Java linux HTML qmail C++ PHP Debian MySQL ASP JavaScript perl Delphi NetBSD Solaris etc etc etc [...]






Page: 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18


Comments:
<webben> (except in so far as UAs don't recognize roles)
<Windrose> webben: in a way, but while it would be difficult to get that construct by WCAG 1.0, it's a breeze to get it by WCAG 2.0
<JibberJim> Yet people haven't got any more able in between
<webben> Hmm ... but accessibility guidelines are not the same as accessibility technologies. The technologies can be good even when the guidelines are rubbish (in theory).
<JibberJim> the WHAT discussion was good in that someone said "I've got some really accessible new content" and the image didn't have an f'ing alt
<webben> That was rather funny.
<JibberJim> and it needed one!
<webben> The MDC Accessible DHTML page doesn't help by making its first example one of reimplementing a checkbox with span.
<webben> http://developer.mozilla.org/en/docs/Accessible_DHTML
<Windrose> Sounds like an excellent discussion, on several levels ... tho possibly not the ones they considered.
<webben> It seems rather important however than Jaws and Window-Eyes are supporting these roles, given that assistive technology often lags even further behind specs than browsers.
<JibberJim> Indeed
<Windrose> webben: it would be, if it wasn't for the fact that neither of them will support user-extended roles ... and that other UAs still won't.
<JibberJim> But not if they need the roles but not an ALT - which is what happens here
<webben> Windrose: If we forget about WCAG 2.0 which from reports sounds useless anyhow ... then the fact that neither will support user-extensions may turn out to be a blessing.
<webben> It means that there would be a reliable subset.



<Windrose> webben: yes. And an unreliable superset. From history, which do you think people will use?
<webben> Windrose: I think people have tended to use unreliable stuff when they can rely on it in the primary user agent (e.g. IE autocomplete).
<webben> Windrose: If Jaws and IE aren't offering it, then the unreliable stuff is unlikely to be used anyway.
<Windrose> Indeed. Tho I must applaud the W3C for finding a way to make people happy.
<webben> It's also noteworthy that the roles extensibility is also nowhere near as simple as just inventing your own role.
<webben> I haven't seen many people keen to delve into RDF. http://www.w3.org/WAI/PF/GUI/roleTaxonomy-20060508.html#Extending
<Windrose> webben: it does make it a wee bit more complicated, yes.
<webben> though I suppose the issue might be that UAs would create RDF which others could use
<nphase> okay im going to restate my question to see if this time someone will help me
<nphase> http://www.bpong.com/store/prod.php?prodid=2 JS/ajax updating of the image works in FF (change the color from black to white and back and forth), but not in IE.. i added a rand() at the end of the image to show that its updating the image div's inners. in IE everything else updates fine though. any ideas?
<webben> How workable those extensions are depends on how completely the behaviour of widget can be defined using RDF.
<Windrose> webben: well ... rather how to make sense of the roles except as simple string labels for attaching styles and scripting.
