| 00:24 | <[SMQ]Mrpsycholo> | hello |
| 00:24 | <[SMQ]Mrpsycholo> | any major differences from xhtml? will there be compatibility issues or differences? |
| 00:25 | jcranmer | snickers |
| 00:25 | <jcranmer> | http://wiki.whatwg.org/wiki/HTML_vs._XHTML |
| 00:28 | <[SMQ]Mrpsycholo> | so document.write can be used in regular without using script in html5? |
| 00:29 | <[SMQ]Mrpsycholo> | will there be compatibility issues with html5 or will xhtml be more compatible across devices or browsers? |
| 00:29 | <[SMQ]Mrpsycholo> | safari recognized <audio> tag but firefox and ie8 didnt |
| 00:29 | <Hixie> | document.write will work the same in html5 as it has in the past |
| 00:29 | <jcranmer> | HTML is designed, in part, to represent current browser semantics |
| 00:30 | <Hixie> | in fact html5 in general is supposed to work the same as in the past, yeah |
| 00:30 | <jcranmer> | besides |
| 00:30 | <jcranmer> | FF nightlies have <audio> support |
| 00:30 | <Hixie> | so html5 will likely be more compatible than xhtml in general |
| 00:30 | <Hixie> | (IE doesn't do XHTML, e.g.) |
| 00:30 | <[SMQ]Mrpsycholo> | i made my site in xhtml |
| 00:30 | <[SMQ]Mrpsycholo> | becuase i was told it was a standard based form |
| 00:31 | <[SMQ]Mrpsycholo> | so i decided to use it and keep strict validation |
| 00:31 | <[SMQ]Mrpsycholo> | as a noob though, but i was udner impression xhtml was the future |
| 00:32 | <Hixie> | it was, for a while |
| 00:32 | <Hixie> | then it wasn't |
| 00:32 | <[SMQ]Mrpsycholo> | html5 will be lowercase? |
| 00:32 | <[SMQ]Mrpsycholo> | will it have a validation feature like xhtml? |
| 00:32 | <Hixie> | html5's tags are case-insensitive |
| 00:32 | <[SMQ]Mrpsycholo> | so they have to be uppercase? |
| 00:32 | <[SMQ]Mrpsycholo> | i like lowercase xhtml |
| 00:32 | <[SMQ]Mrpsycholo> | lol |
| 00:32 | <Hixie> | http://html5.validator.nu/ |
| 00:32 | <Hixie> | no, they can be whatever case you prefer |
| 00:32 | <Hixie> | lowercase, uppercase, both, whatever |
| 00:33 | <[SMQ]Mrpsycholo> | will html5 require all tags to be closed like xhtml? |
| 00:33 | <[SMQ]Mrpsycholo> | <br /> |
| 00:33 | <Hixie> | no, but it allows it |
| 00:33 | <jcranmer> | <A hReF="#asdf"> |
| 00:33 | <Hixie> | <br> and <br/> are equivalent in html5 |
| 00:33 | <[SMQ]Mrpsycholo> | nice its good you can use both :) |
| 00:34 | <[SMQ]Mrpsycholo> | and in html5 y ou dont have to specify type for script? |
| 00:34 | <Hixie> | nope |
| 00:34 | <Hixie> | <script> is fine |
| 00:34 | <[SMQ]Mrpsycholo> | of course css is still good |
| 00:35 | <[SMQ]Mrpsycholo> | www.q4supremacymod.com is my site i am using as a learning tool :P |
| 00:36 | <[SMQ]Mrpsycholo> | so in html5 y ou wont have to use object tag? for video? |
| 00:38 | <Hixie> | for video you'll be able to use <video> |
| 00:38 | <[SMQ]Mrpsycholo> | nice |
| 00:39 | <[SMQ]Mrpsycholo> | came across this from youtube saw a demo on html5 |
| 00:39 | <Hixie> | cool |
| 00:40 | <Hixie> | is that the "features you desperately want but can't yet use" video? |
| 00:40 | <[SMQ]Mrpsycholo> | yeah |
| 00:40 | <Hixie> | i'm the guy giving that presentation :-) |
| 00:40 | <[SMQ]Mrpsycholo> | the car with the laptop on the hood :) |
| 00:40 | <[SMQ]Mrpsycholo> | nice |
| 00:41 | <[SMQ]Mrpsycholo> | i will put a link on my site to www.whatwg.org :) |
| 00:41 | <Hixie> | :-) |
| 00:41 | <[SMQ]Mrpsycholo> | my site is a community for quakers and doomers |
