| 00:23 | <MikeSmith> | anybody seen nessy around today? |
| 00:42 | <MikeSmith> | Hixie: about the broken links in IDLs in the the author view of the spec, for the static W3C copy of the author-view that I'm generating, I guess I can have the script for that rewrite the fragment IDs in the IDLs so that they point to the domintro stuff instead, but not clear to me if/how it's fixable in the dynamic/JS-switchable view of the spec |
| 00:44 | <MikeSmith> | and since not all attributes in the IDLs have corresponding domintro content, I guess on I can just have the build of the static author view un-link them and instead just display them as normal text |
| 00:45 | <MikeSmith> | It seems like that's mainly the case for DOM attributes that have same names as content attributes |
| 00:45 | <MikeSmith> | or whatever they're currently called in the spec to distinguish them from one another |
| 00:46 | <MikeSmith> | hsivonen: bugzilla.validator.nu seems to be responding very slowly |
| 00:54 | <KaOSoFt> | Hello. |
| 00:56 | <MikeSmith> | KaOSoFt: good morning |
| 01:03 | <KaOSoFt> | I've just recently been practicing HTML5, and while thinking in semantics, I thought about formatting (giving it meaning) my data. I read some web-logs which mentioned Microformats, and while it looks like an interesting approach, some others barely wrote about RDFa. |
| 01:03 | <KaOSoFt> | I then checked some documents about RDFa, but I don't seem to find any examples using HTML5. I suppose it's because both need to get the recommendation status, but I wanted to check nevertheless. Given there is a DTD for RDFa with XHTML but since HTML5 doesn't use it, I wanted to know how should I start approaching RDFa in HTML5: is it okay to just start using RDFa properties, or do you rather know of/have an example? |
| 01:04 | <KaOSoFt> | I still need to know exactly how HTML5 is going to work with DTD, or without them. |
| 01:04 | <MikeSmith> | KaOSoFt: you know of the http://dev.w3.org/html5/rdfa/ ? |
| 01:04 | <MikeSmith> | KaOSoFt: HTML5 does not rely on DTDs |
| 01:05 | <MikeSmith> | nobody should be using DTDs at this point, for anything |
| 01:05 | <MikeSmith> | they have been obsolete for a long time |
| 01:05 | <KaOSoFt> | 1. Didn't know about that, thanks. :) |
| 01:05 | <KaOSoFt> | For some reason I never got into that link. :s |
| 01:05 | <Hixie> | MikeSmith: yeah i don't know what to do in the dynamic case |
| 01:05 | <MikeSmith> | if you need a schema to use for validation, the right thing to use these days is either RelaxNG or W3C XML Schema |
| 01:05 | <KaOSoFt> | 2. I see, I wanted to be sure anyways. |
| 01:06 | <Hixie> | MikeSmith: in the static case it might make more sense to start from the pre-anolis draft, change the domintro elements to dfn, and rerun anolis |
| 01:06 | <MikeSmith> | Hixie: hmm, yeah |
| 01:06 | <MikeSmith> | Hixie: for the dynamic case, maybe l e |
| 01:06 | <MikeSmith> | oops |
| 01:08 | <MikeSmith> | Hixie: for the dynamic case, maybe one of your scripts could cause an alert to pop up, saying "The target of this link is not in the author view, do you want to switch to the full (implementor) view instead?" |
| 01:08 | <MikeSmith> | or something like that |
| 01:08 | <MikeSmith> | I don't know how feasible that would be, though |
| 01:09 | <Hixie> | ooh that's a good idea |
| 01:11 | <KaOSoFt> | MikeSmith- Fortunately, that document just mentions the changes to make the XHTML+RDFa recommendation work with HTML5, which implies it is still pretty much XHTML+RDFa. |
| 01:11 | <KaOSoFt> | Thank you. |
| 01:11 | <KaOSoFt> | ^^_ |
| 01:11 | <KaOSoFt> | ^_^ |
| 01:11 | <Hixie> | MikeSmith: can you file a bug to that effect? |
| 01:12 | <MikeSmith> | Hixie: yeah, will do |
| 01:12 | <Hixie> | thanks |
