2013-06-01 [17:01:18.0000] hey we let _you_ edit [17:01:20.0000] /me ducks [18:53:16.0000] Specs todo could probably use a good audit [19:02:19.0000] add it to the spec todo [19:04:17.0000] heh [19:05:04.0000] haha [00:39:00.0000] odinho_: can you look at https://bugzilla.mozilla.org/show_bug.cgi?id=869874 please [01:41:24.0000] good morning all [02:19:49.0000] is this complete list of http-equiv attribute posible values? http://msdn.microsoft.com/en-us/library/ie/ms537417(v=vs.85).aspx [02:28:25.0000] which is best way to cntrol how browsers cahce my web page? [04:05:49.0000] annevk: Yeah, I took a look at it, but I have not a full debugging setup since I've changed computers since moving to Desktop. [04:07:31.0000] I will do that when back at work again where I probably should set up a system again. [04:30:51.0000] what I don't understand about expire http header is this, someone visit my webpage and current date is after the expires date on my webpage so he gets fresh copy but then he visits again in the next day and he again takes fresh copy? or his browser knows that he has good copy on hdd? [05:29:52.0000] odinho: cool [08:27:29.0000] SimonSapin: Were you aware of http://greenbytes.de/tech/tc/datauri/ ? (It's from jreschke.) [08:27:42.0000] /me enjoys talking to people through logs they're unlikely to read. [09:07:50.0000] GPHemsley: annevk seems to read logs (hi there, man!) [09:18:28.0000] odinho: Yeah, but he's the only one I know of. [09:18:38.0000] /me raises hand [09:19:50.0000] good to know [09:21:09.0000] Help me out with phrasing here: " Once a MIME type has been extracted according to the syntax of its containing protocol, it must be parsed in order to obtain the information it contains. " [09:21:53.0000] What are you trying to accomplish? :) [09:22:55.0000] I want to tell anyone implementing the "parse a MIME type" algorithm that they're to handle any protocol-specific stuff (like URL encoding, or multi-line unescaping) before they run the algorithm, and that the only way to extract information from a MIME type is by running this algorithm [09:25:00.0000] Mm [09:26:44.0000] I'd think the first part should be obvious from the layering [09:27:27.0000] you'd think, but you wouldn't always be right ;) [09:37:02.0000] I'm always on, and I frequently read backlog. [09:37:19.0000] That's the same as log reading, only without the irritating log reading. [09:38:40.0000] guys did I talked to real Hixie yesterday? [09:39:54.0000] dekiss, identity on IRC, it is a challenge isn't it? [09:40:22.0000] yep :) [09:40:55.0000] he said he is he was admin here but never replied to my mail i sent him test email and asked for him to reply as I see he is really him :) [09:41:55.0000] he gets a lot of email :) [09:56:37.0000] odinho: Indeed. [09:56:48.0000] dekiss: If we tell you no, will you stop asking? [10:01:59.0000] Ms2ger: No tips? [10:02:43.0000] Can't come up with anything, I'm afraid [10:03:14.0000] GPHemsley im not sure [10:04:25.0000] dekiss, and yes, Hixie is Hixie [10:08:22.0000] have you edited some html css content or something else [11:49:18.0000] GPHemsley: for parameters, you can test parsing in other headers than content-type. For instance, Link: ; type=stylesheet; title=etc [11:49:39.0000] GPHemsley: or Content-Disposition: download; filename="..." [11:51:10.0000] er, attachment [11:51:40.0000] Refresh: ... [11:51:50.0000] Set-Cookie: [12:02:21.0000] Set-Cookie2 [12:02:23.0000] /me ducks [12:08:22.0000] zcorpan: I was wondering about that... are all HTTP headers parsed the same? Because this algorithm assumes it only applies specifically to MIME types. [12:12:13.0000] GPHemsley: i don't know, but it might be worth looking into [12:12:24.0000] perhaps [12:12:34.0000] it would make sense to parse them the same [12:13:14.0000] in some ways yes and in some ways no [12:13:29.0000] MIME types are (now) defined independent of HTTP [12:13:48.0000] and I don't know the extent of coordination between the two [12:13:57.0000] (since HTTP 1.1 was defined) [12:16:07.0000] i'm more interested in whether current implementations use the same parser for all headers [12:26:39.0000] I don't think they do [12:26:49.0000] but that's just my anecdotal impression [12:27:04.0000] (since I know I've look at MIME type-specific code) [12:27:07.0000] +ed [12:37:29.0000] /me wonders why IETF has so often chosen to abbreviate "parameter" as "parm" instead of "param"... [12:37:35.0000] it makes me hungry... [12:42:10.0000] Ms2ger: Is there a way to make anolis use a different timezone than UTC? [12:44:02.0000] GPHemsley, not at this point; patches welcome [12:44:18.0000] Where in the code might that go? [12:44:55.0000] anolislib/processes/sub.py [12:45:05.0000] Where it says time.gmtime() now :) [12:59:42.0000] should I wrap my website in article element? [12:59:55.0000] if I want to make in the left side logo and in right some picture or s;lideshow should I put invisible header element? [13:01:20.0000] in which element should I wrap my website? [13:02:01.0000] body [13:02:26.0000] :) [13:02:35.0000] other, [13:02:43.0000] so I can put my website in middle of screen [13:02:47.0000] div? or [13:04:18.0000] I cant put body in the middle, that way I must put every element in the middle separately [13:04:27.0000] and I will not have control over webwite [13:13:05.0000] Ms2ger: Not the best code, but there you go. [13:16:58.0000] why when I put h1 and h2 heading in hgroup element they are boh same size? [14:00:00.0000] Hixie: What was the difficulty in defining a "resource" again? [14:11:23.0000] should I give ids and classes to html5 semantic section element like header nav section article? [15:08:52.0000] /me grumbles something about bug reporters who don't accept bugmail [15:39:44.0000] is it ok to put articles in sections? 2013-06-02 [17:25:01.0000] Is there by any chance a list of CSS properties that use url() and related resource-requesting methods? [17:52:29.0000] margin: 0px; margin-left: 10px; is this mistake? [18:11:34.0000] dekiss: yes, "0" does not have unit. [18:12:16.0000] (0px still works, of course. Unitless is just shorter.) [18:13:13.0000] dekiss: not a mistake, no. top, bottom and right will be 0px, left 10px [18:15:16.0000] yeah [18:15:39.0000] assigning 10 same properties with different values is not mistake in css it seem so.. [18:17:07.0000] shorthands are equivalent to having declarations for their longhands at the same place, and multiple declarations with the same name are fine: the last valid one is used [18:17:08.0000] dekiss: the rules of specificity define what gets defined as "most important" [18:18:28.0000] yeah [18:19:42.0000] oh guys [18:19:47.0000] im so exited abotu my project :) [18:20:34.0000] im not sure if I am doing this right ill paste code in jsfiddle for you to see ok? [18:20:56.0000] http://pastebin.com/XaBUFhmV [18:21:08.0000] please guys what you think about this html code? [18:21:11.0000] is it good? [18:26:02.0000] SimonSapin: Do you read the logs from when you're offline? [18:26:25.0000] GPHemsley: rarely [18:26:30.0000] why? [18:26:39.0000] I posted something in here for you earlier [18:27:37.0000] http://greenbytes.de/tech/tc/datauri/ this? Yes, I’m aware of it [18:27:51.0000] ah, ok [18:28:12.0000] but haven’t looked into it in details, nor have done anything with data: for some time [18:28:19.0000] oh [18:28:25.0000] alright [18:29:47.0000] Not sure when I’ll get to it … if someone else wants to work on data:, feel free to do so [19:27:13.0000] Is Document restricted to HTML/XML documents, or is it present in any document the browser loads? [19:28:23.0000] And, by extension, then, is a browsing context restricted to HTML/XML documents, or does it apply to any document the browser loads? [01:15:33.0000] GPHemsley, does http://www.whatwg.org/html/#read-text answer your question? [06:55:02.0000] Ms2ger: Actually, http://www.whatwg.org/specs/web-apps/current-work/multipage/history.html#read-media is probably more relevant [06:55:10.0000] But thanks for the pointer [06:55:17.0000] Well, yes, all the subsections there :) [06:55:30.0000] this document is so big, I have no idea these sections even exist [06:55:50.0000] I should really just assume that if I have a question, it's answered in the spec somewhere [06:56:15.0000] (which is probably why Hixie suggested I just keep the one-page version open all the time) [06:57:31.0000] Yeah, the kitchen sink picture is more accurate than is given account for :) [06:58:33.0000] interestingly, in addition to finding out about new sections, I also find out about new places mimesniff is implicitly referenced [07:23:24.0000] Ms2ger: 404 http://ms2ger.freehostia.com/tests/html5/browsing-the-web/load-text-plain.html [07:23:41.0000] from http://www.whatwg.org/specs/web-apps/current-work/show-tests.html?section=read-text [07:23:42.0000] Oh dear [07:24:22.0000] Should be http://w3c-test.org/web-platform-tests/master/html/browsers/browsing-the-web/read-text/load-text-plain.html [07:24:28.0000] But looks like it's broken... [07:27:20.0000] I'll fix when github comes back up [07:34:27.0000] github is down again? [07:34:33.0000] how are they able to get away with that? [08:05:38.0000] GPHemsley, should be fixed now [08:40:20.0000] Ms2ger: Out of curiosity, did you check for others? [08:40:34.0000] No [08:40:44.0000] (and, of course, the other problem is that this 404s too: http://w3c-test.org/web-platform-tests/common/text-plain.txt ) [08:40:46.0000] I need to look through them all at some point [08:41:10.0000] Refresh, I should have fixed that [08:41:25.0000] ah [08:41:27.0000] yup [08:42:09.0000] hmm... [08:42:34.0000] the reason Safari fails this test (which is the reason it caught my eye in the first place) is because it doesn't get "text/plain" out of it [08:42:46.0000] assert_equals: expected (string) "text/plain" but got (undefined) undefined [08:43:35.0000] Chrome is the same [08:44:44.0000] and apparently IE8 doesn't even run the test properly [08:45:05.0000] IE10 gives 404 too [08:46:07.0000] oh, right, I was reading somewhere that document.contentType is Mozilla-only [08:46:18.0000] IE has .mimeType, but WebKit doesn't have either [08:46:23.0000] (from what I've read) [08:46:56.0000] though apparently Opera doesn't have a problem [08:47:57.0000] It's in the spec ;) [08:48:28.0000] is it? [08:48:36.0000] these tests seem to be a problem all over [08:48:46.0000] apparently nobody implements this stuff except Mozilla [08:49:01.0000] (the read-media tests fail even worse than read-text) [08:50:09.0000] IE8 fails simply because it doesn't like data: URLs and redirects straight to them [09:45:28.0000] I can't figure out where multipart/x-mixed-replace is defined... [09:45:43.0000] and/or what its purpose is [09:47:04.0000] There's several assumptions there... [09:50:46.0000] ? [09:52:08.0000] That it is defined [09:52:12.0000] That it has a purpose [09:56:03.0000] some definition in html... [10:23:45.0000] Ms2ger: I'll settle for a usecase or any indication as to why it is exists and is mentioned in the HTML spec :) [10:35:47.0000] GPHemsley: the html spec mentions mjpeg [10:36:38.0000] ah, that instance of "multipart/x-mixed-replace" isn't linked [10:36:49.0000] that's why I didn't find it [10:36:54.0000] /me didn't search manually [10:37:32.0000] zewt: Thanks, though! [10:42:40.0000] looks to be the only instance not linked, despite being the only indication as what "multipart/x-mixed-replace" actually is [10:47:07.0000] Ms2ger: It's further confusing because a MIME type registration card for multipart/x-mixed-replace is included in the spec [10:50:39.0000] perplexed by mail to www-dom [10:51:09.0000] someone @chromium.org who is 1: surprised that exceptions are caught during event invocation and 2: reading the TR of dom level 2 [11:44:08.0000] zewt, hrm, I think the default is to propagate exceptions, actually... [13:49:18.0000] why IE didn't implemented transition in IE 9? [13:49:34.0000] I find this hard to understand [15:20:50.0000] ms2ger (this is a test of the Ms2ger Reads Logs system): i don't think "spec algorithms propagate exceptions by default" makes sense, since that would leave the precise behavior underspecified [15:21:14.0000] If this results in an exception, then abort the overall structured clone algorithm, with that exception being passed through to the caller." [15:21:21.0000] thanks irc client for removing the first half of my line [15:21:29.0000] in many places html says things like ^ [16:02:21.0000] is it dangerious to use new html5 elements? for security [16:02:43.0000] we don't really call it html5 anymore, it's just html [16:03:21.0000] and to answer your question: in general, no, though specific elements may have security concerns you should be aware of [16:03:24.0000] like iframe [16:05:52.0000] hey Hixie thanks for help [16:06:38.0000] when html 5 will be w3c recomendatiton (w3c implements it like html5.) [16:06:52.0000] and why it implements it like html5.. but thats another subject :) [16:07:10.0000] don't ask me, i don't see much point in the w3c's work on html [16:07:22.0000] I plan to use drag drop and content editable and such for my new website [16:07:39.0000] you work together on the spec? [16:07:56.0000] no, they just copy what we do and then screw it up [16:08:03.0000] hah :) [16:08:09.0000] (intentionally sometimes, unintentionally other times) [16:08:14.0000] I tell kids you work together on it ^^ :D [16:08:19.0000] (you think he's joking) [16:08:26.0000] zewt nope :D [16:08:51.0000] hm [16:09:36.0000] Hixie I have though about this subject alot honestly, and I think that w3c had other plans for the web then browser vendors [16:09:57.0000] I think they wanted to make direction for the web in where it will go in the future [16:10:10.0000] it's a mistake to assign to the w3c a single desire or direction [16:10:22.0000] though I never have read xhtml specs entirely [16:10:30.0000] as a venue for multiple companies to discuss matters, there are many opinions [16:14:07.0000] also the w3c can have all the "plans" for the web they want, they can