| 00:06 | <heycam> | TabAtkins++ for calling your spec generator Bikeshed |
| 00:06 | <TabAtkins> | Heh, thank François for the suggestion. |
| 00:06 | <TabAtkins> | Remy, that is. |
| 01:45 | <zewt> | i sure hope "screen orientation" apis are only "lock the current orientation" and never "only allow landscape/portrait", which would be terrible |
| 02:35 | <zewt> | the fact that i can hover over a link and see a url, right click the link and see the same url, then click "copy link location" and get a totally different url: very not okay |
| 06:08 | <Hixie> | zewt: it's trivial for a web app to rotate itself, once the orientation is fixed |
| 06:09 | <Hixie> | zewt: so there's no practical difference to the user |
| 07:00 | <Ms2ger> | Hixie, nsAttrAndChildArray is... interesting |
| 07:04 | <Ms2ger> | It's basically an array of void*s where some of the void*s are actually used to store attributes rather than pointers |
| 11:09 | <annevk> | Hixie: https://www.w3.org/Bugs/Public/show_bug.cgi?id=22496 lacks a reply |
| 11:14 | <annevk> | http://dev.w3.org/cvsweb/2006/webapi/WebIDL/Overview.xml.diff?r1=1.642;r2=1.643;f=h Does getElementsByTagName() always return a new object? I thought we stopped doing that... |
| 11:30 | <zcorpan> | annevk: no, and it was pointed out and fixed in webidl, i think |
| 11:30 | <annevk> | ah I see |
| 11:31 | <annevk> | teaches me reading diffs rather than drafts |
| 12:41 | <hsivonen> | Oh great. Discussion about deprecating <blockquote> on public-html. |
| 12:42 | <odinho> | *shrug |
| 12:42 | <Ms2ger> | Oh great. public-html. |
| 12:45 | hsivonen | hopes that Robin makes it go away |
| 12:49 | <Ms2ger> | I'd rather keep it for the unhelpful people |
| 12:50 | <hsivonen> | oh. I meant making the proposal go away |
| 12:50 | <hsivonen> | making <blockquote> go away won't fly |
| 12:50 | hsivonen | got trolled into replying |
| 12:50 | <hsivonen> | I wonder how long it's been since my previous public-html post |
| 12:51 | <Ms2ger> | Sunday, 10 March 2013 |
| 12:51 | <hsivonen> | shorter time than I thought |
| 12:52 | Ms2ger | wonders if he ever posted there |
| 12:53 | <Ms2ger> | Doesn't appear that way |
| 12:53 | <hsivonen> | Ms2ger: congratulations |
| 12:53 | <Ms2ger> | Thanks :) |
| 13:02 | <jgraham> | Oh that was public-html |
| 13:02 | <jgraham> | makes sense I guess |
| 13:03 | <jgraham> | I should really setup up mail filters again |
| 13:11 | <SteveF> | hsivonen: how did you get trolled into replying? |
| 13:13 | <SteveF> | hsivonen: "hopes that Robin makes it go away" from the current discussion i don't see it happening, but do see possible changes in cite and blockquote definitions |
| 13:19 | <jgraham> | SteveF: That's an obviously silly thread |
| 13:20 | <SteveF> | jgraham: why? |
| 13:21 | <jgraham> | SteveF: Mostly for the reasons hsivonen already gave |
| 13:24 | <SteveF> | jgraham: the idea of obsoleting blockquote is a non starter but other stuff arising from the discussion is not |
| 13:29 | <jgraham> | I'm pretty sure I have seen all the same discussions before. I'm entirely sure that no change you make to the spec here will noticably improve the experience for consumers of HTML, since it won't be used or will at best be used sporadically |
| 13:30 | <SteveF> | jgraham: maybe |
| 13:31 | <SteveF> | jgraham: authors may benefit |
| 13:33 | <hsivonen> | SteveF: I read the initial post to the thread and replied even though I knew I shouldn't |
| 13:34 | <SteveF> | hsivonen: i don't think the people on the thread are trolls but hey thats just me |
| 13:35 | <hsivonen> | SteveF: oh, I'm not suggesting it got intentionally trolled |
| 13:35 | <SteveF> | hsivonen: ok |
| 13:36 | <hsivonen> | is there a term for involuntary trolling that has an effect similar to intentional trolling? |
| 13:37 | <hsivonen> | as in posting something that's from outside the Overton Window without intending to troll |
| 13:38 | <hsivonen> | SteveF: I'm glad we agree that obsoleting the element is a non-starter |