| 01:13 | <KaOSoFt> | Hixie- This example is yous, isn't it? http://www.w3.org/TR/xhtml-rdfa-primer/alice-example.html |
| 01:13 | <Hixie> | lol, laura just implied that if the WAI isn't consulted on something, it can't be good for accessibility |
| 01:13 | <MikeSmith> | KaOSoFt: there are some unresolved issues with full integration of RDFa into text/html |
| 01:13 | <Hixie> | KaOSoFt: never seen it before |
| 01:14 | <KaOSoFt> | MikeSmith- Yeah, I thought so. I know they are still not Recommendations, but wanted to be sure how far they were. Thanks. |
| 01:16 | <MikeSmith> | KaOSoFt: have you looked at the HTML5 Microdata feature? |
| 01:17 | <KaOSoFt> | Hixie- Oh, sorry, I got my tabs mixed, and thought it was yours. :P |
| 01:19 | <KaOSoFt> | MikeSmith- Well, I was reading what some search engine crawlers understood, and that's why I knew of Microformats and RDFa. |
| 01:19 | <KaOSoFt> | Now that you mentioned HTML5 Microdata, I'm currently reading this: http://www.whatwg.org/specs/web-apps/current-work/multipage/microdata.html |
| 01:22 | <KaOSoFt> | The current HTML5 specification is quite large. I have to check the multipage version, otherwise the browsers take a while to load it all. |
| 01:24 | <KaOSoFt> | Hmmm, looks interesting. |
| 01:29 | <MikeSmith> | Google's rich-snippets feature also supports HTML5 Microdata now |
| 01:29 | <MikeSmith> | if your goal is to have search-engine support |
| 01:30 | <KaOSoFt> | Yes, exactly. :) |
| 01:33 | <KaOSoFt> | By the way, is this the correct channel for these inquiries? I don't mean to disturb whatever you do here. |
| 01:33 | <KaOSoFt> | ._. |
| 01:37 | <MikeSmith> | KaOSoFt: here is fine |
| 01:38 | <MikeSmith> | especially if you have questions about the Microdata feature |
| 01:38 | <MikeSmith> | foolip and others have used it some and can help a bit probably |
| 01:39 | <KaOSoFt> | Yeah, I'm currently searching using "HTML5 Microdata", and have found some articles. :D |
| 01:39 | <KaOSoFt> | The first step was knowing about it. :P |
| 01:59 | <KaOSoFt> | Wow, there are already a lot of tools for testing/trying/enjoying Microdata. |
| 01:59 | <KaOSoFt> | I'm currently checking foolip's Live Microdata. :D |
| 02:00 | <kennyluck> | KaOSoFt, any pointer? |
| 02:00 | <KaOSoFt> | Pointer? |
| 02:00 | <KaOSoFt> | http://gitorious.org/microdatajs |
| 02:00 | <kennyluck> | yeah, like this, thatnks. |
| 02:00 | <KaOSoFt> | http://james.html5.org/microdata/ |
| 02:00 | <KaOSoFt> | http://philip.html5.org/demos/microdata/demo.html |
| 02:00 | <KaOSoFt> | :D |
| 02:01 | <kennyluck> | Thanks! :) |
| 02:10 | <MikeSmith> | kennyluck: hey, you at t45 already? |
| 02:10 | <kennyluck> | MikeSmith, yes I am! Why? |
| 02:11 | <MikeSmith> | just wondering |
| 02:11 | <MikeSmith> | some trains seem to be delayed |
| 02:11 | <kennyluck> | The rain wasn't that big when I came here. |
| 02:16 | <MikeSmith> | kennyluck: if you have time in working on some HTML5-related stuff, there's a number of things that might be interesting |
| 02:16 | <MikeSmith> | work on the validator.nu backend, for one |
| 02:16 | <MikeSmith> | also, the RDFa WG is working on and RDFa API |
| 02:17 | <kennyluck> | MikeSmith, cool. |
| 02:17 | <kennyluck> | Yep, I would like to join RDFa WG, I just got my W3C member account. |
| 02:17 | <MikeSmith> | great |
| 02:19 | <kennyluck> | MikeSmith: I am also now discussing with Fumi about holding an event for Semantic Web & Linked Data. I wonder whether I can invite you to talk about Microdata. |
| 02:20 | <kennyluck> | Actually not an event, but gatherings twice a month. |
| 02:27 | <MikeSmith> | if you do the event you talked about, make sure to invite Kanzaki-san |