| 00:41 | <[SMQ]Mrpsycholo> | :) |
| 00:42 | <jcranmer> | and my site is an intranet where my latest addition was a theming style that derives from an april fools' day prank derived from google |
| 00:42 | <[SMQ]Mrpsycholo> | i dont think i saw that april fools one |
| 00:43 | <jcranmer> | well, it's an intranet |
| 00:43 | <jcranmer> | I don't think you'd see intranet april fools' day pranks |
| 00:43 | <[SMQ]Mrpsycholo> | ahh |
| 00:44 | <[SMQ]Mrpsycholo> | intranet being your personal local net not the global one? |
| 00:44 | <[SMQ]Mrpsycholo> | i should have went and studied computer science instead of psychology |
| 00:44 | <[SMQ]Mrpsycholo> | but my eye did catch intra vs inter |
| 00:44 | <jcranmer> | intranet = intra-net; intra- = within |
| 00:45 | <[SMQ]Mrpsycholo> | yeah latin idea |
| 00:45 | <[SMQ]Mrpsycholo> | endocrine exocrine |
| 00:45 | <[SMQ]Mrpsycholo> | internal secrete versus external |
| 00:48 | <[SMQ]Mrpsycholo> | so for <video src="firefox.ogg" controls autoplay></video> |
| 00:48 | <[SMQ]Mrpsycholo> | you dont need to separate with commas? |
| 00:48 | <[SMQ]Mrpsycholo> | controls and autoplay you can list them after the src? |
| 00:49 | <Hixie> | yup |
| 00:49 | <Hixie> | it's like with <input type=text value="test" readonly disabled> |
| 00:50 | <[SMQ]Mrpsycholo> | and some habits i have with xhtml i can still keep in html5 |
| 00:51 | <[SMQ]Mrpsycholo> | its good i dont have to have the doctype in 5 |
| 00:51 | <Hixie> | yeah just the short one |
| 00:51 | <Hixie> | <!DOCTYPE HTML> |
| 00:51 | <Hixie> | it's actually possible to memorise it! |
| 00:52 | <[SMQ]Mrpsycholo> | hehehe yeah |
| 00:52 | <[SMQ]Mrpsycholo> | <!DOCTYPE html PUBLIC "-//W3C//DTD XHTML 1.0 Strict//EN" |
| 00:52 | <[SMQ]Mrpsycholo> | "http://www.w3.org/TR/xhtml1/DTD/xhtml1-strict.dtd">; |
| 00:52 | <[SMQ]Mrpsycholo> | <html xmlns="http://www.w3.org/1999/xhtml"; xml:lang="en" lang="en"> |
| 00:52 | <[SMQ]Mrpsycholo> | is a bit of a pain lol |
| 00:53 | <[SMQ]Mrpsycholo> | i havent memorized it yet |
| 00:54 | <Rik`> | Hixie: is the lang attribute required on <html> ? |
| 00:54 | <Rik`> | oh, if i remember, <html> is not even required |
| 00:54 | <Hixie> | lang="" isn't required, but it's a good idea to put it in to declare the language, that way things like :lang() in CSS work |
| 00:55 | <Rik`> | or assistive technology can use it |
| 00:57 | <Hixie> | indeed |
| 00:57 | <[SMQ]Mrpsycholo> | php will be usable to? i am looking at the file open and save |
| 00:57 | <Rik`> | I'd like to see it required but it's maybe more a WCAG concern |
| 00:57 | <Hixie> | [SMQ]Mrpsycholo: html5 doesn't really affect php |
| 00:58 | <[SMQ]Mrpsycholo> | cause i wanted to have users to be able to upload files |
| 00:58 | <[SMQ]Mrpsycholo> | but my site doesnt do php with windows i guess |
| 00:58 | <Hixie> | Rik`: i don't think in practice there's a big enough disadvantage to not including it to warrant requiring it |
| 00:58 | <[SMQ]Mrpsycholo> | people in the community who have content could upload the content is what i wanted to do originally |
| 01:01 | <[SMQ]Mrpsycholo> | will html ever use // for comment lines? or /* */ ? |
| 01:02 | <[SMQ]Mrpsycholo> | that would interfere wouldnt it with what could be on the page? |
| 01:02 | <Rik`> | Hixie: accessibility concerns ? good practices ? but like I said, it's maybe not the role of html5 |
