Why is our RSS feed language setting set to en-GB?
> 
> Don't know. Why is the current site set to en-US? Ah,
> that's because in the current site, most of the
> developers had that as their primary language.
><><><><><><><><><><><><><><><><><><><><><><><><><><><><><

Or, perhaps because the bulk of the site's users are US-based.

><><><><><><><><><><><><><><><><><><><><><><><><><><><><><
> > Why can't commenting be turned off by Content for
> > certain articles?
> 
> It can, but it needs better labelling in the edit form.
><><><><><><><><><><><><><><><><><><><><><><><><><><><><><

Let's get it labeled better then.

><><><><><><><><><><><><><><><><><><><><><><><><><><><><><
> > What happened to being able to "Grrr!" a comment?
> 
> Functionality is there, with much more fine-grained
> control than the blunt 'grr' (and for the record, I
> always thought that label was ridiculous)
><><><><><><><><><><><><><><><><><><><><><><><><><><><><><

I find the term "spam" ridiculous as that doesn't adequately cover all the
possible reasons a comment is undesirable.

><><><><><><><><><><><><><><><><><><><><><><><><><><><><><
> Needs better definition of moderation status.
> 
> >  What is the "Helpful Comments" box for?
> 
> See, you got it.
><><><><><><><><><><><><><><><><><><><><><><><><><><><><><

Got what?  Sorry, I still don't know what the "Helpful Comments" box is for.

><><><><><><><><><><><><><><><><><><><><><><><><><><><><><
> > Why are many articles with code samples royally messed
> > up in the print version?  For that matter, why are
> > *so* many articles royally messed up in the print
> > version, period?!
> 
> Examples please. Documented in the tracker.
><><><><><><><><><><><><><><><><><><><><><><><><><><><><><

Done.

><><><><><><><><><><><><><><><><><><><><><><><><><><><><><
> But I'll put money it being the content filter. When we
> get the final data dump, they will all be imported
> without filtering, and it'll get turned back on for new
> content.
><><><><><><><><><><><><><><><><><><><><><><><><><><><><><

No, I've verified that the markup is correct in the actual data.  However,
it's messed up when pumped into the article template.

><><><><><><><><><><><><><><><><><><><><><><><><><><><><><
> > Why aren't article comments available in the
> > print version?
> 
> Because someone complained about us implementing bright
> ideas on the print template? That was one of the ones
> you killed, Jeff.
><><><><><><><><><><><><><><><><><><><><><><><><><><><><><

Newsflash!  The existing site *already* offers the ability to print comments
with the article in print mode.  It also prints the reader rating.
Optionally, either one or both can be turned off in print mode.

><><><><><><><><><><><><><><><><><><><><><><><><><><><><><
> For the record, printing the normal page works much
> better.
><><><><><><><><><><><><><><><><><><><><><><><><><><><><><

Sometimes, yes.  However, if you want the nicely annotated links from the
article, you're SOL if you're printing from the non-print template.

><><><><><><><><><><><><><><><><><><><><><><><><><><><><><
> > Can we not only display how long a user has been a
> > member, but also the date they joined (I personally
> > hate doing date-related math)?
> 
> Why yes, we can implement that feature request. When
> it's logged on the tracker.
><><><><><><><><><><><><><><><><><><><><><><><><><><><><><

It's a bug, not a feature request.  The existing site currently displays the
date a user joined, *not* how long they've been a member.  The least you
could do is duplicate existing functionality rather than making it up as you
go.

><><><><><><><><><><><><><><><><><><><><><><><><><><><><><
> > Why can't members of Content edit users?
> 
> Same reason they can't at present?
><><><><><><><><><><><><><><><><><><><><><><><><><><><><><

Who has "Full Admin" access then, as I clearly am only able to test a subset
of admin privileges?

><><><><><><><><><><><><><><><><><><><><><><><><><><><><><
> >  Can users be made inactive?
> 
> Yes. Also blocked from adding any new content.
> 
> > Are comments editable?  Can they be deleted?
> 
> Yes and Yes. Role-based.
><><><><><><><><><><><><><><><><><><><><><><><><><><><><><

As a member of Content, why am I unable to do these things then?

><><><><><><><><><><><><><><><><><><><><><><><><><><><><><
> > Why is UK English and Icelandic the only available
> > language options for the site?
> 
> Well that's more options than for the existing site,
> which doesn't even have the option.
><><><><><><><><><><><><><><><><><><><><><><><><><><><><><

If you're going to offer any, at least offer enough to cover 99% of our
users.  Does it do anything beyond changing the value of the "lang" and
"xml:lang" attributes of the <HTML> tag?  If not, then why are you making
such a big deal about it?  Any idea what % of our current user-base would
benefit from non en-us lang settings?