| 02:30 | <kennyluck> | MikeSmith, yup, that's on the plan. It was very sad that I didn't meet him last year. I even bought 5 of his books, and I gave Tim one of them (should have got his signature) |
| 02:33 | <MikeSmith> | anyway, I may not be the best person to present about Microdata |
| 02:34 | <MikeSmith> | I'm pretty sure we could find someone who's already using it in pages |
| 02:34 | <kennyluck> | OK, but you are always welcome :) I just forwarded my plan to you. |
| 02:34 | <MikeSmith> | OK |
| 02:34 | <MikeSmith> | adding RDFa support to validator.nu would be another thing worth exploring |
| 02:34 | <MikeSmith> | i'm not sure that's completely possible right now without some changes being made to the HTML5 parsing algorithm |
| 02:35 | <MikeSmith> | but hsivonen or Hixie would know better |
| 02:35 | <MikeSmith> | hsivonen did an implementation a while back that didn't support CURIEs but the RDFa community did not seem to be too thrilled with that |
| 02:35 | <kennyluck> | We should be happy...don't know why. |
| 02:36 | <MikeSmith> | heh |
| 02:36 | <MikeSmith> | http://www.html5.jp/blog/2010/04/04/html5-api/ is interesting |
| 02:36 | <kennyluck> | But I should say, I have little knowledge about the difference of HTML+RDFa and XHTML+RDFa. |
| 02:36 | <kennyluck> | Does HTML+RDFa means RDFa 1.1 ? |
| 02:38 | <kennyluck> | By the way, CURIE like "[wiki:Test]" is a mess, we RDF community people can't live without "foaf:knows". Maybe that's what happened. |
| 02:38 | <kennyluck> | s/, we/, but we/ |
| 02:38 | <kennyluck> | Ughhhh.. |
| 02:39 | <MikeSmith> | hsivonen could give you the details about what the validator implementations issues are with respect to supporting RDFa |
| 02:39 | <MikeSmith> | HTML+RDFa means supporting whatever version of RDFa in text/html MIME type |
| 02:40 | <MikeSmith> | as opposed to XML MIME type |
| 02:40 | <kennyluck> | Yes, so HTML+RDFa doesn't have to meet the "copy and paste" requirement. Am I right? |
| 02:40 | <MikeSmith> | yeah, arguably |
| 02:41 | <MikeSmith> | to put it in other terms, RDFa+HTML means supporting it RDFa documents that are parsed with HTML parsers instead of XML parsers |
| 02:42 | <MikeSmith> | or, I guess, in any tools that don't rely on the pseudo-namespace mechanism that RDFa uses |
| 02:43 | <kennyluck> | OK, I need to look it up. I am not familiar with the difference between HTML DOM and XML DOM.. |
| 02:55 | <KaOSoFt> | Well, I've read enough for the day. Tomorrow I'll practice a little bit more, and try to implement Microdata in a project I'm working on. Thank you all. |
| 02:55 | <KaOSoFt> | Goodbye. |
| 02:55 | <kennyluck> | KaOSoFt, bye. |
| 03:20 | <othermaciej> | good evening everyone |
| 04:43 | <MikeSmith> | Hixie: I'm only now noticing that hostname on the From address of mail generated by W3C bugzilla has changed from wiggum.w3.org to jessica.w3.org |
| 04:43 | <MikeSmith> | I don't know if that makes any difference for any mail filtering you have set up |
| 04:43 | <MikeSmith> | it does for me |
| 04:44 | <MikeSmith> | I think it's also breaking some of the hacky commit-notification stuff I have set up |
| 04:44 | <MikeSmith> | I'll have to check and see |
| 04:45 | <Hixie> | that was one of the few changes that didn't actually break anything for me :-) |
| 04:45 | <MikeSmith> | OK |
| 04:46 | <MikeSmith> | the may be some other stuff that's borked due to the bugzilla upgrade |
| 04:46 | <MikeSmith> | anyway, I'm going to be away for the next 12+ hours or so |
| 04:46 | <MikeSmith> | traveling |
| 04:47 | <MikeSmith> | so if in the mean time you discover anything busted, please ping shepazu about it |
| 04:48 | <Hixie> | k |
| 04:50 | <MikeSmith> | Hixie: btw, as far as the general rationale for having document-conformance criteria in the spec, I think another reason that could possibly be cited is that non-browser tools that consume HTML also benefit from the existence of constraints that bound the possible markup cases they are designed to process |