| 01:03 | <Hixie> | Rik`: in practice, ATs don't need the language, they'll assume the user's language, which will almost always be right |
| 01:03 | <Hixie> | [SMQ]Mrpsycholo: too many pages already include those for us to do that i think |
| 01:04 | <[SMQ]Mrpsycholo> | do you guys have an icon? |
| 01:04 | <Rik`> | "almost always" ? i go to french and english websites, in the same proportions |
| 01:04 | <[SMQ]Mrpsycholo> | when i put the link on the page i could put an icon |
| 01:05 | <Hixie> | [SMQ]Mrpsycholo: http://whatwg.org/images/ |
| 01:05 | <Rik`> | so if i used ATs, half the time it would be wrong |
| 01:06 | <Hixie> | Rik`: true. though in those cases, basic heuristics could guess the language anyway. |
| 01:07 | <Rik`> | I don't know if they have this mechanism, I believe not from the last demo I had |
| 01:09 | <Rik`> | has this subject been discussed already ? |
| 01:12 | <Hixie> | Rik`: which, requiring language declarations? |
| 01:12 | <Rik`> | yep |
| 01:13 | <[SMQ]Mrpsycholo> | i appreciate the help i will defintiely read more on that demo site |
| 01:14 | <[SMQ]Mrpsycholo> | its good you also provided the text too :) |
| 01:15 | <[SMQ]Mrpsycholo> | any of you guys play quake? lol |
| 01:19 | <[SMQ]Mrpsycholo> | when you guys need a break from coding play this :P http://www.youtube.com/watch?v=2hqWgRx7QgM :) |
| 01:25 | <Hixie> | Rik`: don't think so |
| 01:29 | <[SMQ]Mrpsycholo> | quakelive is the new version ported into a browser actually |
| 01:29 | <[SMQ]Mrpsycholo> | i was impressed with how unreal tournament used web for control of a server in unreal tournament 3 |
| 04:15 | <[SMQ]Mrpsycholo> | is it possible to make your own tags yet? |
| 04:16 | <[SMQ]Mrpsycholo> | document.getElementByTagName('div') could you possibly define your own? |
| 04:23 | <Hixie> | [SMQ]Mrpsycholo: to make up your own tag, you pick a tag that is closest to what you want your tag to mean, and then you put your tag's name in the "class" attribute |
| 05:32 | <[SMQ]Mrpsycholo> | oh yeah that makes sense :) |
| 05:33 | <[SMQ]Mrpsycholo> | under type in css you would define it as a block right? |
| 05:33 | <Hixie> | depends what kind of presentation you want, but yes |
| 05:34 | <[SMQ]Mrpsycholo> | the most difficulty i have had so far is the positioning |
| 05:34 | <[SMQ]Mrpsycholo> | thats why i went with frameset |
| 05:35 | <[SMQ]Mrpsycholo> | trying that layout wiht css without using frames was such a pain for me |
| 05:35 | <[SMQ]Mrpsycholo> | it was easier just to define columsand rows |
| 05:35 | <[SMQ]Mrpsycholo> | make several pages to accomplish it |
| 05:46 | <kanchan_Tripathi> | Hello to all can anybody please tell me how to post a form value using telnet |
| 05:46 | <kanchan_Tripathi> | ?? |
| 05:59 | <kanchan_Tripathi> | Hello to all can anybody please tell me how to post a form value using telnet?? |
| 07:45 | <hsivonen> | I had dinner with friends who make software |
| 07:45 | <hsivonen> | two of them have done XML signatures |
| 07:45 | <hsivonen> | and said that one should avoid XML signatures if one wants interop |
| 07:48 | <MikeSmith> | hsivonen: why is that? because not enough implementations handle XML signatures, or because they don't handle them interoperably? |
| 07:51 | <hsivonen> | MikeSmith: because XML signatures depend on canonicalized serialization of XML and everyone has bugs in the canonicalizing serializer |
| 07:51 | <MikeSmith> | ah |
| 07:52 | <MikeSmith> | yeah, that seems like a weak design |