also want a singing unicorn [16:15:56.0000] they are good people [16:16:10.0000] they mean good for web and people, its just different people different opinions [16:16:41.0000] everyone means well, that doesn't go very far :) [16:17:05.0000] not sure what you mean by "good people" in this context [16:17:12.0000] if yo uask me, I want web to be free and opened to as much peopel as posible -(peopel with assistive technology most), [16:17:27.0000] i have definitely met people through the w3c who were quite happy to put their corporation's interests ahead of the web's [16:17:36.0000] which doesn't seem "good" to me [16:17:52.0000] q.v. the w3c working on drm now [16:17:54.0000] but that is only way web will go I think it will develop a lot in the future as the demand for it is enormous [16:18:08.0000] Hixie hm [16:18:17.0000] yeah there are such people everywhere [16:18:28.0000] when I say 'they' I refer to Tim :0 [16:18:46.0000] and people liek him, I beleive they hold key positions, but ye I am against paid membership [16:19:04.0000] its only around 2000 $ yearly for my country for 1 year membership [16:19:20.0000] not clear that tim does anything for the web these days, unless you count the rdf stuff, and i'm not even sure he's much involved in that anymore [16:19:22.0000] that means almost anyone can get in [16:19:56.0000] if you ask me I think there should not be paid membership, but only membership with some work done in the past, people with titles etc prizes [16:20:00.0000] not money [16:21:07.0000] Hixie I think its time for new ceo of w3c [16:21:26.0000] or he must take work in his hands [16:21:31.0000] and fix things [16:22:17.0000] I think web lacks a lot of legislation [16:22:51.0000] in my country for example they only added few laws that include web in 2-3 years ago [16:23:01.0000] and they have no idea how to regulate other stuff that is on the web [16:23:21.0000] first thing msut be done is forbid adult content from web at all liek chinese did [16:24:10.0000] sounds uncontroversial [16:24:13.0000] here I watch 12 year old kids go on some "web portal" some bad web site and see - News, Poliics, Economy, SEX & EROTICS, sport, culture, showbiz.. [16:24:30.0000] I mean cmon.. every so called web portal has sex and erotics section these days [16:24:33.0000] at least in my country [16:24:45.0000] and next to the website there are ads from t-mobile and such.. [16:25:09.0000] second thing is microsoft and viruses [16:25:16.0000] and hacks [16:25:19.0000] ddos third [16:25:25.0000] etc [16:25:51.0000] imo icann iana ietf w3c wahtwg should come in one body and regulate this stuff asap [16:26:06.0000] add soem security work with countries to make legislations [16:26:17.0000] ... standards bodies != governments ... [16:27:02.0000] zewt they must colaborate [16:27:14.0000] (and the idea of the w3c influencing legislation is a bit terrifying) [16:27:17.0000] to make web good place [16:27:46.0000] not influencing but workign together with governments to make laws [16:27:56.0000] working together != influence [16:28:09.0000] well no working together == influence [16:28:21.0000] someone with clear mind must forbid adult and bizzare content from web [16:28:26.0000] and all countries must apply that [16:28:36.0000] yo ucan find literally anything these days on the web [16:28:50.0000] and web is given to kids of 12 years or younger and they can easy see this content [16:28:57.0000] just write it on google and boom you have it.. [16:29:05.0000] that is very very bad for the world [16:29:08.0000] lolwut [16:29:12.0000] TabAtkins: don't ask me [16:29:54.0000] someone must stand up and say from tomorow there is no adult content and bizzare content on the web and content is censored [16:30:23.0000] Lots of people stand up and say that today. They're all stupid, wrong, and don't get anything done besides wasting money. [16:30:39.0000] otherwise mee too will keep my kid away from the web, because I know what he can see easy on the web.. [16:30:54.0000] TabAtkins why you want bizzare content on the web? [16:31:06.0000] like videos of executions [16:31:25.0000] more bizzare is wehn I see these things on cnn but thats another subject [16:31:53.0000] TabAtkins chinese did something.. [16:32:23.0000] That's... not a model to follow. Massive, government-run censorship is a terrible, terrible idea, and you're going back on ignore now. [16:32:44.0000] someone must decide what web is and where it should go and make it go that way, and I think web can go in thousand directions other then atracting visitors with videos of executioning people [16:32:58.0000] and adult content and trading pirated material and getting money for that [16:33:00.0000] And if you actually think Chinese people can't get access to porn and "bizarre" things, you're extremely naive. [16:33:27.0000] TabAtkins im not sure what chinese people can or canot view [16:33:38.0000] but there must be some change on the top level for the web [16:33:43.0000] this got out of control already [16:33:52.0000] TabAtkins and ok I dont care if you put me on ignore :D [16:33:54.0000] indeed, the internet badly needs more pornography [16:34:02.0000] i think the w3c needs to see to that [16:34:09.0000] zewt [16:34:11.0000] hah nvm [16:34:15.0000] I disagree with you man [16:34:28.0000] but i wont put you on ignore coz im not frustrated nerd :D [16:34:43.0000] I respect other opinions as I value people that I sepak to [16:34:51.0000] al have points [16:34:55.0000] zewt tell me your point [16:35:06.0000] Unless the other people are "frustrated nerds", apparently. [16:35:35.0000] TabAtkins: I think your ignore feature is broken 2013-06-03 [17:30:04.0000] /me wonders what to do with http://wiki.whatwg.org/wiki/AAFC_Metadata_Application_Profile [23:41:47.0000] zewt, test succeeded :) [23:42:35.0000] The issue is that checking for exceptions at every step is terribly error-prone [23:54:12.0000] this dekiss guy is adding a lot of noise [00:08:30.0000] can we make anolis warn about xref-able elements without a title that don't get xreffed? [00:12:56.0000] Actually, you can make it error [00:13:35.0000] --use-strict [01:47:58.0000] argh [01:58:02.0000] annevk: ping [01:58:13.0000] smaug____: whoa, why are you awake? [01:58:36.0000] (someone called me around 9am and tried to sell something) [01:59:05.0000] and hey, it is noon [01:59:13.0000] smaug____, exactly :) [01:59:21.0000] annevk: https://bugzilla.mozilla.org/show_bug.cgi?id=780953 [01:59:38.0000] do you recall if that has been discussed somewhere [01:59:42.0000] whatwg or w3 [02:00:21.0000] smaug____: hasn't been discussed [02:00:43.0000] smaug____: pretty sure anyway, I recommend raising on WHATWG [02:01:46.0000] I don't think OS-level changes have had much consideration in general [02:02:45.0000] smaug____: this has been discussed in sysapps [02:03:21.0000] darobin: well, needs to happen in WhatWG, at least partially [02:03:27.0000] smaug____: https://www.w3.org/Bugs/Public/show_bug.cgi?id=21290 [02:03:34.0000] since the change requires Navigator