| 04:50 | <Hixie> | are there any conformance criteria that we've added for that reason? |
| 04:51 | <MikeSmith> | I don't know if there are some that have been added for that specific reason |
| 04:51 | <MikeSmith> | but it is at least a beneficial side effect |
| 04:51 | <MikeSmith> | tools can't be expected to process every insane markup case that some author wants to use for whatever bizarre reasons |
| 04:52 | <MikeSmith> | e.g., whatever recent bug was arguing that div should be allowed as a child of span |
| 04:53 | <Hixie> | depends on the tool |
| 04:53 | <MikeSmith> | true |
| 04:53 | <Hixie> | not really sure what kind of tool would have problem with that kind of thing |
| 04:53 | <MikeSmith> | I guess cases like that actually break the element semantics |
| 04:55 | <Hixie> | anyone know of a character that in a particular character encoding other than UTF-16 or -32 becomes a multibyte character with one of the bytes being 0x20? |
| 05:09 | <wirepair> | http://www.fileformat.info/info/unicode/char/2022/index.htm |
| 05:09 | <wirepair> | %u2022 |
| 05:09 | <wirepair> | er sorry that is utf-165 |
| 05:09 | <wirepair> | er 16 ;> |
| 05:12 | <wirepair> | looks like shift_jis and euc_jp don't have any |
| 05:29 | <Hixie> | well if there aren't any then that makes my life easier. :-) |
| 07:32 | <JonathanNeal> | ahoy! |
| 09:20 | <Hixie> | othermaciej: 94? |
| 09:20 | <othermaciej> | Hixie: pardon? |
| 09:20 | <Hixie> | how did you count 94 open bugs? |
| 09:20 | <Hixie> | i count 43 |
| 09:20 | <othermaciej> | Hixie: did I say 94? |
| 09:20 | <othermaciej> | I meant 43 |
| 09:21 | <Hixie> | that's quite the typo :-P |
| 09:21 | <othermaciej> | Hixie: d''oh, I forgot to edit it from last month's value |
| 09:21 | <Hixie> | you had me momentarily worried that my scripts were missing some huge number of bugs |
| 09:22 | <othermaciej> | Hixie: no, I distinctly remember doing a query and getting 43 |
| 09:23 | <Hixie> | good, good |
| 09:23 | <othermaciej> | and indeed the ones I called out as oldest reopened and oldest non-reopened will be in that set |
| 09:23 | <othermaciej> | sorry for scaring you |
| 09:23 | <Hixie> | so long as i'm not blithely ignoring half the bugs out there, i'm happy :-) |
| 10:50 | annevk | wonders if it would be better to have File/LocalFile/Blob rather than File/Blob with Blob overloaded with type/URL |
| 10:56 | <annevk> | defining maps for properties is still hard considering forward extensibility |
| 10:57 | <annevk> | at least if we do things like .property.partOfMap |
| 10:57 | <annevk> | since partOfMap might conflict with a future value API |
| 10:57 | <annevk> | maybe maps and list should be on their own attributes |
| 10:57 | <annevk> | e.g. margin.m.left |
| 10:57 | <annevk> | or b.l[0].m.color |
| 10:58 | <annevk> | euh, I meant background.l[0].m.color |
| 10:59 | <annevk> | meh, complicated |
| 10:59 | <annevk> | also fun |
| 11:41 | <annevk> | more thinking time tomorrow |
| 16:20 | <ment> | is there a sheet somewhere listing supported <audio>, <video> codecs in various browsers? |
| 16:25 | <paul_irish> | ment: theres a page on html5doctor that tries to capture audio codecs.. video support.. that one should be easier to find. btw you can use Modernizr to detect these on the fly. |
| 16:34 | <ment> | paul_irish: modernizr it is! thanks |
| 16:56 | <annevk> | "I dont think there is so much difference between say :hover and mouseover, or :visited and click (except that click is more general), or state and event in general." |
| 16:56 | <annevk> | sounds like it is time to give up on this thread |