| 07:53 | <MikeSmith> | I wonder how many of the bugs are due to ambiguities in the Canonicalization spec? |
| 07:54 | <hsivonen> | as I understand it, it's mainly that UTF-8 is tough, Namespaces are tough and string interning is tough |
| 07:54 | <hsivonen> | all in all, serializing XML is tough |
| 07:57 | <MikeSmith> | so in the case of the Widgets work, it really doesn't seem like it's necessary to use XML signatures if something less complex and more robust will work instead |
| 07:58 | <hsivonen> | seems that way to me |
| 07:59 | <MikeSmith> | hsivonen: btw, I notice that the xhtml5core.rnc whattf driver doesn't toggle off the <mark> element the way that the html5core.rnc driver does |
| 07:59 | <MikeSmith> | do you recall of that's by design for some reason? |
| 08:00 | <hsivonen> | MikeSmith: seems like a bug |
| 08:00 | <MikeSmith> | OK |
| 08:01 | <MikeSmith> | I'll try to fix it in the patch I'm working on for generating the driver files |
| 08:14 | <MikeSmith> | hsivonen: I notice also that xhtml5core.rnc has "v5only = notAllowed" but html5core.rnc does not |
| 08:16 | <MikeSmith> | hsivonen: ah, I see that v5only is, among other things, used to toggle the <mark> element |
| 08:18 | <MikeSmith> | so seems like the 'include "phrase.rnc" { mark.elem = empty }' part is maybe not necessary |
| 08:18 | <MikeSmith> | if html5core.rnc uses "v5only = notAllowed" instead too |
| 08:19 | <hsivonen> | MikeSmith: yeah, the bug is probably the other way round than first thought |
| 08:20 | <MikeSmith> | hsivonen: OK |
| 08:47 | <hsivonen> | annevk2: why do CSS namespaces allow @namespace url(http://www.w3.org/1999/xhtml); ? |
| 08:47 | <hsivonen> | annevk2: after all, the NS "URI" is a string and not an address |
| 08:47 | <hsivonen> | (probably too late to change now. I'm just curious.) |
| 08:50 | <annevk2> | because it's too late to change now |
| 08:51 | <annevk2> | CSS namespaces is from 99 |
| 08:51 | <hsivonen> | annevk2: ok |
| 08:53 | <hsivonen> | http://www.w3.org/2005/Incubator/socialweb/charter |
| 08:53 | <hsivonen> | how come W3C Widgets are now a social Web topic? |
| 08:54 | <MikeSmith> | hsivonen: no clue from me |
| 08:54 | <MikeSmith> | I guess I should actually take time to read charters |
| 08:56 | <MikeSmith> | "The scope includes issues such as widget platforms (such as OpenSocial, Facebook and W3C Widgets).." |
| 08:58 | <MikeSmith> | I didn't even know that widget platforms were an "issue", let alone one related to the idea of social Web |
| 08:58 | <hsivonen> | I guess the kind of widgets that are embedded on Web pages are commonly used on Social Media Web 2.0 sites |
| 08:59 | <hsivonen> | but W3C Widgets aren't that kind of widgets |
| 09:11 | <jgraham> | There's discussion going on of ecmascript versioning on es-discuss, for those not following along at home |
| 09:18 | hsivonen | predicts that out of band version won't be adopted by authors. |
| 09:33 | hendry | discovers http://syntax.whattf.org/relaxng/ |
| 09:34 | <MikeSmith> | hendry: don't get too used to that |
| 09:34 | <MikeSmith> | the schema drivers are likely going to disappear |
| 09:34 | <hendry> | MikeSmith: drivers? |
| 09:35 | <MikeSmith> | html5full.rnc and others |
| 09:35 | <MikeSmith> | those drivers are really intended to be used within the context of validator.nu |