| 13:38 | <hsivonen> | I'll try to use expressions like "succumb to 386" in order to avoid the offense of "getting trolled" |
| 13:39 | <SteveF> | hsivonen: sure |
| 13:40 | <annevk> | hsivonen: it happens when you encounter people who missed a decade of debate, as with the XML 1.1 guy |
| 13:40 | <SteveF> | hsivonen: poeple can propose anything they like, but there are lots of hoops to jump through before stuff may happen |
| 13:40 | <Ms2ger> | "detractor performant" |
| 13:42 | <hsivonen> | Ms2ger: I believe "detractor performant" refers to suspected-intentional trolls without the word "troll" |
| 14:16 | GPHemsley | just closed a 13-year-old Mozilla bug reported by Hixie as WFM. |
| 14:17 | <Lachy> | GPHemsley, which bug? |
| 14:17 | <GPHemsley> | https://bugzilla.mozilla.org/show_bug.cgi?id=34822 |
| 14:18 | <GPHemsley> | And then there's this bug, which annevk said he would get to back in 2004: https://bugzilla.mozilla.org/show_bug.cgi?id=74263 |
| 14:20 | <GPHemsley> | resolved that WFM, too |
| 14:20 | <zewt> | Hixie: having a way for apps to say "landscape only" is going to mean lots of pages randomly doing that (in the same way that pages once resized windows to 640x480), which will be a horrible UX |
| 14:22 | <zewt> | i suspect people might be thinking of that as if it's analogous to mobile apps locking one orientation, but it's very different (having your orientation change out from underneath you normally only happens when you explicitly change apps; this would happen merely from browsing, and even hitting back a few times could spin back and forth between different orientations for each page you go through) |
| 14:25 | <zewt> | re "trivial to rotate itself: rotating the page itself isn't the same as forcing an orientation, since the browser and OS UIs won't be affected, which makes it much less intrusive (it's also not at all what you want, if you really *do* want that orientation) |
| 14:34 | <zewt> | gar @ pages that intercept f5 to try to keep you from refreshing |
| 15:20 | <annevk> | GPHemsley: only if people wanted it ;) |
| 15:20 | <GPHemsley> | ;) |
| 15:23 | <jgraham> | GPHemsley: trying to eliminate all the 5 digit bugs? |
| 15:23 | <GPHemsley> | Not exactly, though that might be worthwhile. |
| 15:23 | <GPHemsley> | I was just going through the list of open bugs that hadn't been touched in 5 years. |
| 15:24 | <GPHemsley> | And a few jumped out at me as being clearly obsolete. |
| 15:24 | <GPHemsley> | (And/or something that I had to the knowledge to assess.) |
| 15:27 | <GPHemsley> | jgraham: Incidentally, there are actually still a handful of open 4-digit bugs |
| 15:27 | <GPHemsley> | opened in 1999 |
| 15:28 | <GPHemsley> | but I don't think I'm qualified to assess any of them |
| 15:29 | <GPHemsley> | Oh, here's a good one: https://bugzilla.mozilla.org/show_bug.cgi?id=7954 |
| 15:29 | <GPHemsley> | "outstanding issues for full HTML 4.01 support" |
| 15:30 | <GPHemsley> | (though that one still gets updated every so often) |
| 15:32 | <GPHemsley> | oh, and 3 3-digit bugs |
| 15:33 | <GPHemsley> | two are on Bugzilla |
| 15:33 | <GPHemsley> | but the third is "implement inheritance of alignment attributes from columns (align, valign, char, charoff, (lang, dir)?) " |
| 15:33 | <Ms2ger> | Oh, 915 |
| 15:33 | <GPHemsley> | yeah, the infamous 915 |
| 15:34 | <Ms2ger> | It doesn't close easy |
| 15:35 | <GPHemsley> | no it doesn't |
| 15:36 | <GPHemsley> | the first attempt was 2 days after it was file, back in 1998 |
| 15:36 | <GPHemsley> | +d |
| 15:37 | <Ms2ger> | Should have a party next month |
| 15:38 | <Ms2ger> | Not many bugs make it to their 15th birthday |
| 15:39 | <GPHemsley> | heh |
| 15:39 | <GPHemsley> | 540 and 554 will get there first |
| 15:40 | <GPHemsley> | though it's possible 540 won't make it |
| 15:40 | <GPHemsley> | "I will create a new patch for this very soon." – dkl back in June |
