Options:
- # Session Start: Fri Aug 28 00:00:00 2009
- # Session Ident: #html-wg
- # [00:01] * Quits: adele (adele@17.246.18.111) (Quit: adele)
- # [00:02] * Joins: J_Voracek (irchon@173.74.111.41)
- # [00:02] * Quits: J_Voracek (irchon@173.74.111.41) (Client exited)
- # [00:03] * Quits: heycam (cam@210.84.56.211) (Quit: bye)
- # [00:10] * Joins: adele (adele@17.246.18.111)
- # [00:13] * Joins: Zeros (Zeros-Elip@98.231.211.61)
- # [00:15] * Joins: aroben_ (aroben@17.203.14.197)
- # [00:15] * Quits: aroben (aroben@17.203.14.197) (Connection reset by peer)
- # [00:15] * Quits: tlr (tlr@128.30.52.169) (Quit: tlr)
- # [00:23] * Quits: aroben_ (aroben@17.203.14.197) (Quit: aroben_)
- # [00:26] * Joins: aroben (aroben@17.203.14.197)
- # [00:37] * Quits: adele (adele@17.246.18.111) (Quit: adele)
- # [00:46] * Joins: adele (adele@17.246.18.111)
- # [00:48] * Quits: adele (adele@17.246.18.111) (Quit: adele)
- # [00:50] * Quits: smedero (smedero@128.208.133.191) (Quit: smedero)
- # [00:50] * Quits: MikeSmith (MikeSmith@mcclure.w3.org) (Client exited)
- # [00:54] * Joins: heycam (cam@130.194.72.84)
- # [00:54] * Quits: tH (Rob@82.4.89.172) (Quit: ChatZilla 0.9.85-rdmsoft [XULRunner 1.9.0.1/2008072406])
- # [01:08] * Joins: adele (adele@17.244.214.209)
- # [01:10] <pimpbot> bugmail: [Bug 7059] [blocked on xpathwg] Forking XPath <http://lists.w3.org/Archives/Public/public-html-bugzilla/2009Aug/0346.html>
- # [01:34] * Joins: aroben_ (aroben@17.246.19.32)
- # [01:34] * Quits: aroben_ (aroben@17.246.19.32) (Client exited)
- # [01:35] * Joins: aroben_ (aroben@17.203.14.197)
- # [01:35] * Quits: aroben (aroben@17.203.14.197) (Connection reset by peer)
- # [01:37] * Joins: mjs_ (mjs@17.246.16.249)
- # [01:40] * Quits: mjs (mjs@17.203.15.179) (Ping timeout)
- # [01:40] * mjs_ is now known as mjs
- # [01:48] * Quits: adele (adele@17.244.214.209) (Quit: adele)
- # [01:52] * Joins: aroben (aroben@17.246.19.32)
- # [01:54] * Quits: aroben_ (aroben@17.203.14.197) (Ping timeout)
- # [01:54] * Quits: aroben (aroben@17.246.19.32) (Client exited)
- # [01:54] * Joins: aroben (aroben@17.203.14.197)
- # [02:36] * Quits: aroben (aroben@17.203.14.197) (Quit: aroben)
- # [03:30] * Joins: mjs_ (mjs@17.246.16.249)
- # [03:30] * Quits: mjs (mjs@17.246.16.249) (Connection reset by peer)
- # [03:31] * mjs_ is now known as mjs
- # [03:41] <pimpbot> bugmail: [Bug 7451] New: "In the example above" should be "In the example below" <http://lists.w3.org/Archives/Public/public-html-bugzilla/2009Aug/0347.html>
- # [03:50] * Quits: mjs (mjs@17.246.16.249) (Quit: mjs)
- # [03:52] * Quits: jwatt (roslea@63.245.220.240) (Ping timeout)
- # [04:23] * Quits: Zeros (Zeros-Elip@98.231.211.61) (Quit: This computer has gone to sleep)
- # [04:33] * Parts: richardschwerdtfe (RichS@99.39.114.91)
- # [04:41] <pimpbot> bugmail: [Bug 7452] New: "The collection of Documents is the browsing context's session history. " -- What do you mean by "the collection of documents" ? not clear <http://lists.w3.org/Archives/Public/public-html-bugzilla/2009Aug/0348.html>
- # [04:59] * Joins: mjs (mjs@69.181.42.237)
- # [06:01] * Joins: MikeSmith (MikeSmith@mcclure.w3.org)
- # [06:11] * Quits: MikeSmith (MikeSmith@mcclure.w3.org) (Ping timeout)
- # [06:15] * Joins: MikeSmith (MikeSmith@mcclure.w3.org)
- # [06:44] * Quits: MikeSmith (MikeSmith@mcclure.w3.org) (Quit: Tomorrow to fresh woods, and pastures new.)