| 09:36 | <MikeSmith> | and the legacy.rnc file in particular is intended to be used within the validator.nu context and not for schema-only checking |
| 09:36 | <hendry> | just thinking it would be nice to validate html5 locally with rnv and one of these schemas. some Debian Web apps (ikiwiki) want to validate generated HTML offline and rnv would be lightweight. validator.nu is too heavy to get running on a buld machine. |
| 09:37 | zcorpan | notes that 'text-shadow' does not support the effect used at http://syntax.whattf.org/ |
| 09:37 | <MikeSmith> | hendry: yeah, I can understand that |
| 09:37 | <hsivonen> | hendry: if you come up with a regexp hack to replace the proper datatypes with rnv-compatible approximations, please share it |
| 09:37 | <hsivonen> | hendry: does rnv support XSD regexps? |
| 09:38 | <zcorpan> | oh the shadow has different text |
| 09:38 | <MikeSmith> | I don't think it does |
| 09:38 | <hendry> | XSD? i thought we don't use XSD :) |
| 09:38 | <MikeSmith> | I mean I don't think RNV supports pluggable datatypes |
| 09:39 | <MikeSmith> | hendry: it's debatable whether we want to be doing anything to facilitate schema-only validation outside of the context of a complete conformance checker |
| 09:41 | <hsivonen> | hendry: we don't use XSD. However, the XSD regexps aren't all bad, although they have surprising quirks. |
| 09:41 | <hendry> | i prefer a 'less is more' approach. or i want a non-complete lightweight conformance checker. |
| 09:42 | <hsivonen> | hendry: so if you don't port the HTML5 datatype library for rnv and rnv supports XSD regexps, you are better off using regexps than no attribute value checking at all |
| 09:43 | <hsivonen> | hendry: I'd expect it to be feasible to run a perl or python script over the schemas so that the proper datatypes are trashed and the commented regexps uncommented |
| 09:44 | <MikeSmith> | hendry: outside of the datatype issue, you'll still be able to do that kind of checking with the schemas, by taking the schema modules and writing your own drivers |
| 09:47 | <annevk2> | One problem with ECMAScript versions is that Mozilla uses its own versioning stuff for JavaScript. E.g. JavaScript 1.7. (Not too different from UA=IE8...) The other of course that the specification does not actually document how to react to different version identifiers and such. |
| 09:47 | <hendry> | MikeSmith: sorry, bit confused by terminology; so a driver = .rnc = RelaxNG Schema ? |
| 09:48 | <hsivonen> | hendry: driver as in XHTML2 WG Modularization speak :-) |
| 09:48 | <hsivonen> | hendry: a schema that just imports other schema files |
| 09:49 | <hendry> | hsivonen: and i'm confused about you talk of regexps. were you saying you could not to use rnv and use regexps to validate? |
| 09:49 | <MikeSmith> | hendry: the drivers are the files there that don't define any content models themselves, but just import the other schema files that have the actual definitions |
| 09:49 | <hendry> | MikeSmith: gotcha |
| 09:50 | <MikeSmith> | http://www.davidashen.net/rnv.html says: |
| 09:50 | <MikeSmith> | "Besides built-in support for XML Schema datatypes, RNV provides two different ways to add user-defined datatype libraries." |
| 09:51 | <hsivonen> | hendry: I'm saying that it'll be easier for you to approximate the HTML5 datatype with XSD regexps than to write a an implementation of the HTML5 Datatype Library in such a way that it can be used with RNV |
| 09:52 | <annevk2> | you in the US hsivonen? |