| 15:47 | <gsnedders> | Well, Opera's CORE-1 got fixed a couple of years back. |
| 15:48 | <Ms2ger> | Our first bug got wontfixed |
| 15:49 | <jgraham> | Did DSK-1 ever get fixed? |
| 15:49 | <jgraham> | I don't know what it was |
| 15:49 | <gsnedders> | DSK-1 was "test" |
| 15:49 | <jgraham> | Oh |
| 15:49 | <gsnedders> | Not sure what status it is in. |
| 15:50 | <gsnedders> | No, sorry, it was "Test Test Test", a feature request, rejected. |
| 15:51 | <gsnedders> | DSK-2 is "test", resolved invalid. |
| 15:51 | <gsnedders> | AndDSK-3 is "TEST". |
| 15:51 | <gsnedders> | (also invalid) |
| 16:04 | <GPHemsley> | hmm... http://dev.w3.org/csswg/css-text/ is a 401 |
| 16:05 | <annevk> | wfm |
| 16:07 | <SimonSapin> | GPHemsley: it’s a pre-processor fail. The actual file contains the error message: https://dvcs.w3.org/hg/csswg/file/f7456b5a5ed0/css-text/Overview.html |
| 16:08 | <GPHemsley> | interesting |
| 16:10 | <SimonSapin> | GPHemsley: fixed |
| 16:10 | <GPHemsley> | thanks |
| 16:19 | <Ms2ger> | gsnedders, ah, those DSK people... |
| 16:22 | <annevk> | TIL: browsers don't care about empty domain labels |
| 16:37 | <Ms2ger> | Lachy: I must recommend a FxOS phone ;) |
| 16:46 | jgraham | thinks the temptation to call a class TestRunnerRunner might indicate something went wrong somewhere |
| 16:47 | <Ms2ger> | TestRunnerRunnerFactory? |
| 16:47 | <gsnedders> | jgraham: You approved a review of html5lib stuff with moduleFactoryFactory! |
| 16:48 | <jgraham> | Yeah well |
| 16:50 | <jgraham> | In this case TestRunners want to live in their own process and if they are forcibly killed it must be by the owner process. So I have a TestRunnerRunner that starts the TestRunners and kills them if they become unresponsive. But it doens't feel good, so I guess something should change |
| 16:52 | <gsnedders> | That seems sane. |
| 16:53 | <Ms2ger> | TestRunnerManager |
| 16:54 | <jgraham> | "manager" makes it sound better already |
| 16:54 | <Ms2ger> | TestManagerRunner |
| 16:55 | <jgraham> | Worse again :p |
| 16:55 | Ms2ger | switches some glasses around |
| 16:55 | <Ms2ger> | This? |
| 16:55 | Ms2ger | switches more |
| 16:55 | <Ms2ger> | Or this? |
| 16:59 | Ms2ger | finds http://hg.hoppipolla.co.uk/hgwebdir.cgi/domharness/ |
| 17:11 | <tobie> | jgraham: not sure what you're working on, but the runner and server prob need to be two different projects. |
| 17:14 | <Ms2ger> | tobie, he's got a job now ;) |
| 17:16 | <jgraham> | tobie: I am working on all the things that Mozilla need to run the tests |
| 17:16 | <jgraham> | Not neccessarily sequentially |
| 17:18 | <tobie> | :) |
| 17:33 | <Hixie> | zewt: i agree |
| 17:35 | <Hixie> | hsivonen: why are you even reading that list, sheesh |
| 17:43 | <annevk> | I checked in IDNA support for the host parser. I didn't do anything I wanted to do today though: https://www.w3.org/Bugs/Public/show_bug.cgi?id=22986 |
| 17:44 | <annevk> | Hmm, I meant everything there... but just as well |
| 17:54 | <annevk> | Hixie: replied |
| 17:54 | <Hixie> | url? |
| 17:56 | <annevk> | Hixie: https://www.w3.org/Bugs/Public/show_bug.cgi?id=22496 |
| 18:29 | <GPHemsley> | There is a legitimate bug entitled "Flour missing from pancake recipe" |
| 18:34 | GPHemsley | 's head explodes |
| 18:34 | <GPHemsley> | recursive definitions don't work well in plain English |
| 18:36 | <GPHemsley> | an inclusive ancestor can either be host-including or non-hosting including? |
| 18:36 | <GPHemsley> | +- |
| 18:36 | <GPHemsley> | an inclusive ancestor can either be host-including or non-hosting-including? |
| 18:36 | <GPHemsley> | argh |
| 18:37 | <GPHemsley> | an inclusive ancestor can either be host-including or non-host-including? |
| 18:38 | <GPHemsley> | meh, that's the wrong question to ask anyway |
| 18:42 | <GPHemsley> | yeah, I think adding "of an object's root's host" after "an inclusive ancestor" might clarify things a little bit |
| 18:42 | GPHemsley | just realized annevk isn't even here |