<hax> ajnewbold: one of my couses is "Internet Applications"
<hax> *courses
<ajnewbold> haha
<hax> ajnewbold: i swear i'm gonna lose it if i have to use frontpage
<ajnewbold> "Internet Explorer"
<ajnewbold> "Microsoft Outlook"
<hax> ajnewbold: clearly worth the $8k/semster it probably is gonna cost
<ajnewbold> I have never used frontpage
<webben> It's not inconceivable that you could create a syntax that would allow proper description of widget behaviour.
<ajnewbold> is it that bad?
<defbyte> hi, is there an event that tells me if the window size has changed ??
<hax> ajnewbold: its that bad
<ajnewbold> hax: it is better or worse than just using Word to make your web pages? :)
<webben> In any case, I suppose there's nothing prevent WHATWG specs from limiting roles to those included in the spec.
<hax> ajnewbold: worse
<ajnewbold> hax: oh, heh, then it must really blow
<hax> ajnewbold: word atleast recognizes its not a web app
<webben> s/in the spec/in the W3C spec
<Windrose> WHATWG, aren't they the lot behind the 'ping' attribute too?
<webben> Windrose: that's in their web apps spec. Which is decidely less advanced than their Web Forms 2.0 spec, which seems to be transitioning to W3C.
<JibberJim> "the lot"
<JibberJim> ITYM Hixie
<Windrose> "Aye. 'im."
<krisp> can anyone point in the direction of some really nice advertising agency sites?
<webben> Does the potential problem with extensibility make the role technology less or more desirable than 1) standardized delimited class names (which is how roles will be implemented in html4 anyhow) or 2) XBL2
<Windrose> webben: frankly I don't see a need for either roles or class names - I would rather see HTML extended with more elements.
<Windrose> Same problem, of course, with UA support and understanding.
<webben> well, that's my option 1) ... and i guess it's what Web Forms 2.0 does to some extent.
<webben> e.g input type="range" etc
<webben> (which could be _displayed_ as a slider for example)
<Windrose> "more of them".
<webben> Windrose: isn't that typographic stuff coming in CSS3? (I realize that's like waiting for Godot).
<webben> Windrose: if you had to add one more semantic element to the HTML4 set, what would it be?
<GwaiLo> Windrose: *hugs* hello :)
<Fenix|work> Greetings :)
<Fenix|work> Can someone educate me on when one should use XML and when one shouldn't?
<Windrose> ... but CSS 3 has also - last I looked - made an effort to make itself an ass - and I mean the donkey kind. Specifically the word-wrap property"
<Windrose> Fenix|work: that's a very openended question. When you have a need to create a markup language of some sort, and doesn't need the full power of SGML.
<webben> Fenix|work: in what context?
<Fenix|work> Windrose, it's open ended, cus I'm just learning. :)
<Fenix|work> webben, that... I'm not sure...
<webben> Fenix|work: are we talking ajax here? web pages? internal data formats?
<Fenix|work> but I'll come up with some examples.
<Fenix|work> and yes, web pages
<Fenix|work> just plain php... no ajax
<Fenix|work> I understand that xml's purpose is to carry data
<Fenix|work> and it complements html ... where you can load in the data into html and display it