| 17:26 | <dglazkov> | ah what the heck |
| 17:26 | <dglazkov> | good morning, Whatwg! |
| 17:28 | <annevk> | morning dglazkov! |
| 17:28 | annevk | wonders if everyone has Easter Monday off |
| 17:28 | <dglazkov> | nope. |
| 17:29 | <dglazkov> | but some peeps do. |
| 17:30 | <annevk> | we do, though I've done some work anyway |
| 17:32 | <JonathanNeal> | You're a Unicorn! |
| 17:34 | <annevk> | orly? |
| 17:48 | <JonathanNeal> | yes, annevk, rly. |
| 18:09 | <JonathanNeal> | Is there no -moz-linear-gradient equiv in Opera, still? |
| 18:15 | <jwalden> | wasn't there a CSS Opera logo recently that I thought used one? or did someone just make one for the heck of it? |
| 18:16 | <Philip`> | http://desandro.com/articles/opera-logo-css/ |
| 18:16 | <Philip`> | It doesn't work in Opera |
| 18:16 | <paul_irish> | http://people.opera.com/howcome/2010/tests/operalogo.html |
| 18:17 | <Dashiva> | Absolut semantic markup |
| 18:19 | <annevk> | we don't do gradients yet |
| 18:19 | <JonathanNeal> | ah, got it. |
| 18:20 | <JonathanNeal> | iz cool, just making sure I was supporting Opera's syntax if it was there. |
| 18:21 | <Dashiva> | How would you make that accessible the best way? |
| 18:23 | <JonathanNeal> | Dashiva, was that question for me? If so, I don't quite understand the question. |
| 18:23 | <Dashiva> | <div><div></div><div></div><div></div><div></div><div></div><div></div><div></div><div></div></div> doesn't explain very much what's going on |
| 18:26 | <paul_irish> | Yah where's the semantic O in the middle of it? |
| 18:26 | <paul_irish> | a buncha slackers. |
| 18:30 | <JonathanNeal> | They have classnames, Dashiva :) |
| 18:32 | gsnedders | notes we can render the Opera logo using SVG |
| 18:37 | <JonathanNeal> | You'll have gradient soon enough, fo sho. |
| 18:39 | <TabAtkins> | Moz just happened to be the first to implement the new gradient syntax. Others will get on the ball soon enough. |
| 18:44 | <TabAtkins> | exit |
| 18:45 | <JonathanNeal> | entrance? |
| 18:46 | <TabAtkins> | Eh, forgot the /. Wanted to reload my prefs. |
| 18:50 | <JonathanNeal> | Oh, you weren't just being dramatic :) |
| 19:04 | <TabAtkins> | othermaciej: So should I wait on writing counterproposals? I was going to block out some time this week for them. |
| 19:04 | <othermaciej> | TabAtkins: it's never a problem to start early, or to submit proposals without a formal call |
| 19:05 | <TabAtkins> | Sure, but if something might get rejected before having counterproposals asked for, it's a waste of time for me to do anything. Is that a possibility? |
| 19:05 | <TabAtkins> | Basically I am trying to minimize the amount of time I waste on countering ridiculousness. |
| 19:06 | <othermaciej> | TabAtkins: I don't know. I have suggested to the other co-chairs that if a proposal has no support besides the author, and significant opposition, we should consider just doing a call for consensus to reject it. |
| 19:06 | <othermaciej> | TabAtkins: (a) Not sure what they will think. (b) Not sure if Laura's comments can be counted as "support". |
| 19:06 | <TabAtkins> | Kk. |
| 19:06 | <othermaciej> | we will probably decide on some course of action for at least some of the proposals by early this week |
| 19:07 | <TabAtkins> | K. I'll stay in the loop, but not plan on anything yet. |
| 19:12 | <othermaciej> | Hixie said he thought one counter-proposal might be sufficient to counter all the semantic element/attribute proposals |
| 19:12 | <othermaciej> | perhaps not literally identical text, but at least a template |
| 19:12 | <othermaciej> | you could maybe consider using a wiki, since others (e.g. John Foliot) have expressed interest in contributing |
| 19:13 | <TabAtkins> | Yeah, I'll put it up on the wiki if I start on them. |