- # [07:34] * Joins: MikeSmith (MikeSmith@mcclure.w3.org)
- # [07:50] * Quits: MikeSmith (MikeSmith@mcclure.w3.org) (Ping timeout)
- # [08:30] * Quits: heycam (cam@130.194.72.84) (Quit: bye)
- # [08:48] * Joins: tlr (tlr@128.30.52.169)
- # [09:00] <jgraham> mjs: In colloquial en-gb "boffin" can have slightly unfriendly connotations. It tends to be a word used by lowbrow publications to refer to anyone vaugely scientific (but usually in a good-ish way) but is also a common insult at schools for anyone percieved to be uncool and academic
- # [09:04] * Joins: tH (Rob@82.4.89.172)
- # [09:07] <jgraham> Hixie: BTW the annotations service seems to be working for me now. So the spec gen should be working too
- # [09:08] * Joins: heycam (cam@210.84.56.211)
- # [09:12] <pimpbot> bugmail: [Bug 7059] [blocked on xpathwg] Forking XPath <http://lists.w3.org/Archives/Public/public-html-bugzilla/2009Aug/0349.html>
- # [09:42] <pimpbot> bugmail: [Bug 7453] New: "preferred name" should be "preferred MIME name" but it should also state that if there is no preferred MIME name for the encoding the name of the encoding should be used. <http://lists.w3.org/Archives/Public/public-html-bugzilla/2009Aug/0350.html>
- # [10:13] <pimpbot> bugmail: "[Bug 7059] [blocked on xpathwg] Forking XPath" (2 messages in thread) <http://lists.w3.org/Archives/Public/public-html-bugzilla/2009Aug/0352.html>
- # [10:20] <Hixie> jgraham: yeah it fixed itself
- # [10:26] <jgraham> Hixie: As far as I could tell it was timing out trying to read the issues from the issue tracker. Obviously it should be more robust against such things
- # [10:26] <Hixie> eh, don't worry about it (unless you want to of course)
- # [10:27] <Hixie> if it becomes a serious issue i'll just make the w3 copy not get checked in when it happens
- # [10:27] <hsivonen> Hixie: shouldn't the part about fetching profile URL say split on whitespace and then resolve each resulting token and fetch those?
- # [10:28] <Hixie> no?
- # [10:28] <Hixie> html4 only allowed one url
- # [10:28] <Hixie> and html5 doesn't allow any
- # [10:28] <Hixie> so...
- # [10:28] <hsivonen> Hixie: ok
- # [10:28] <hsivonen> Hixie: when HTML5 did allow profile, it allowed many URLs
- # [10:28] <anne> (some people think that was a bug in HTML4)
- # [10:29] <Hixie> yeah, back when i didn't realise that profile="" sucked, i was going to expand it as suggested in some html4 comments
- # [10:29] <mjs> didn't some spec at some point allow multiple profiles?
- # [10:29] <mjs> one of the XHTMLs?
- # [10:29] <Hixie> but turns out the bug is having profile="" at all
- # [10:29] <Hixie> html4 mentions that profile="" in future might be expanded to take multiple urls
- # [10:29] * hsivonen wonders if gmpg.org get a lot of automated profile fetches
- # [10:29] * Hixie really dislikes the idea of <main> or <content>
- # [10:29] <Philip> http://philip.html5.org/data/profile-values-2.txt
- # [10:29] <Philip> Seems people like multiple URLs
- # [10:29] <Philip> (A few people, anyway)
- # [10:30] <Hixie> Philip: but is anyone fetching them?