| 09:53 | <hsivonen> | annevk2: not yet |
| 09:53 | <hsivonen> | annevk2: flying on Saturday |
| 09:53 | <hendry> | hsivonen: ok, i grok .rncs, though i have no real exp with 'XSD regexps'. do you have XSD examples i could peer at? and what tool uses them? xmlstarlet? |
| 09:53 | <hendry> | hsivonen: San Francisco? :-) |
| 09:53 | <hsivonen> | hendry: there are commented-out examples in common.rnc |
| 09:53 | <hsivonen> | hendry: first Mountain View, then Alexandria, VA |
| 09:56 | <hendry> | MikeSmith: so you think rnv can support XDF with an external command or something? what tool grok XSD regexps then? |
| 09:56 | <hendry> | s/XDF/XSD/ |
| 09:56 | <hendry> | hsivonen: what's in Alexandria? |
| 09:57 | <MikeSmith> | hendry: to use that feature in rnv, I think you have to roll the external command yourself |
| 09:57 | <MikeSmith> | or use RNV's alternate support for expressing the datatype-checking in scheme |
| 09:57 | <MikeSmith> | if you know scheme or lisp already, that might be a good way to go |
| 09:58 | <MikeSmith> | or a useful project for somebody who does know lisp or scheme well |
| 09:58 | <hsivonen> | hendry: friend's wedding |
| 10:01 | <hendry> | MikeSmith: i don't know lisp well. i didn't like my dr scheme class and University. or emacs. drat. |
| 10:01 | <hendry> | s/and/at :-) University was great |
| 10:02 | <MikeSmith> | hendry: anyway, there are some examples and more info in the README that's distributed with rnv |
| 10:04 | <hendry> | MikeSmith: yup. I see them. |
| 10:41 | <hsivonen> | Philip`: I was able to find and fix the bug you reported about parsing http://www.giftology.co.uk |
| 10:42 | <hsivonen> | Philip`: I'm unable to reproduce the bugs about http://www.villatraining.ca/ and http://www.autobanga.lt/ |
| 11:20 | <hsivonen> | Philip`: do you have additional clues on how to track down http://krijnhoetmer.nl/irc-logs/whatwg/20090401#l-410 ? |
| 11:30 | <hsivonen> | "Use aria-label only if the interface is such that it is not possible to have a visible label on the screen" |
| 11:30 | <hsivonen> | I wonder how strict "possible" is there |
| 11:33 | hsivonen | wonders how aria-label use cases differ from title attribute use cases |
| 11:34 | zcorpan | wonders why one can't use css positioning to hide labels off-screen |
| 11:35 | <zcorpan> | hsivonen: have you had a look at http://philip.html5.org/data/table-implicitly-closed-p.txt ? |
| 11:36 | <hsivonen> | zcorpan: only took a look at a tiny sample so far |
| 11:36 | <hsivonen> | zcorpan: tiny as in 10 |
| 11:37 | <zcorpan> | hsivonen: maybe it's possible to set up a script that takes a screenshot using the old moz parser and using the html5 parser and then comparing the screenshots |
| 11:43 | <hsivonen> | zcorpan: I guess the reftest framework could be changed to do something like that |
| 11:44 | <hsivonen> | why did ARIA hois aria-invalid to a global state??? |
| 11:44 | <hsivonen> | *hoist |
| 11:44 | <hsivonen> | and required |
| 12:22 | <Philip`> | hsivonen: Hmm, that might just be because I was using a really old version |
| 12:22 | <Philip`> | (1.0.7 apparently) |
| 12:22 | <hsivonen> | Philip`: oh. I thought you were using my special Philip` build |
| 12:23 | <Philip`> | hsivonen: Oh, maybe I was |
| 12:24 | <Philip`> | hsivonen: I can try testing it some time later, if I don't forget |
| 12:34 | <hsivonen> | Philip`: thanks |