| 19:22 | <hober> | I had started my ISSUE-95 zero-edit CP a while ago; I'd be willing to generalize it into a broader one for the several related issues |
| 19:23 | <hober> | here's the existing text; I have several recent emails flagged in gmail as source material for updating the CP this week: http://hober.jottit.com/issue-95_change_proposal |
| 20:05 | <annevk> | http://lists.w3.org/Archives/Public/www-html-editor/2010JanMar/0013.html o_O |
| 20:05 | <annevk> | http://lists.w3.org/Archives/Public/www-html-editor/2010JanMar/att-0011/00-part apparently a common thing |
| 20:05 | annevk | remembers someone from the W3C telling him about it |
| 20:06 | <Dashiva> | What is happening? |
| 20:06 | <annevk> | nothing much, it's a free day |
| 20:23 | <annevk> | whoa: http://www.rijksoverheid.nl/copyright |
| 20:23 | <annevk> | that's pretty cool |
| 20:40 | <annevk> | "PS Notice lack of smiley" lol |
| 20:41 | <annevk> | are all her emails like that? maybe I should read some more |
| 20:42 | <othermaciej> | annevk: what email is that from? |
| 20:46 | <annevk> | http://lists.w3.org/Archives/Public/www-archive/2010Apr/0007.html |
| 20:53 | <annevk> | http://twitter.com/shelleypowers/status/11657402980 -- "It begins to look like a month of my work will be rejected out of hand at the HTML5 WG http://bit.ly/cgsEb9 because of five lazy assholes" |
| 20:53 | <annevk> | which is followed with |
| 20:54 | <annevk> | http://twitter.com/shelleypowers/status/11657500563 -- "I wonder when the new boss at the W3C will finally so something about the atmosphere of hostility and bias at the HTML5 WG" |
| 20:54 | <annevk> | classic |
| 20:56 | <othermaciej> | I'm definitely sensing some hostility |
| 20:57 | <dglazkov> | TabAtkins: are you to blame for everything? |
| 20:58 | <annevk> | dglazkov, you guys must regret what you got yourselves into by hiring this TabAtkins figure |
| 20:58 | <annevk> | dglazkov, nasty Texans :p |
| 20:58 | <JonathanNeal> | I blame Tab when little things go wrong. |
| 20:58 | <dglazkov> | what? we hired him?! I thought it was just April Fool's |
| 20:58 | <JonathanNeal> | My girlfriend misunderstands me. |
| 20:58 | <JonathanNeal> | I hit a red light even though there are no cars coming the other way. |
| 21:00 | <JonathanNeal> | I say to myself, "Tabbit." |
| 21:03 | <TabAtkins> | I am, basically, to blame for all of your small ills. |
| 21:03 | <TabAtkins> | Re Shelley, sigh. |
| 21:21 | <JonathanNeal> | So whatwg has no rendering requirement or recommendation for :link { color }, right? |
| 21:24 | <Hixie> | http://www.whatwg.org/specs/web-apps/current-work/complete.html#fonts-and-colors |
| 21:26 | <JonathanNeal> | Hixie, you know I always look at that guy ... it's just that sometimes I miss 'em. |
| 21:26 | <JonathanNeal> | Like body having a margin, it's not writen out css style. |
| 21:29 | <JonathanNeal> | Oh, there it is, blue and purple |
| 21:31 | <JonathanNeal> | brain fail |
| 21:31 | <dglazkov> | hey, standards brains, is there anything more modern and specific than this on appendChild()? http://www.w3.org/TR/DOM-Level-3-Core/core.html#ID-184E7107 |
| 21:32 | <dglazkov> | specifically, the actual algorithm of appending document fragments. |
| 21:32 | <othermaciej> | dglazkov: DOM Level 3 Core is the most recent official standard |
| 21:33 | <othermaciej> | dglazkov: inserting DocumentFragments into the DOM is a particularly hairy area though, due to mutation events |
| 21:33 | <dglazkov> | othermaciej: ok. If I wanted to specify more, how would I go about it? |
| 21:33 | <dglazkov> | othermaciej: yes indeedy :) |
| 21:33 | <othermaciej> | dglazkov: volunteer to edit DOM Level 4 Core? |
| 21:33 | <dglazkov> | never! |
| 21:33 | <dglazkov> | :) |