- # [10:30] <hsivonen> Hixie: isn't that an unscientific way to address <main>? :-)
- # [10:30] <Hixie> can't we get people to just use <div> instead of <main>
- # [10:30] <Hixie> the element is completely meaningless
- # [10:30] * Philip copies and pastes one of the URLs into his browser
- # [10:30] <Hixie> it has less meaning than <div>
- # [10:30] * anne really likes the idea of <main> :)
- # [10:30] <Philip> Someone is fetching them now!
- # [10:30] <Hixie> it's as bad as <di>
- # [10:30] <jgraham> Hixie: It improves source readbility and will prevent misuse of sectioning elements
- # [10:30] <Hixie> in fact it's worse than <di>
- # [10:30] <hsivonen> Hixie: a future version of Opera Mobile could use it for the skip to content feature
- # [10:31] <Hixie> hsivonen: you can already skip to content! it adds _nothing_ for skip to content.
- # [10:31] <anne> Hixie, come to think of it, I like <di> too, although the compat story is not nearly as good with <di>
- # [10:31] <jgraham> I also suspect that it will help with "skip to content" even though that is already theoretically possible
- # [10:31] <Hixie> preventing misuse is its only redeeming feature
- # [10:31] <mjs> it seems to be a common pattern to wrap the main content in an element, and it seems to be a recurring question how to do that with HTML5
- # [10:31] <hsivonen> Hixie: it adds the same thing to skip to content that rel=next adds to Fast Forward
- # [10:32] <Hixie> hsivonen: no, it doesn't. we already have that today.
- # [10:32] <Hixie> hsivonen: you can unambiguously tell where the content starts today.
- # [10:32] <hsivonen> Hixie: after </header>?
- # [10:32] <mjs> (I didn't know that <div> was the party line answer to that until recently)
- # [10:32] <Hixie> see my e-mail
- # [10:32] <Hixie> mjs: <div> is the element to use to wrap anything without adding semantics
- # [10:32] <jgraham> Hixie: Using subtractive mechanisms to identify main content strikes me as more fragile than explicit mechanisms
- # [10:33] <mjs> I think Hixie would say the first piece of content in <body> that's not one of the special non-main elements is how to skip to main
- # [10:33] <Hixie> jgraham: it's not substractive. You just keep skipping until you find the first element you're not skipping. It's more resilient than <main> because it lets you do it multiple times, regardless of nesting.
- # [10:33] <Hixie> in fact it's better because right now you get "skip to next interesting content" instead of "skip to main content", which is a subset
- # [10:34] <jgraham> Hixie: It's subtractive in the sense that :the main content is all the content except that which is a descendant of {some list of elements}"
- # [10:34] <Hixie> maybe we just need to stick <div>s in a bunch of examples, if people are all obsessing about styling their contents for some reason
- # [10:34] <jgraham> Hixie: I'm pretty sure that evangalism woin't work any more than it does for accessibility
- # [10:35] <Hixie> jgraham: what you're saying then is that the <main> proposal actually is a negative semantic, it says, "everything not in this element is not important"
- # [10:35] <jgraham> I really don't see the harm in giving people what they want here
- # [10:35] <Hixie> which is even worse
- # [10:35] <Hixie> how are people going to learn about <main>?
- # [10:36] <jgraham> The same way they learn about <section> i.e. from loosely worded, slightly inaccurate, blog posts
- # [10:36] <mjs> I think it just seems oddly non-parallel to people for everything but the main content to have a designated element, though I admit that pattern is found in other places in HTML on a smaller scale
- # [10:36] <Hixie> well let's have blog posts about <div>
- # [10:37] <mjs> "Hixie encourages people to use <div> more"
- # [10:37] <mjs> never thought I'd see the day :-)
- # [10:37] <Hixie> i really view this as exactly the same as <di>. It adds nothing semantically, it's only needed for styling if at all, and what it really does is add lots of possible ways for people to screw things up even more.