| 13:00 | <hsivonen> | how does JAWS 10 get the landmark roles from a browser? |
| 13:11 | <annevk2> | if they don't map to an accessibility API direcltly I believe accessibility APIs (maybe not all) have some way to store unknown roles |
| 13:12 | <annevk2> | are there no posts on the benefits of e.g. using role=navigation in current screen readers? |
| 13:12 | <annevk2> | seems quite logical that some would have toyed with it already and posted results |
| 13:17 | <jgraham> | http://www.paciellogroup.com/blog/misc/landmarks.html |
| 13:20 | jgraham | makes a note to self that evolving the language seperatly from the implementation requirements is bad beacuse feedback from implementation can't affect the language design |
| 13:21 | <annevk2> | I'm thinking that dealing with screen readers must suck. I hate learning shortcuts but there you'd have to learn whole sets of words and their meaning to figure out what the screen reader is talking about. |
| 13:21 | <Philip`> | annevk2: That's probably why most people aren't voluntarily blind |
| 13:23 | <annevk2> | You think better software is holding them back? |
| 13:23 | <Philip`> | (Well, not specifically because of that issue with screen readers, but the general issue of it making life much harder) |
| 13:23 | <annevk2> | So only JAWS supports them. |
| 13:24 | <jgraham> | I guess it is interesting that there is no common use of audio to supplement the browsing experience |
| 13:24 | <jgraham> | Presumably that means that visuasl input is so high bandwidth that trying to add audio would just cause blocking |
| 13:25 | <Philip`> | jgraham: That's not true - a large number of pages use audio to provide information such as "this page was designed in the 90s" to their readers |
| 13:26 | <Philip`> | Also there's the click sound when you click links (at least in IE, at some point in the past) |
| 13:27 | <annevk2> | hsivonen, "IO Error: <http://annevankesteren.nl/2009/04/html5-wai-aria#comment-6768>; Code: 59/PROHIBITED_COMPONENT_PRESENT in FRAGMENT: A component that is prohibited by the scheme is present." is annoying |
| 13:27 | <annevk2> | hsivonen, just ignore the fragment (maybe say you did if you really want to) |
| 13:27 | <jgraham> | Philip`: I'm not sure author-supplied midi files count :) |
| 13:33 | <annevk2> | "I worked hard to get a good list of landmarks from the PF group, but what it ended up as was spec writers deciding what users should want." -- https://bugzilla.mozilla.org/show_bug.cgi?id=459395 |
| 13:34 | <annevk2> | What is it again they criticize the WHATWG for? |
| 13:41 | <annevk2> | http://wiki.whatwg.org/index.php?title=MetaExtensions&diff=3626&oldid=3467 o_O |
| 13:44 | <Lachy> | woah. That's a lot of proposals |
| 13:45 | <Lachy> | I thought he only added the useless addmark and addmarklocal extensions to address the non-existent problem he imagined in bug 6774 |
| 13:45 | <annevk2> | it's from the guy "spamming" W3C bugzilla |
| 13:46 | <Lachy> | yeah, I realised that |
| 13:47 | <annevk2> | hmm, per https://bugzilla.mozilla.org/show_bug.cgi?id=459395 browsers don't support ARIA roles. It's the AT being creative |
| 13:48 | <Lachy> | dammit, my email on dreamhost is down. That explains why I haven't received anything for a few hours :-( |
| 14:07 | <zcorpan> | annevk2: s/roles/landmarks/ |
| 14:08 | <zcorpan> | annevk2: the value of the role="" attribute is, in firefox, always exposed as-is via xml-roles |