><><><><><><><><><><><><><><><><><><><><><><><><><><><><><
> > Can we *please* address the inline heading tag on
> > article pages (revisions, edit, preview, etc.)?
> 
> I say po-tay-to.  You say po-tah-to.
><><><><><><><><><><><><><><><><><><><><><><><><><><><><><

Actually, *I* say po-tay-to, and *you* say po-tah-to.  If you'd stop trying
to be an ass for 30 seconds you'd have noticed that.  I'm just trying to be
helpful and make sure we "go to market" with something that resembles a
finished product.  You, apparently, are more interested in being a dick to
me.

><><><><><><><><><><><><><><><><><><><><><><><><><><><><><
> > Why does a rather specific search like "javascript"
> > yield absolutely zero results?
> 
> Indexing will happen when we get the final data.
><><><><><><><><><><><><><><><><><><><><><><><><><><><><><

How exactly is the search feature supposed to be tested if it isn't indexed
prior to launch?

><><><><><><><><><><><><><><><><><><><><><><><><><><><><><
> > The date/time an article is published and the
> > categor(y|ies) it's published in should be below the
> > article title, not above.
> 
> Let me refer you to the agreed design.
><><><><><><><><><><><><><><><><><><><><><><><><><><><><><

I see that's how it is in the "agreed design", however if you view the page
without CSS or with a screenreader it'll be immediately apparent that this
aspect of the design is flawed and needs to be addressed.

><><><><><><><><><><><><><><><><><><><><><><><><><><><><><
> > Why do we have to manually map URLs to make old ones
> > work?  Why aren't we developing a system that
> > recognizes them automatically like we did last time
> > we changed the URL scheme?
> 
> Dupe, see above.
><><><><><><><><><><><><><><><><><><><><><><><><><><><><><

Not a dupe, a different question.  In order to handle existing URLs that are
currently able to retrieve content from WEO, a dynamic system must be put in
place.  As it will require a lot of testing, it is something that should be
built now rather than later or we'll suffer at the hands of the search
engines.

For example, the following should all bring up the same content:

/article/10_Steps_To_Higher_Search_Engine_Positioning/20/60390/index.html
/article/10_Steps_To_Higher_Search_Engine_Positioning/20/60390/
/article/rating/20/60390/
/article/foo/20/60390/
/node/60390/
/10-steps-to-higher-search-engine-positioning

Obviously, in order to support that, something dynamic (or regex-based) will
have to be built.  Simply relying on Drupal's rather clumsy URL aliases
won't get the job done.

Additionally, it'd be ideal (as there are still lots of these links out
there) to support the old style of links, which WEO currently supports.

/index.cfm?menu=8&cid=60390

><><><><><><><><><><><><><><><><><><><><><><><><><><><><><
> > Why do I have to delete "username" and "********" from
> > the login form everytime I login?
> 
> I thought you knew your WAI guidelines, sorry.
><><><><><><><><><><><><><><><><><><><><><><><><><><><><><

Apparently you don't know them as well as you think then.  If it were, you'd
know you're already failing WAI Checkpoint 12.4 which says that "form
controls and their labels must be explicitly associated with the LABEL
element".  I don't see any label for the password form control, nor do I see
a label for the search box adjacent to the login form.

However, the guideline you're suggesting you're following is WAI Checkpoint
10.4 which says to "include default, place-holding characters in edit boxes
and text areas".  However, it does not say you must use the value to further
label the form control.  The rational is not to label the form control with
the default value, but to aid in some devices that will ordinarily skip form
controls that are empty, a rather retarded model, if you ask me.
Interestingly, Bobby is just fine with a value of " " (a single space or
Chr(32)) for the form controls and will pass you on WAI Checkpoint 10.4
without issue.  In fact, refer to Techniques for WCAG2 which actually states
that "This is a negative technique, an example of something you shouldn't
do" [1]. The test case has also been rejected [2] and erratum proposed [3]
for the WCAG Errata [4].

[1] http://www.w3.org/TR/WCAG20-HTML-TECHS/#emptycontrol
[2] http://www.w3.org/WAI/GL/WCAG20/tests/test63.html
[3] http://lists.w3.org/Archives/Public/w3c-wai-gl/2003AprJun/0216.html
[4] http://www.w3.org/WAI/GL/WAI-WEBCONTENT-ERRATA

Regards,

Jeff



More information about the theforum mailing list