- # [10:37] <jgraham> A blog post that says "use <section> instead of <div> to mark up structure in HTML 5" will lead to the kind of misuse that we're seeing
- # [10:38] <jgraham> And I have written a blog post telling people not to use <section> where they mean <div>
- # [10:38] <anne> Hixie, it does add something
- # [10:38] <Hixie> anyway gotta go
- # [10:38] <anne> Hixie, it would map to role=main
- # [10:38] <mjs> the current answer to what to use for the main content is actually "<div> or <article> depending on what you're doing or maybe <section> or maybe you don't need a special element at all"
- # [10:38] <jgraham> I've done the evangalism thing
- # [10:39] <mjs> which is too complicated
- # [10:40] <jgraham> I would rather remove <article> and have <main> than have <section> and <article> but not <main>
- # [10:40] <anne> +1
- # [10:40] <jgraham> Since people find the whole <section>/<article> thing rather confusing
- # [10:40] <anne> I was thinking of proposing that actually, initially
- # [10:40] <jgraham> (sadly)
- # [10:41] <mjs> there is the blog case where you'd have multiple <article>s inside <main>
- # [10:41] * Quits: Lachy (Lachlan@85.196.122.246) (Quit: Leaving)
- # [10:42] <jgraham> mjs: I think <article> is a good idea and I am happy to keep it. But I'm not convinced it will be used correctly
- # [10:42] <mjs> not sure how relevant that is, but bloggers are particularly fond of making pretty structural markup it seems
- # [10:42] <mjs> jgraham: if both <main> and <article> existed, what errors do you think would be likely?
- # [10:43] <jgraham> mjs: Same as if <main> doesn't exist; people will use <article> where they mean <section>
- # [10:43] <mjs> jgraham: in what cases do you think people would say <article> instead of <section>
- # [10:43] <mjs> ?
- # [10:43] <jgraham> (the point of removing it would jsut be to reduce the language surface and so the number of things to get confused about. It is only tangentially related to the <main> element)
- # [10:44] <jgraham> (except insofar as one should be able to suggest something to remove when adding something to keep the overall language tractable)
- # [10:44] <mjs> jgraham: you made it sound like a choice between <article> and <main> - just trying to understand your thinking on this
- # [10:44] <jgraham> mjs: Yeah I was unclear
- # [10:45] <mjs> I imagine the difference between an article and a section of an article (or section in general) is clear
- # [10:45] <mjs> are you worried people will use <article> for article excerpts/summaries that are not truly full articles?
- # [10:46] <jgraham> mjs: There have been a few questions along the lines of "how do I nest <article> and <section> relative to one another" as well as "can I use <article> for a synopsis of an article"
- # [10:46] <mjs> what confuses me somewhat is when to use <section> vs <div> - some people have said you should only use <section> for its effect on the outline algorithm, but the spec also says applications can have <sections>, yet it's surely nonsensical to have an outline of an application UI
- # [10:47] <jgraham> So whilst <article> seems clear enough to me in theory I am not sure that in the real world it will ever get treated differently to <section> by UAs
- # [10:48] <jgraham> (I should state that I am more than happy to leave <article> in. I just don't think it's essential)
- # [10:48] * hsivonen learns that new Foo does stuff in JS without ()
- # [10:48] <mjs> (my favorite example on this is Google Calendar, which clearly if you look at it has visually identifiable "sections", but it would not really make sense to present those as an outline)
- # [10:49] <jgraham> mjs: Would it make sense if clicking on the heading in the outline view focused the first focusable child of that section?
- # [10:49] * jgraham doesn't use Google calendar
- # [10:50] <mjs> jgraham: you don't have to use it as such to get a look at what its main UI looks like
- # [10:50] <mjs> jgraham: the idea of an outline view of an application UI just strikes me as a category error
- # [10:51] <jgraham> mjs: Maybe.
- # [10:51] <jgraham> Maybe for some webpages you are so far over into application land that document semantics don't make much sense
- # [10:52] <jgraham> But there are a big class of pages, even those with some application-like behaviour, where headings and outlines and so on do make sense
- # [10:52] <mjs> so maybe you can say not to use <section> in that case, but that's not what the draft says at the moment
- # [10:53] <jgraham> The other question is "would an outlineish view help AT users?"