<Fenix|work> but how you use xml is solely up to the designer/developer
<Windrose> Fenix|work: not really. XML is a meta-language with which you create other markup languages such as XHTML.
<Fenix|work> ok
<webben> Fenix|work: so what we're talking about here is having an xml file with data (rather than a database)
<webben> and construct (x)html web pages with that data
<Fenix|work> why would you use an xml file instead of a database?
<Fenix|work> (for a webpage)?
<webben> Fenix|work: hmm, depends partly on what's easiest
<webben> Fenix|work: for instance it may prove easy to have an XML file which is "transformed"/"converted" quite directly into (X)HTML for the web
<Fenix|work> I can understand speed in the sense that opening up a data connection to a database, then querying, etc... using xml would be faster for a small amount of data
<webben> Fenix|work: also XML is more versatile, in that it can actually be used as a medium of data exchange
<Fenix|work> like with RSS feeds
<webben> Fenix|work: yeah
<webben> Fenix|work: a lot of Ajax using XML too
<Fenix|work> yeah, I saw
<W_> speed? _very_ few sites on the internet have to worry about cpu resources
<webben> Fenix|work: usually that XML is created out of a database, but sometimes it may make sense to just use XML as your backing data store
<mattmcc> Actually it's amusing how often ajax _doesn't_ use XML. Because it's overkill compared to, say, JSON.
<webben> For instance, I'm currently working on an XHTML FAQ which would use a custom XML format for FAQ.
<webben> That would be trivial to transform to XHTML or HTML.
<Fenix|work> ok... webben in terms I fully understand... how would one use XML and MySQL for a blog?
<Windrose> webben: you are using FaqML?
<Fenix|work> right now, everything I've ever done is PHP and MySQL
<W_> IMO the greatest benefit with XML (or good databases for that matter) is that you get validation for free
<webben> Windrose: no... i should look into that, thanks .... my custom format is just for my own use, really
<Windrose> Well ... one advantage of XML - and I don't see too many - is that there exist a whole lot of languages already defined.
<webben> Fenix|work: if you're using MySQL to store your blog entries, then you'd be creating XML, HTML, XHTML etc from that data
<mattmcc> W_: XML gives you well-formedness for free, but not validation. That requires the additional work of something like an XML Schema.
<Fenix|work> to use to make my blog available for data exchange with programs that understand xml
<W_> mattmcc, I consider that part of the benefit of using XML
<Fenix|work> like an RSS newsfeed
<W_> and it's certainly low-cost if not free
<mattmcc> W_: Heh. Have you ever written an XSD? :)
<W_> always
<W_> just pure XML doesn't make much sense to me; you have to code up against a schema
<mattmcc> Nevermind that lots of XML parsers don't actually support it.
<webben> Fenix|work: but you wouldn't want to store your data in both MySQL and XML
<Fenix|work> no... you'd use the database to create xml
<Fenix|work> to then transfer
<W_> My usual use of XML goes something like: Create schema->compile to java classes->code rest of app
<webben> W_: depends, writing my little FAQ in a custom a format even sans schema is massively easier than writing it straight into XHTML
<mattmcc> Fenix|work: If your content is in a DB already, why introduce the extra step of converting it to XML, only to convert to other formats?
<webben> W_: because it allows me to focus on relevant markup
<Windrose> webben: sorry, missed your earlier question - <menu>
<W_> webben, well I'm a fan of enforced consistency
<webben> W_: oh, i agree a schema is a good thing :)
<Fenix|work> mattmcc, the only reason I can think of is to prepare it for RSS
<W_> if you have a schema when you start, you can't make a mistake in your document, use the wrong tag or something
<Windrose> ... the element W3C took /out/ of HTML because, quote "it renders just like an UL".
<webben> Windrose: a la XHTML2's navigation lists? or different?
<mattmcc> Fenix|work: Generating RSS from a DB is just like generating HTML from a DB.
<W_> lol Windrose
<W_> is that for real?
<VolVE> goddamn crazy spam
<Windrose> W_: I wish it wasn't.
<Fenix|work> mattmcc, I've never done it before... perhaps a better example is to create an xml sitemap from content in the db :)
<Windrose> W_: http://www.w3.org/TR/html401/struct/lists.html#edef-MENU
<Windrose> "Both elements have the same structure as UL, just different rendering ... "
<W_> heh
<Windrose> "We strongly recommend using UL instead of these elements." - the pain.
<W_> perhaps we should stop using <p> tags as well, go back to separating paragraphs with <br>
<mattmcc> Windrose: Took out <menu>, then in XHTML 2 put in <nl>? :)
<Zeros> nah, use <p>&nbsp;</p> to separate blocks of text like DW does!
<Windrose> mattmcc: yeah.
<danor_> It was a long time since I built a homepage, what languages to use?
<mattmcc> Um. HTML?
<danor_> hehe ya, beside that ;)
<Windrose> danor_: ... CSS?
<mattmcc> English?
<Fenix|work> webben, so basically... xml is best used to store data like in a database, but without the database :)
<hax> brb, going to visit grandma
<danor_> "languages" ya, css but is there a better lnguage than php I should choose?
<webben> Fenix|work: sort of, or for converting stuff from the database for data exchange


Name:

Comments:

Please enter the result of the sum 63 + 46 (to avoid spam):






Return to #web
or
Go to some related logs:

erland debian package
python like conch
no cd boot pxe
debian
css
python
debian amd64 codeblocks
ubuntu terminal background color
suse100.i686 iso
OWM premature end of script headers