| 21:34 | <dglazkov> | file a bug mebbe? |
| 21:34 | gsnedders | notes shepauz did |
| 21:34 | <gsnedders> | (or whatever his IRC nick is) |
| 21:39 | <JonathanNeal> | fieldset is display inline according to you folks, right? |
| 21:40 | <JonathanNeal> | I never see where it is recommended or required to be display block. |
| 21:40 | <JonathanNeal> | Though I want to infer that from the margin and padding. |
| 21:40 | <dglazkov> | othermaciej: FF has the sanest behavior on document fragments + mutation events, so I'll just use theirs |
| 21:41 | <jwalden> | dear lazyirc, anyone feel like reporting to Opera that they've mis-implemented calculation of the background positioning area in 10.51? testcase is <http://whereswalden.com/files/mozilla/background-size/bg-position.html>;, cowbell should scale over entire canvas, not over the area of the rounded dashed rectangle+margins |
| 21:41 | <othermaciej> | dglazkov: is there anything wrong with the way WebKit does it currently? |
| 21:41 | <othermaciej> | dglazkov: is there a compat issue caused by this? |
| 21:43 | <dglazkov> | othermaciej: in a word, yes |
| 21:43 | <gsnedders> | jwalden: Are you sure we actually support cover at all? |
| 21:43 | <jwalden> | gsnedders: yes |
| 21:44 | <jwalden> | (make the window tall and narrow to demonstrate the bug, to elaborate) |
| 21:44 | gsnedders | doesn't have 10.5x at hand |
| 21:47 | <TabAtkins> | JonathanNeal: fieldset is crazytimes, and not actually done with CSS in any browser |
| 21:47 | <JonathanNeal> | Sure, but you're recommending margins and padding, so why not display type? |
| 21:48 | <TabAtkins> | The hardest part is positioning <legend> (which, though it is required to be the first child of <fieldset>, for legacy reasons must be accepted as any child) and styling the border (since it breaks behind the <legend>). |
| 21:48 | <TabAtkins> | Shrug, dunno. |
| 21:53 | <othermaciej> | I think <fieldset> is essentially a block other than the special handling of <legend> |
| 21:54 | <TabAtkins> | It is. If we could find a way to specify the border rendering, and the positioning of <legend> without making all <fieldset>s into positioning roots, we could do it all in CSS. |
| 21:56 | <othermaciej> | <fieldset> is kind of legacy cruft - it isn't very useful in modern UI design |
| 21:56 | <JonathanNeal> | It is essentially a block, so is it a bad idea just to write it in? |
| 21:56 | <othermaciej> | so not worth spending too much effort on it |
| 21:56 | <JonathanNeal> | It's not legacy if you're putting it in the new spec. |
| 21:56 | <TabAtkins> | (While I don't really like selectors appearing in values, I suspect that's what would be required to specify <legend>'s positioning. position-to: "fieldset"; would grab the closest ancestor that matches the selector, and treat it as the positioning root (unless there was a closer ancestor who is actually a positioning root from normal means). |
| 21:57 | <TabAtkins> | That would be useful in more than just <fieldset>/<legend>. |
| 21:57 | <othermaciej> | in WebKit, fieldset in fact has a display: block rule in the UA stylesheet |
| 21:57 | <othermaciej> | TabAtkins: to really do fieldset/legend, you need the notion of an interrupted border with something positioned in it, and an item that acts as the border caption for the relevant container |
| 21:58 | <othermaciej> | TabAtkins: so I think it can be done without selectors in values |
| 21:58 | <KrocCamen> | Aren’t fieldsets needed to get radio buttons to know each other? (press one, and the others depress) |
| 21:58 | <othermaciej> | TabAtkins: same basic concept as display: table and display: table-caption |
| 21:58 | <KrocCamen> | (Or am I thinking of VB6, perhaps I am) |
| 21:58 | <TabAtkins> | border-caption... That's an interesting way to capture that. |