| 20:36 | <TabAtkins> | gsnedders: Yo, html5lib is throwing "Coercing non-XML name" warnings, but not telling me anything about the error. No line number, no name, nothing. |
| 20:36 | <annevk> | Hixie: I don't follow, recursive terminology around trees has been in standards since at least CSS 2.1 |
| 20:43 | <gsnedders> | TabAtkins: You want useful error messages? |
| 20:43 | <TabAtkins> | gsnedders: Yes. |
| 20:43 | <gsnedders> | Bah! |
| 20:43 | <gsnedders> | TabAtkins: GitHub issue, plz. |
| 20:43 | <TabAtkins> | kk |
| 20:47 | <Hixie> | annevk: responded |
| 21:00 | <annevk> | ta, will look later |
| 21:01 | <annevk> | I mean, that comment looks useful, will look into using it later |
| 21:05 | <Hixie> | heh k |
| 22:31 | <MikeSmith> | Hixie: question about document conformance and "script-supporting elements", per wording of the hgroup content model "One or more h1, h2, h3, h4, h5, h6, and script-supporting elements.", is <hgroup><script>...</script></hgroup> meant to be valid? |
| 22:32 | <Hixie> | yeah, the idea being that you might use the <script> to document.write() the h1 or whatever |
| 22:32 | <Hixie> | there's a lot of things where it's not clear exactly where the line should be drawn, that's one of them |
| 22:32 | <MikeSmith> | OK |
| 22:33 | <MikeSmith> | and same for dl, right? I can have <dl><dt>foo<script>..</script></dl> without <dd>, right? |
| 22:35 | <MikeSmith> | but not <dl><dt>foo</dl> |
| 22:48 | <Hixie> | MikeSmith: hm, let me check, i remember that <dl> was hard |
| 22:49 | <MikeSmith> | ok |
| 22:49 | <Hixie> | "Zero or more groups each consisting of one or more dt elements followed by one or more dd elements, optionally intermixed with script-supporting elements." |
| 22:49 | <Hixie> | so no, if you have a <dt> you have to have a <dd> |
| 22:49 | <Hixie> | but you can have <Script> and <template>s all over the place |
| 22:50 | <MikeSmith> | OK |
| 22:51 | <MikeSmith> | Hixie: that seems inconsistent with the <hgroup> case, where the <script> or <template> can essentially be used in place of an <h1>-<h6> |
| 22:51 | <Hixie> | <dl><script></script></dl> is fine |
| 22:52 | <MikeSmith> | OK |
| 23:00 | <MikeSmith> | Hixie: but I note that's a different model than what you had before you made the http://html5.org/r/8022 change, "Make <script> be accepted in most places <template> is now accepted" |
| 23:00 | <MikeSmith> | https://github.com/w3c/html/commit/e92161bbeb11d2ed1a2e3f5f45e63171f4102d83#L2L19501 |
| 23:01 | <MikeSmith> | prior to that, <dl><dt>foo</dt><template></template></dl> and <dl><template></template><dd></dd></dl> were valid |
| 23:03 | <Hixie> | MikeSmith: yeah, that was a conscious change |
| 23:03 | <MikeSmith> | ok, thanks, that's what I wanted to confirm |
| 23:04 | <MikeSmith> | I'll update the validator |
| 23:04 | <TabAtkins> | Hixie: What was the conscious behind that change? |
| 23:07 | <Hixie> | i basically ended up in a state where i was getting rid of any checks at all in <dl> |
| 23:07 | <Hixie> | and i didn't think that was useful to authors |
| 23:08 | <Hixie> | given the number of times i accidentally type <dd> instead of <dt> and so on |
| 23:11 | <TabAtkins> | Not any checks at all, just any checks *while there's a script or template around*. |
| 23:12 | <TabAtkins> | Because you could output anything from them. |
| 23:14 | <Hixie> | the <Script> doesn't have to be inside the <dl> for that argument to work |
| 23:15 | <Hixie> | so i settled on a compromise, whereby the validator checks for matching groups |
| 23:15 | <Hixie> | like i said above, there's a lot of things where it's not clear exactly where the line should be drawn |
| 23:17 | <TabAtkins> | The script does have to be in the <dl> for that argument to work *during parsing*. |
| 23:38 | <GPHemsley> | Hixie: Maybe my new comment makes it clearer? |
| 23:38 | GPHemsley | hopes he understands correctly |
| 23:40 | <GPHemsley> | d'oh |
| 23:40 | <GPHemsley> | I keep typing "ancestory" instead of "ancestor" |