| 15:10 | <takkaria> | I think that Nick guy thinks the mark element is going to be used like MS' smart tags were going to be |
| 15:11 | <takkaria> | but I'm not quite sure where he thinks the third-party servers are going to come in to it from |
| 15:26 | <Philip`> | takkaria: Maybe the third-party servers are like Google's cache servers, that serve pages and insert extra markup to highlight search term matches? |
| 15:27 | Philip` | assumes this is related to bug 6774, but he only skimmed the first half of the first comment before getting terminally bored |
| 15:31 | <jgraham> | If you got terminally bored, are you posting from beyond the grave? |
| 15:36 | <Philip`> | It was terminal to my reading of the bug, not to my life |
| 15:38 | <annevk2> | zcorpan, yeah, that's what I meant; hoped it was clear that the statement was scoped by the bug; guess not :) |
| 15:45 | <jgraham> | http://socialcollider.net/ is a kind of interesting way of killing browsers |
| 15:45 | <jgraham> | (because it's so slow) |
| 15:46 | <jgraham> | (especially in gecko) |
| 15:48 | <annevk2> | hsivonen, your HTML parser is not used for everything in Gecko? |
| 16:37 | <zcorpan> | hsivonen: still appears to be crashy for me |
| 16:52 | <zcorpan> | hsivonen: http://software.hixie.ch/utilities/js/live-dom-viewer/saved/87 |
| 16:53 | <annevk2> | whoa, Firefox 3.6something uppercases ö to Ö |
| 16:56 | <gsnedders> | WebKit does too |
| 16:56 | <gsnedders> | (Saf4b, this is) |
| 17:29 | gsnedders | is trying to get lyrics from music |
| 17:30 | gsnedders | supposes he could just ask the vocalist |
| 17:34 | gsnedders | is listening to Epic Fail by Hit the Breaks |
| 17:34 | <Philip`> | You could try listening to it and writing down the words |
| 17:34 | <gsnedders> | I expect a lot of you would find this amusing. |
| 17:34 | <gsnedders> | http://www.myspace.com/hitthebrakesno5 |
| 17:48 | <gsnedders> | Philip`: I was trying to listen to it and write down the world |
| 17:48 | <gsnedders> | *words |
| 19:30 | <nitroball> | Hey, wassup? Anyone try using the canvas tag with the Wii brower? It works very well except the color is missing for me on the lines. |
| 19:31 | <nitroball> | Color does show up in the IE, Opera, Firefox, etc. |
| 19:46 | <weinig> | annevk2: ping |
| 20:01 | <nitroball> | I think I figured it out... ctx.lineStyle = "rgba(255,165,0,1)"; doesn't work on the Wii browser. ctx.lineStyle = "rgb(255,165,0)"; does |
| 20:01 | <nitroball> | its the alpha value that was turning everything black |
| 20:51 | <Hixie> | so i'm going through the rdf folder looking for actual use cases and requirements |
| 20:51 | <Hixie> | 99% of the e-mails have absolutely nothing to do with use cases, despite the e-mail i sent saying that that's all i was looking for |
| 20:51 | <Hixie> | of the remainder, there are use cases like "provide information useful for improving the accessibility of Web content" |
| 20:52 | <Hixie> | which are the most vague, under-defined, and useless use case descriptions in the history of mankind |
| 22:06 | <gsnedders> | Hixie: Time to beat that: "to improve the long-term direction of the human race" |
| 22:08 | <Dashiva> | gsnedders: Can you prove that axiomatically? |
| 22:08 | <gsnedders> | Proof? Peh. |
| 22:30 | <gsnedders> | http://www.tesco.com/books/search.aspx?Ntt=%EF%BF%BF&Ntk=primary&VSI=1&Ntx=mode%2Bmatchall&Nty=1&N=0 |
| 22:43 | gsnedders | has interview 8am tomorrow morning, yay |