| 21:58 | <TabAtkins> | KrocCamen: No, radios know each other by @name. |
| 21:58 | <othermaciej> | TabAtkins: the latter finds its containing table without the table having to act as a containing block and without specifying it explicitly by selector |
| 22:02 | <TabAtkins> | Aw man, I love Template. So good! |
| 22:22 | <JonathanNeal> | I also noticed that HR has no recommendation for margin, so no margin is recommended. |
| 22:23 | <TabAtkins> | Hm, really? Pretty sure <hr> *does* have margin. |
| 22:24 | <TabAtkins> | It's got ".5em auto" in Chrome, at least. |
| 22:26 | <TabAtkins> | <hr> has some odd legacy issues, but none that would actually require any changes to CSS. Everybody's implementation of it is just an empty block box with an inset border and a small vertical margin. |
| 22:26 | <TabAtkins> | It's just that, for some reason, some browsers don't actually expose it as such through CSS. It's possible that all the major ones work correctly now, though? I'd have to test to be sure. |
| 22:26 | <TabAtkins> | And I can't test properly until I get my Windows box. |
| 22:38 | <JonathanNeal> | I was just wondering because I couldn't find a requirement or recommendation in the spec (maybe I'm blind again) |
| 22:38 | <JonathanNeal> | http://www.iecss.com/whatwg.html experimental comparison |
| 22:39 | <Hixie> | http://www.whatwg.org/specs/web-apps/current-work/complete.html#the-hr-element-0 is all there is about <hr> styles |
| 22:39 | <Hixie> | no vertical margins currently |
| 22:40 | <JonathanNeal> | Hixie, right .. that's what I was curious about. |
| 22:40 | <TabAtkins> | IE* and 9 seem to be setting up the properties correctly, they just set a few additional ones as well for some reason. No clue why it has overflow or text-align on it. |
| 22:40 | <TabAtkins> | Hixie: I think modern browsers have converged on .5em vertical margins on <hr>. Can we have that in the spec? |
| 22:41 | <Hixie> | file a bug |
| 22:41 | <Hixie> | but yes |
| 22:41 | <TabAtkins> | Will do. |
| 22:42 | <Hixie> | also mention that i don't set 'height' if size="" isn't present currently |
| 22:42 | <Hixie> | oh i guess i don't need to, since i don't set box-sizing |
| 22:42 | <Hixie> | so nevermind that |
| 22:43 | <TabAtkins> | Never minded. |
| 22:45 | <TabAtkins> | You don't need to talk about height *anyway*. You never manipulate height or width on it - it just acts like a normal empty block box. |
| 22:49 | <JonathanNeal> | but vertical margin may still added? |
| 22:50 | <TabAtkins> | Yes. <hr> should have "margin: .5em auto;" on it. |
| 22:51 | <JonathanNeal> | Since I'm so hit and miss, I'll ask another question I couldn't directly identify in the spec --- is there a recommended font-family (serif?), font-size (16px?), line-height (1.2)? |
| 22:52 | <Hixie> | no |
| 22:52 | <Hixie> | that's css's territory |
| 22:56 | <JonathanNeal> | Most browsers settle on something like 16px/1.2 and some kind of serif. |
| 22:56 | <JonathanNeal> | By most I mean all major browsers (though I'm not sure if they do it by px) |
| 22:58 | <TabAtkins> | I think we leave that explicitly up to the browser. It relies a lot on user defaults and such. |
| 23:00 | <Hixie> | 16px really should be made the only allowed value for font-size:medium, imho |
| 23:00 | <Hixie> | http://www.hixie.ch/specs/css/font-size-ui/font-size-ui |
| 23:06 | <JonathanNeal> | neato Hixie |
| 23:19 | <JonathanNeal> | All right, so these are what I gather would be the requirements and recommendations of whatwg @ http://www.iecss.com/whatwg.css |
| 23:29 | <TabAtkins> | JonathanNeal: I'll go through and vet it in a bit. |
| 23:29 | <JonathanNeal> | TabAtkins, that would be killer awesome, I would totally appreciate it. |