- # [10:54] <mjs> I don't think an outline of GMail makes sense, even though in some sense the search area, the folder list, the currently viewed folder, the chat box and the contact list are separate sections
- # [10:54] <jgraham> mjs: That seems sort of helpful to me
- # [10:55] <jgraham> I can imagine that if I was using AT I would like the ability to jump straight to the chat box
- # [10:55] <jgraham> And calling each different bit of the UI a "section" and giving each a heading seems like a reasonable way to achieve that
- # [10:55] <mjs> maybe, but I don't think native apps with different areas of the UI expose it in a form resembling a document outline
- # [10:56] <jgraham> Right, I don't know what native apps do with AT.
- # [10:56] <mjs> I would guess giving each of those sections a heading is something Google would not be up for, since it would mess up the layout
- # [10:56] <jgraham> They could always hide the heading
- # [10:57] <jgraham> Native applications like Lightroom have a very clealy define concept of "sections"
- # [10:57] <jgraham> *clearly
- # [10:58] <mjs> I believe on Mac OS X with VoiceOver, you can just cycle through the focusable controls, and there is no jumping around sections of the app; but I have not tried a super complicated app
- # [11:05] * Joins: Lachy (Lachy@213.236.208.22)
- # [11:10] <Julian> not allowing multiple URIs in head/@profile is a known bug in HTML4, and if HTML5 defines it it should do it properly
- # [11:11] <anne> is it in html4 errata?
- # [11:12] <Julian> it should have been, but there was nobody working on errata; we had the discussion on the HTML WG mailing list around April I think.
- # [11:13] <pimpbot> bugmail: [Bug 7454] New: Please do not remove these useful features - or are alternative mechanisms planned? <http://lists.w3.org/Archives/Public/public-html-bugzilla/2009Aug/0353.html>
- # [11:19] * Quits: Julian (chatzilla@217.91.35.233) (Ping timeout)
- # [11:22] * Joins: Julian (chatzilla@217.91.35.233)
- # [11:26] * Quits: Lachy (Lachy@213.236.208.22) (Quit: Leaving)
- # [11:28] <Julian> BTW - from http://www.w3.org/TR/html401/struct/global.html#adef-profile
- # [11:28] <pimpbot> Title: The global structure of an HTML document (at www.w3.org)
- # [11:28] <Julian> "This attribute specifies the location of one or more meta data profiles, separated by white space."
- # [11:28] <Julian> The next sentence of course sucks-
- # [11:30] * Joins: Lachy (Lachy@213.236.208.22)
- # [11:37] <anne> or the <dt> that proceeds it
- # [11:37] * anne has read HTML4
- # [11:38] <Julian> yes, and the DTD
- # [11:38] <Julian> HTML4 certainly is inconsistent here, and this has been a known issue for ages.
- # [11:40] <anne> HTML4 is pretty bad in general
- # [11:41] * Joins: webben (benh@217.12.14.240)
- # [11:43] <hsivonen> Hixie: I think if you forbid authors from using certain language features, those features don't suddenly stop having meaning
- # [12:04] <anne> you'd think it be easy to spell my first name correctly...
- # [12:05] <gsnedders|work> What, anna thinks people can spell her name write?
- # [12:33] * Joins: myakura (myakura@222.144.70.237)
- # [12:57] * Joins: J_Voracek (irchon@166.205.4.54)
- # [12:57] * Quits: J_Voracek (irchon@166.205.4.54) (Client exited)
- # [14:08] * Joins: J_Voracek (irchon@166.205.6.38)
- # [14:08] * Quits: J_Voracek (irchon@166.205.6.38) (Client exited)
- # [14:13] * Joins: aguest (d53d9aea@64.62.228.82)
- # [14:16] * Quits: aguest (d53d9aea@64.62.228.82) (Quit: http://www.mibbit.com ajax IRC Client)
- # [14:45] * Joins: J_Voracek (irchon@173.74.111.41)
- # [14:45] * Quits: J_Voracek (irchon@173.74.111.41) (Client exited)
- # [14:58] * Quits: tH (Rob@82.4.89.172) (Ping timeout)
- # [15:18] * Joins: jjok (jonathan_j@87.194.204.86)
- # [15:37] * Joins: maddiin (mc@87.185.222.55)
- # [16:15] * Quits: cyns (627da612@128.30.52.43) (Quit: CGI:IRC)
- # [16:49] * Quits: Julian (chatzilla@217.91.35.233) (Quit: ChatZilla 0.9.85 [Firefox 3.5.2/20090729225027])
- # [16:51] * Quits: ChrisWilson (cwilso@131.107.0.69) (Ping timeout)
- # [16:55] * Parts: jjok (jonathan_j@87.194.204.86)
- # [16:57] * Joins: ChrisWilson (cwilso@131.107.0.75)
- # [17:27] * Quits: mjs (mjs@69.181.42.237) (Quit: mjs)
- # [18:01] * Quits: gsnedders|work (gsnedders@88.131.66.80) (Quit: Adios Intarwebs!)