to inherit EventTarget [02:03:43.0000] ah [02:03:46.0000] thanks [02:03:58.0000] it needs to happen in the HTML spec for sure I would say [02:04:01.0000] uh, could I change the product [02:04:05.0000] who does it, I don't care [02:04:23.0000] if you want to put it on the WHATWG's plate, please don't change the product, but instead clone it [02:04:41.0000] I reckon it'll happen anyway, it's useful [02:06:02.0000] yeah, it is useful [02:06:07.0000] and simple thing [02:06:16.0000] indeed [02:07:40.0000] MikeSmith: https://dvcs.w3.org/hg/csswg/rev/20c14a9da2c1 gives me 500 [02:10:36.0000] Ms2ger: thanks. however, it gives an error for :-( [02:10:47.0000] Lovely [02:11:54.0000] /me puts that on his todo list [02:13:38.0000] probably shouldn't be an xref element [02:14:03.0000] smaug____: it's already filed against WHATWG, darobin referenced the wrong bug [02:14:20.0000] smaug____: https://www.w3.org/Bugs/Public/show_bug.cgi?id=21289 [02:14:33.0000] Yeah, I think it's based on Bert's xref elements [02:15:02.0000] i'd like it if wasn't an xref element, too [02:15:02.0000] Ms2ger: should move anolis to GitHub [02:15:12.0000] yeah, we should not have be xref [02:15:18.0000] or [02:15:38.0000] why not ? [02:15:49.0000] annevk, you know I'm not much of a git fan :) [02:16:18.0000] zcorpan: is for algorithm flags and they're typically not xreffed [02:16:29.0000] Ms2ger: that's a losing proposition [02:17:12.0000] Some people like to think so [02:18:14.0000] /me didn't reference the wrong bug, just the bug he knew of (which is also the correct one ;) [02:18:32.0000] darobin, not in this channel ;) [02:19:01.0000] zcorpan: are you still getting 500 on dvcs.w3? it seems to be back up here [02:19:20.0000] darobin: ah, works now [02:20:26.0000] /me likes being an xref element, strangely [02:22:20.0000] annevk: ah, thanks [02:22:50.0000] darobin: instead of ? [02:22:55.0000] darobin: that might actually be nice [02:23:00.0000] Ms2ger: ^^ [02:23:26.0000] File a bug :) [02:23:55.0000] annevk: that's what ReSpec uses [02:24:01.0000] it makes more sense I find [02:24:14.0000] it's even a correct use of HTML!!!!1 [02:37:11.0000] Ms2ger: https://bitbucket.org/ms2ger/anolis/pull-request/6/remove-and-from-the-list-of-xref-elements/diff [02:38:00.0000] darobin: Don't be silly, it's not possible to use HTML correctly. [02:41:42.0000] jgraham: I was waiting for that one :) [02:42:52.0000] We have a great future as a comedy double act. [02:44:01.0000] Dry British wit... And darobin? [02:48:04.0000] Drunk French humour? [02:48:35.0000] Sounds about right [02:48:51.0000] /me passes darobin a glass of wine [02:51:27.0000] should i add in the same PR? [02:51:58.0000] /me puts on a béret and camembert sandwich [02:52:07.0000] i guess would need more work than just adding it to the list to not screw up [02:54:34.0000] Sure [02:54:40.0000] (Did you run tests?) [02:55:02.0000] no, i didn't [02:55:34.0000] FAILED (failures=1) [02:57:51.0000] FAIL: test_tests/toc-basic.src.html, however that seems unaffected by the change [03:06:31.0000] zcorpan: I been getting reports of intermittent 500s from dvcs.w3.org but can't see any problem on the server side [03:06:47.0000] MikeSmith: ok [03:07:06.0000] the only troubleshooting solution that's worked in the past is for use to just restart apache there [03:07:47.0000] but in the past the only time we had 500s was when the server was completely wedged and consistently unresponsive [03:36:05.0000] Unsure if this is the right place to ask, but an element on my page (an "a" element with an "img" inside) simply isn't being focused in the Tab cycle, how can I fix this? I've tried adding a "tabindex" value, but it doesn't work, instead the textbox just next to it in the markup always gets focused first, and then everything else on the page after that. [03:37:45.0000] joesavage: perhaps not the best forum for such questions, but does your have an href with something useful in it? [03:37:49.0000] if not, it's not focusable [03:39:41.0000] It's just linking to "/", but I've tried a bunch of other values in my troubleshooting process, so I guess it's not this. [03:40:39.0000] Interestingly, if I add some text after the image (i.e. test), then it focuses first. [03:42:43.0000] I also notice that taking the "float: left;" away from it in the CSS makes it focus correctly too. Very odd. [04:58:01.0000] fwiw: http://annevankesteren.nl/2013/06/london-tag (not comprehensive) [05:03:35.0000] annevk: "Getting there from where we are today is though" [05:03:56.0000] Sounds right [05:04:37.0000] Also, I'm not sure why the "champions" model is a good thing [05:05:06.0000] jgraham: not sure what you meant by quoting that? [05:05:09.0000] tough? [05:06:05.0000] You either misspelt or just missed a word [05:06:26.0000] ooh [05:06:50.0000] Damn English [05:07:17.0000] jgraham: thanks [05:07:39.0000] jgraham: not sure if champions is good, but is there anything else that scales? [05:11:50.0000] annevk: why a separate GitHub org for W3C tag? [05:12:20.0000] See discussion in this channel a few days ago [05:14:01.0000] tobie: easier to setup repos [05:14:33.0000] annevk: how so? [05:15:30.0000] I mean, if TAG members can't get admin rights on gh.com/w3c... [05:15:48.0000] we'd need to fix that. :) [05:16:41.0000] annevk: feels like a missed opportunity for visibility and transparency. [05:17:17.0000] /me thinks it doesn't make that much difference [05:17:20.0000] that sounds very opaque [05:34:50.0000] annevk: what does? [06:39:54.0000] Ms2ger: not explicitly saying what to do with exceptions is terribly not saying what to do--if you don't know exactly which steps in an algorithm might throw an exception, you can't see all of the code paths [06:40:26.0000] and if the spec author doesn't know all of those places, that's error prone as well [07:11:44.0000] tobie: what you said about the TAG not sharing a GitHub account [07:12:15.0000] what's opaque about that? [07:14:30.0000] it seems important our work is on GitHub, it doesn't matter much where afaict [07:15:01.0000] and in fact, for people interested in what the TAG works on, a single page seems a lot clearer [07:15:52.0000] No one cares about that though [07:16:04.0000] Or no one should [07:16:25.0000] They should care about "are good web technologies being standardised and shipped" [07:18:04.0000] (a similar argument could be made that the CSS WG needs it own repos, and the HTML WG and the WebApps WG, and the Web Performance working group and…) [07:18:14.0000] s/repos/account/ [07:20:32.0000] I think it helps to have some granularity so you know who is accountable [07:20:57.0000] That we have a gazillion groups doing APIs these days is indeed not helping [07:21:36.0000] But organizations don't do work; individuals do. [07:21:43.0000] I don't understand your argument [07:22:17.0000] Even if we only had HTML/WebApps/CSS it would still be strictly worse to split up all their stuff [07:22:46.0000] If you want namespacing that is possible by naming the repos groupname-whatever [07:22:56.0000] /me