- # [18:35] * Quits: ChrisWilson (cwilso@131.107.0.75) (Ping timeout)
- # [18:41] * Joins: ChrisWilson (cwilso@131.107.0.77)
- # [18:57] * Quits: tlr (tlr@128.30.52.169) (Quit: tlr)
- # [19:04] * Joins: aroben (aroben@17.203.14.197)
- # [19:13] * Quits: webben (benh@217.12.14.240) (Ping timeout)
- # [19:16] * Joins: adele (adele@17.246.18.111)
- # [19:48] * Quits: Lachy (Lachy@213.236.208.22) (Quit: Leaving)
- # [19:54] * Joins: tH (Rob@82.4.89.172)
- # [19:57] * Joins: Lachy (Lachlan@85.196.122.246)
- # [20:01] * Joins: laplink (link@193.157.66.7)
- # [20:07] * Joins: aroben_ (aroben@17.246.19.32)
- # [20:07] * Joins: kliehm (kliehm@217.86.146.74)
- # [20:09] * Quits: aroben (aroben@17.203.14.197) (Ping timeout)
- # [20:10] * Joins: richardschwerdtfe (RichS@32.97.110.51)
- # [20:21] * Quits: myakura (myakura@222.144.70.237) (Quit: Leaving...)
- # [20:25] * Quits: laplink (link@193.157.66.7) (Quit: This computer has gone to sleep)
- # [20:50] * Joins: ROBOd (robod@89.122.216.38)
- # [20:52] * Quits: kliehm (kliehm@217.86.146.74) (Quit: ChatZilla 0.9.85 [Firefox 3.5.2/20090729225027])
- # [21:08] * Quits: adele (adele@17.246.18.111) (Quit: adele)
- # [21:10] * Quits: aroben_ (aroben@17.246.19.32) (Quit: aroben_)
- # [21:12] * Joins: laplink (link@193.157.66.7)
- # [21:26] <anne> http://www.w3.org/mid/OFD88BE4F2.25EFC17B-ON85257620.00666324-85257620.0067320D@lotus.com
- # [21:26] <pimpbot> Title: W3C TAG interest in Content-type "sniffing" from noah_mendelsohnus.ibm.com on 2009-08-28 (www-tag@w3.org from August 2009) (at www.w3.org)
- # [21:26] * anne wonders if the TAG has kept up with recent events
- # [21:33] * Quits: tH (Rob@82.4.89.172) (Ping timeout)
- # [21:48] * Joins: tH (Rob@82.4.89.172)
- # [21:54] * Quits: laplink (link@193.157.66.7) (Quit: This computer has gone to sleep)
- # [22:03] * Joins: adele (adele@17.246.18.111)
- # [22:08] * Joins: aroben (aroben@17.246.19.32)
- # [22:34] * Joins: laplink (link@193.157.66.7)
- # [22:47] * Joins: gsnedders (gsnedders@83.252.192.255)
- # [23:00] * Quits: adele (adele@17.246.18.111) (Quit: adele)
- # [23:06] * Quits: ROBOd (robod@89.122.216.38) (Quit: http://www.robodesign.ro )
- # [23:10] * Quits: laplink (link@193.157.66.7) (Quit: This computer has gone to sleep)
- # [23:28] * Quits: aroben (aroben@17.246.19.32) (Quit: aroben)
- # [23:31] * Parts: richardschwerdtfe (RichS@32.97.110.51)
- # [23:43] * Joins: aroben (aroben@17.246.19.32)
- # Session Close: Sat Aug 29 00:00:00 2009
The end :)