doesn't really know what's going on, being on holiday and all [07:25:45.0000] (c.f. Conway's Law) [07:39:40.0000] jgraham: I see it more as being about size [07:39:47.0000] jgraham: if it's too large it becomes hard to follow [07:40:16.0000] jgraham: e.g. that's why some specs have their own twitter account, or why we have separate specs to begin with [07:40:25.0000] jgraham: even though they're all representing the same platform [07:40:41.0000] I thought Conway's law was if you arrange black squares on a grid in a special way, you get spaceships [07:41:50.0000] want [07:42:03.0000] sounds accurate [07:42:07.0000] Fun fact: the game of life is Turing-complete [07:42:53.0000] /me found that part of that textbook more interesting than the part about RDF [07:43:00.0000] Fun fact: So is W3C Process [07:44:04.0000] Dammit I think I just nerd-sniped myself. Now I wonder if you can abuse W3C Process to perform computations [07:44:36.0000] Implementing a Turing machine in Life is unfortunately slightly less practical than implementing one in C++ templates [07:44:42.0000] jgraham: ROFL [07:45:02.0000] well, we have loops for sure [07:45:17.0000] I don't think you need the full process, Rec track might be T-complete [07:45:54.0000] /me wonders what textbook Ms2ger had that featured both the GoL and RDF [07:46:06.0000] someone must've been smoking something pretty damn good at the textbook factory [07:46:56.0000] darobin, Rich's Automata, Computability and Complexity [07:47:10.0000] It also featured claims that HTML is a regular language [07:47:57.0000] I guess it was more than just smoking then [07:48:40.0000] mmm, that book ain't cheap, too [07:48:50.0000] Indeed it isn't [07:51:49.0000] I reckon that if you treat drafts as storage space then given LC is a conditional loop you have Turing-completeness [07:53:04.0000] Unforunately it doesn't define the computation engine itself, merely everything else. [07:54:51.0000] :) [09:33:28.0000] So, just how many testing-related IRC channels does W3C need? [09:34:01.0000] 42 [09:35:04.0000] That would explain a lot [09:36:29.0000] jgraham: #htmlt, #test, #testing what else? [09:47:04.0000] tobie: I have no idea. Until today I didn't know that #test existed. [09:47:25.0000] me neither [09:47:37.0000] what's #test for? [09:47:40.0000] jgraham: with the move to GitHub, it feels like some reconciling would be useful. [09:49:16.0000] jgraham: including mailing lists [09:49:21.0000] tracking that here: [09:49:35.0000] https://github.com/w3c/ttwf-docs/issues/7 [09:49:50.0000] MikeSmith: used by the testing tf [09:50:06.0000] ok [09:50:30.0000] #css-test [09:56:20.0000] Ms2ger: ty [09:56:30.0000] Np [11:17:00.0000] tobie: Anyway I would happily close them all apart from #testing [11:17:53.0000] Mind you, I wouldn't spin up a bunch of closed membership task forces either [11:23:40.0000] jgraham: neither would I, but corps are touchy when it comes to $$$. [11:32:23.0000] how is html language made? who is making it? html editors? [11:32:49.0000] I have hard time understanding it, how that mailing lists work etc. [11:37:55.0000] Unrelatedly, is there some reason that people are pushing everything to happen at "end of microtask" rather than just using normal tasks? [11:38:18.0000] (for small values of "everything" that includes a bunch of new stuff) [11:53:16.0000] So we're at the point where we're doing d3e conf calls without anything to talk about again? [12:00:05.0000] Someone else is doing D3E calls [12:12:36.0000] Also, "we were forced to do bad thing $X because of the $$$" seems to be the standard W3C excuse at the moment [12:18:08.0000] is there any chance that Goolge will be trying to standardise DOM improvements made in Dart? [12:19:00.0000] I mean things like query() or events as streams [12:19:21.0000] The idea behind those was that there's no legacy dart content, so breaking stuff is free, right? [12:19:33.0000] Breaking stuff on the actual web is far from free [12:21:25.0000] I'm not aware of Dart internals, but it looks like most of the improvements are implemented as wrappers on top of standard DOM [12:21:29.0000] http://www.dartlang.org/articles/improving-the-dom/ [12:22:29.0000] jgraham - w3cexcuses.tumblr.com ? [12:23:13.0000] jarek: Nor am I, but I doubt they want to reimplement the C++ bits [12:24:46.0000] I mean, the DOM still has to exist. So having something fundamentally different exposed to Dart and JS seems hard [12:25:00.0000] why NodeList and DOMTokenList don't inherit from Array.prototype? Does the spec prohibit this explicitly? [12:25:16.0000] NodeList explicitly inherits from Array.prototype in the spec [12:25:34.0000] We're waiting on Blink to try implementing it [12:25:46.0000] Last time we tried, a lot of Google sites broke [12:26:59.0000] And having wrappers on top of the DOM only increases the API surface for little gain, making the platform harder to learn and to maintain [12:27:39.0000] Ms2ger: Wait, what? Why does the spec say something that's known not to work? [12:28:01.0000] TabAtkins: yo, if you're around, if you could reply to the www-dom thread with MM that'd be grand [12:28:24.0000] jgraham, because Google could maybe have fixed its site [12:28:25.0000] s [12:28:33.0000] It should at best say Note: It has been suggested that (foo). Experiments with (foo) showed a significant web compat impact. [12:28:58.0000] Ms2ger: If it broke Google sites it seems highly likely it also broke other sites [12:29:18.0000] Perhaps [12:29:23.0000] I haven't looked into it closely [12:29:43.0000] and expecting Google to spend resources fixing its sites to make some blink people happy seems unreasonably optimistic [12:30:10.0000] In any case I don't think the spec should be left in a state where it requires knwon-broken behaviour [12:30:13.0000] *known [12:41:26.0000] jgraham, notice added [12:43:13.0000] Ms2ger: Thanks [13:37:03.0000] Hixie: Ah, one catch you forgot to mention about keeping the whole spec open: I have to reload it all when changes are made! ;) [14:52:04.0000] good morning, Whatwg! [16:18:18.0000] can I send informations from server to client every 1 min while web page is open? websockets I think is used for this? is that ok if I use it or which are my options? [16:25:33.0000] dekiss: for web authoring questions you probably want another forum, like #html5 or some such [16:25:51.0000] dekiss: here you're more likely to get snarky comments and venting than advice :-) [16:25:54.0000] can I restrict some web page so not all can see it? and if yes how please? [16:26:27.0000] should I check for origin or? [16:26:48.0000] Hixie ah ok man sry :) [16:27:27.0000] I never understood who make html and how it's made and this mailing system [16:27:59.0000] i'm not saying you can't ask here, just that if you want answers, this isn't hte best place for you :-) [16:28:06.0000] at least, if you want useful answers [16:28:19.0000] ok thanks [16:28:34.0000] man who makes HTML specification? [16:28:39.0000] you? ^^ [16:28:43.0000] editors? [16:29:34.0000] i write the text [16:29:40.0000] if that's what you mean [16:32:36.0000] Hixie: I honestly have a difficulty responding to . There is a word "associate" in the algorithm that I don't understand. What is the expected result of doing an "associate"? [16:33:10.0000] Hixie: is it that both sides will immediately know that they are "associated"? [16:33:21.0000] I ment who is making the web these days? [16:33:29.0000] HTML standard [16:33:43.0000] Hixie: in such case, I just don't know how to implement that [16:35:03.0000] ap: is there any observable way to tell if it's instant or not? [16:35:26.0000] dekiss: if you mean http://whatwg.org/html, then i write the text, based on feedback here and on the mailing list, etc [16:36:10.0000] thanks that was what I ment [16:36:56.0000] Hixie one more quetion, will menu be disbanded? [16:37:04.0000] and rarely used elements? [16:37:05.0000] menu? [16:37:11.0000] Hixie: unsure - the string "associat" is not found elsewhere in text (other than as "association list", which I think is different). So the effect of "association" is implicit and thus difficult for me to verify [16:38:04.0000] yes "menu" element [16:38:48.0000] hm I am not sure I tihink I saw in w3c.org spec that it was not used much and is considered to be disbanded [16:38:53.0000] also hgroup [16:38:54.0000] ap: it means "let them be entangled" [16:39:11.0000] dekiss: and
are new, i have no plans to remove them [16:39:23.0000]
in particular is used in over 100,000,000 pages already [16:39:46.0000] Hixie: the spec defines "entangle" through "associate", so explaining "associate" through "entangle" doesn [16:39:46.0000] t [16:39:48.0000] help [16:40:19.0000] ap: elsewhere in the spec there are algorithms that refer to whether a port is entangled with another or not, and that refer to the entangled port. [16:40:29.0000] ap: this algorithm is attempting to create that connection. [16:40:34.0000] Hixie: or is "entagle" only called on ports that have not been shipped yet? [16:41:22.0000] grr stumbling over new Event() because of the bad option defaults [16:42:44.0000] Hixie: so there is the clone algotithm: [16:42:46.0000] If the original port is entangled with another port, then run these substeps: [16:42:47.0000] ap: entangling of a port is something that happens upon its creation, and when its entangled port is cloned (when you clone a port A into a new port B, where A and C are entangled, A and C are disentangled and B and C are entangled) [16:42:51.0000] Let the remote port be the port with which the original port is entangled. [16:43:58.0000] Hixie: how can one detect if the original port is entangled? maybe another side has already initiated disentanglement, but it's still unknown to this side [16:44:46.0000] Hixie: e.g. ports A and B are entangled, and are in separate processes already. Scripts call clone() on them at the same time [16:44:56.0000] ap: you have to act as if it's instantaneous. Suppose you have A-B entangled, and you clone A to A' and B to B' at the "same time". You should end up with A'-B' entangled. [16:45:25.0000] Hixie: I doubt that there are implementations that do that [16:45:41.0000] Hixie: WebKit's certainly doesn't seems like it does [16:46:33.0000] ap: does webkit support cloning ports at all yet? [16:48:32.0000] Hixie: is this what you do when passing a port via postMessage? yes, that works (hopefully correctly when workers are not involved, and kind of maybe when they are) [16:49:04.0000] ap: well if it doesn't work i guess you have a bug :-) [16:50:55.0000] Hixie: it is of course something we can think of ourselves eventually, but it would be helpful if someone could think about whether the spec is implementable without unacceptable performance problems [16:51:12.0000] Hixie: such as pausing all processes to synchronize [16:52:27.0000] Hixie: anyway, I think that you answered bug 21416 - in that it's supposed to behave as if there was a cross-process mutex around all MessagePort operations [16:52:51.0000] ap: this doesn't require any synchronisation [16:52:53.0000] drawing a blank for some reason: any examples of APIs that dispatch events, other than really weirdo things (in particular other than element.click) [16:52:57.0000] er [16:53:01.0000] that dispatch events synchronously [16:53:26.0000] zewt: execCommand :-p [16:53:28.0000] ap: just make sure you forward all your messages appropriately (and essentially leave a forwarding address) [16:53:36.0000] ap: i'll bundle that in with weirdo things :P [16:53:51.0000] zewt: seach for "fire a " in the html spec [16:54:04.0000] that'll have hundreds of hits, most of which are async [16:54:28.0000] oh synchronously [16:54:28.0000] sorry, I mean JS entry points that fire events synchronously to script [16:54:29.0000] my bad [16:54:38.0000] dispatchEvent() is the main one [16:55:04.0000] Hixie: no, I understand that leaving a forwarding address is the next step. It's just that then I need to rip apart most of the other algorithms in the spec that don't explicitly account for that (most importantly those that explain object lifetime) [16:55:05.0000] yeah (that's the one I was looking to compare to, actually) [16:56:24.0000] ap: like what? i'm happy to make the spec clearer, i just don't understand what the problem is here [16:56:55.0000] (and I don't want to use click() as an example, there's so much horribly wrong with onclick it makes a bad comparison) [16:57:27.0000] Hixie: I _think_ that for every sentence that has the word "entangled" in it, I will need to think about whether it's actually "entangled or a forwarder to entangled" [16:58:05.0000] Hixie: and devise a protocol for dropping the forwarder privileges eventually [16:58:50.0000] ap: sure, that's like for every object that mentioned HTMLElement objects, you have to think about whether it means the C++-side of the object or the C++-side plus its JS-engine wrapper, or whatever is going on in the rendering engine. Figuring out these things is kind of your job. :-) [16:58:52.0000] Hixie: because when I have process A and process B, and a port gets shipped across the boundary and then back, we probably don't want to keep doing two cross-process messages any more [16:59:01.0000] ap: sure [16:59:37.0000] ap: doing cross-process IPC with pipes whose endpoints can be moved around the world is definitely non-trivial [16:59:43.0000] ap: i don't think anyone would argue otherwise :-) [16:59:58.0000] ap: our goal here is to try to make it as trivial as possible for Web authors, but that means eating a lot of pain 2013-06-04 [17:00:26.0000] Hixie: perhaps the spec would be easier to follow if the lifetime algorithms were just removed from it [17:00:43.0000] Hixie: as they will just be so different in practice [17:01:02.0000] well we need to define the detectable aspects of lifetime [17:01:05.0000] that's all the spec is trying to define [17:01:17.0000] obviously the actual implementations will likely be different [17:01:19.0000] Hixie: what are the detectable aspects? [17:01:26.0000] e.g. when does the worker stop running [17:01:27.0000] Hixie: usually the goal that there are none :) [17:01:31.0000] Hixie ok thanks for info man [17:02:00.0000] annevk: Which thread is this? Have I already responded to it? [17:02:09.0000] Hixie assistive technology devices can interpret javascript and dom? [17:02:39.0000] dekiss: in practice, or in theory? [17:02:45.0000] Hixie: is there a way to explain the shutdown procedure without the algorithms? Because they seem unlikely to help implementations converge in this case [17:02:51.0000] um [17:03:00.0000] both [17:03:24.0000] Hixie: people will just have another barrier trying to understand what the algorithm means in spec's synchronous world, and then implement differently anyway [17:03:51.0000] ap: i tried various ways to define it, but what's in the spec is the simplest, least confusing, most declarative manner i was able to come up with [17:04:00.0000] Hixie: ok [17:04:08.0000] ap: i'm very open to alternative phrasings if you can come up with any though, they're pretty horrific [17:04:32.0000] ap: definitely not a part of the spec i would point to when going up for promotion, shall we say [17:05:47.0000] ap: see the diff i just posted to the bug (https://www.w3.org/Bugs/Public/show_bug.cgi?id=21416), in case that helps... not sure what else to add [17:06:41.0000] Hixie: I'll see what I can do if I get a chance to work on channel messaging again. The last time, I had a hard time comparing the spec and the existing WebKit implementation to understand if it was doing something at least related to what's correct [17:07:18.0000] Hixie: I think that the change you made is helpful [17:08:08.0000] ap: cool, thanks (on both fronts) [17:08:24.0000] thank you! [17:08:36.0000] ap: totally agreed that this isn't easy to think about. It's probably the second most complicated part of web platform i've worked on [17:08:54.0000] Hixie? [17:08:54.0000] most complicated being xbl2, if we can count that as part of the "web platform" :-| [17:09:20.0000] dekiss: sorry, had to help ap first, he's an implementor, they're my peeps :-P to answe you, yes, ATs can be part of browsers. [17:12:43.0000] ok thanks man [17:12:46.0000] np [17:23:05.0000] ap: fwiw, safari seems to handle this test ok: http://www.hixie.ch/tests/adhoc/dom/workers/002.html [17:23:40.0000] ap: what i do there is create two workers, then a message channel, and i pass one side of each message channel to each worker at the same time, then wait a second, then have them send their parts back, then i send a message from one side to the other [17:24:37.0000] (Firefox doesn't seem to support MessageChannel, sadly) [17:27:09.0000] bbl [17:32:36.0000] Hixie: I think that have an actual mutex for dedicated workers (unsure if we lock it in all the right places, or what we do with shared ones) [17:33:19.0000] as shared ones were only out of process in chromium port [17:37:58.0000] it seems like there's a gap between what the spec wants to happen and how to actually do it in practice--i wonder if a small sample implementation would help narrow that (on top of unix pipes or something) [17:58:34.0000] (for that matter, I don't know if it can be sanely implemented with simple pipes, or if it wants some form of shm) [18:02:14.0000] heh cute, apparently mousewheel scrolling in chrome closes the context menu [18:02:46.0000] handy way when I right-click by accident, want to close the menu, but ... can't find anywhere to click that won't have side-effects [18:03:20.0000] (and more natural than hitting esc) [20:16:19.0000] Why does the meta tag not allow virtually any name/content pairing? [20:16:59.0000] Ideally, should Twiter and Facebook be dropping the cards and the open graphs? Realistically, should the spec reflect the true state of ? [20:23:27.0000] I'm just reacting to validator errors versus warnings. [21:02:28.0000] JonathanNeal: about arbitrary names not being allowed for meta, it's because the spec says so [21:02:31.0000] currently [21:02:52.0000] I personally have come around to thinking it's a lost cause [21:03:12.0000] that is, it's a lost cause to disallow arbitrary name values [21:03:33.0000] but Hixie believes differently, I think [21:03:43.0000] I guess I should file a bug [21:04:16.0000] btw, you can put anything you want into the value of the content attribute, thoug [21:04:22.0000] there are no restrictions on that [21:21:54.0000] ok [21:21:55.0000] no, i agree that the current state is not viable [21:22:14.0000] the wiki experiment has been a great success [21:22:58.0000] it's shown that a lot of people really are wasting their time, and a lot of other people (far more than i expected) are doing things with meta and link rel and so on that are specific to their deployments [21:23:39.0000] there's a bunch of bugs about how to deal with this [21:26:21.0000] well, I just added one more [21:26:23.0000] https://www.w3.org/Bugs/Public/show_bug.cgi?id=22257 [21:27:03.0000] Hixie: you meant to say that the the wiki experiment has *not* been a great success? [21:27:14.0000] no, it's been a success [21:27:19.0000] oh [21:27:20.0000] it has taught us much [21:27:26.0000] at least, it's taught me much :-) [21:27:50.0000] yeah [21:27:59.0000] I agree about that too [21:28:20.0000] anyway I hope hsivonen can weigh in with his opinion on it [21:28:33.0000] I'll send a message to the list [21:28:46.0000] i'm not looking at this until sometime next year at the earliest, fwiw [21:28:58.0000] trying to get all the bugs that relate to interop dealt with first [21:29:03.0000] browser interop [21:29:22.0000] ok [22:34:31.0000] excellent, twitter cards will soon be valid, thanks to Hixie and MikeSmith and the rest of the gang. [22:40:15.0000] are they invalid now? [22:41:54.0000] Hixie: https://dev.twitter.com/docs/cards [22:42:03.0000] https://dev.twitter.com/docs/cards/markup-reference [22:42:42.0000] Bad value twitter:card for attribute name on element meta: Keyword twitter:card is not registered. [22:42:57.0000] why doesn't someone just register them? they, unlike almost every other meta tag, actually have a spec! [22:43:20.0000] hmm I see that some of them are now registered, actually [22:43:21.0000] http://wiki.whatwg.org/wiki/MetaExtensions [22:43:28.0000] They are http://wiki.whatwg.org/wiki/MetaExtensions [22:43:36.0000] ah. then they're valid already [22:43:44.0000] /me goes to add those to the validator source [22:43:50.0000] heh [22:44:01.0000] that's definitely one part of this that doesn't scale well [22:44:27.0000] Ooo, better sneak benschwarz' in there now. [22:45:18.0000] or not, but that's what made me start thinking about this https://github.com/benschwarz/metaquery [22:45:56.0000] You really ought to just make the element's name attribute open. [22:46:17.0000] the problem is that that would lead to a lot of authors wasting a lot of time [22:46:30.0000] because a lot of authors spend ages including things that are totally pointless [22:46:46.0000] to whom? [22:46:55.0000] pointless to anyone [22:47:05.0000] as in, tags that are never used by any software at all [22:47:38.0000] Hixie: they waste time now running stuff throught the validator now and getting errors they don't care about [22:47:58.0000] That's a relative statement. For instance, HTML5 Boilerplate rejected adding twitter cards because "most sites or apps will never use this and too many people will accidentally leave it in there." [22:48:00.0000] and then complaining through bug reports and mailing-list messages and e-mail to me [22:48:33.0000] MikeSmith: i disagree with the premise that it's a waste of their time to be told that something they're doing is pointless. [22:48:52.0000] but i agree that the current system isn't working [22:49:03.0000] "Keyword X is not registered." hardly communicates what you are saying. [22:49:30.0000] Hixie: It's not an absolute waste of time, it's a relative/net waste of time [22:49:30.0000] yeah, the current system isn't working, like i said [22:50:13.0000] it's like the validator complaining aobut table@border is a net waste of time, I think [22:50:18.0000] MikeSmith: if an author spends 10 minutes per project on a that gives nobody value, and one day they validate and the validator says "this is a waste of your time, stop doing it", then they can save 10 minutes per future project. [22:50:40.0000] Hixie: the thing is, they don't change their markup [22:50:46.0000] Then you pick which meta tags are a waste of time, or are they all? [22:50:49.0000] MikeSmith: then we're not communicating it well, like JonathanNeal said [22:50:52.0000] Also, is that really an error, or maybe just a warning? [22:51:14.0000] twitter cards might not be so arbitrary from the point of view of, say, twitter. [22:51:15.0000] Hixie: they keep the meta@name values, and they keep table@border. The main thing they learn is to ignore the validator for those cases [22:51:24.0000] JonathanNeal: well, right now we're in a kind of learning state where we are finding out what the proportion of useful values to not-useful values is [22:51:56.0000] MikeSmith: then we're not communicating to them why those are errors well enough. [22:52:16.0000] MikeSmith: especially for , that's unsurprising, given that it's not well defined in the first place :-) [22:52:26.0000] true [22:58:08.0000] IMHO, as long as HTML is the language of the browsers, the tag will continue down a liberal path, as developers continually extend it to communicate data about the document. [22:59:06.0000] sure, but in reality it's not that simple. [22:59:16.0000] metadata that nobody reads ends up bitrotting [22:59:30.0000] JonathanNeal: please try your twitter cards with http://validator.w3.org/nu/ or http://validator.w3.org/ now [22:59:47.0000] metadata that isn't in a specified format starts useless (with everyone using different syntaxes, etc) [22:59:55.0000] And the variations can be quite popular, as they have been for apple, facebook, and now twitter. [22:59:59.0000] I will, thanks. [23:00:33.0000] whoops, was using validator.nu, will try w3's [23:00:40.0000] VALID [23:00:59.0000] yeah I'll ping hsivonen about pulling the changes to validator.nu too [23:01:26.0000] JonathanNeal: only twitter ones I added are the ones that are actually registered [23:01:26.0000] The error is still "Bad value X for attribute name on element meta: Keyword X is not registered." I tried throwing something random in there. [23:01:44.0000] bbl, sleep [23:01:45.0000] not the other 50,000 other ones in their developer docs [23:01:49.0000] Hixie: nn [23:02:05.0000] Yea, cause now you are maintaining a spec. [23:06:14.0000] No sneak required [23:06:30.0000] :) [23:07:04.0000] I think this is a step forward, and hopefully opens folks up to the idea of letting meta name be free. [23:34:58.0000] http://lists.xml.org/archives/xml-dev/201306/msg00002.html trying to make sense of the black text on black background [23:39:30.0000] hsivonen, I read "I am as high as a kite" [23:40:01.0000] "You are using an unregistered meta keyword. Go read Cory Doctorow's metacrap essay." [23:41:42.0000] so much opportunity for 386 on xml-dev [02:59:53.0000] /me wonders who writes like TimBL speaks [03:03:17.0000] /me wonders the context for Ms2ger's wondering [03:05:52.0000] Kerouac? [03:06:14.0000] Not sure that's really a good fit… [03:09:56.0000] TimBL writes like TimBL speaks [03:10:30.0000] otherwise, I'm tempted to suggest Robert Shea + Robert Anton Wilson [03:17:14.0000] https://twitter.com/glazou/status/341819288182726657 [04:14:18.0000] what is the best way to make something like Facebook posts? [04:17:58.0000] content editable? [04:22:11.0000] I want to make sokmething like Facebook posts, how can I make this what is the best way? Using content editable or using textarea or? [04:23:56.0000] marcosc: No plan because testharness.js has been on github for a long time already [04:24:15.0000] jgraham, ah, should have known :) [04:34:53.0000] anolis doesn't support referencing a section, or does it? [04:35:59.0000] dekiss: try asking in #html5 or something instead [04:36:28.0000] ok [04:53:17.0000] tobie: So am I allowed to merge things in ttwf-docs, or is that only for special people? [04:53:43.0000] (also I wonder why the documentation is branded as "ttwf") [04:54:30.0000] (not only because the TTWF people settled on TestTWF which doesn't have such a bad reputation on urban dictionary) [04:55:19.0000] Darn. forgot about that. [04:55:27.0000] (not that I really mind, but at least at present TestTWF is a branch of the testing effort rather than the whole thing.) [04:56:44.0000] jgraham: plan is to use TestTWF for the overall web platform testing effort [04:57:02.0000] and have the event organization be a part of that. [04:57:11.0000] I see [04:57:53.0000] but yeah need to fix the urban dic ref. [04:57:56.0000] :( [04:58:31.0000] jgraham: I think everyone relates to the dinos really well [05:00:17.0000] The dinos are cool [05:00:22.0000] and you can go ahead and merge stuff there. [05:00:35.0000] I added you to the project with that in mind. :) [05:02:54.0000] dinos++ [06:17:59.0000] annevk: shouldn't stuff like "enum XMLHttpRequestResponseType" be marked with [NoInterfaceObject] in the IDL? [06:18:11.0000] no [06:18:20.0000] http://xhr.spec.whatwg.org/#interface-xmlhttprequest [06:19:06.0000] Why not? I see no point in having window.XMLHttpRequestOptions and window.XMLHttpRequestResponseType around [06:21:19.0000] hallvors: it appears you're either not reading WebIDL, or WebIDL says something silly [06:21:48.0000] Oh, maybe I'm making assumptions here.. [06:23:30.0000] ah, only relevant for "interface" [06:23:36.0000] ofc ;) [06:33:09.0000] hallvors: your pull requests are super unclear btw [06:33:16.0000] hallvors: you should really rebase or some such first [06:33:19.0000] sorry [06:33:33.0000] I know. I've just forgotten to do so [06:33:38.0000] :-/ [06:34:28.0000] but I've appointed jgraham as my Git guru, so the future is bright [06:40:09.0000] heycam|away: how should i spec style['font-size']? attribute CSSStyleDeclaration font-size; isn't valid webidl [06:42:25.0000] DOMString* [06:42:30.0000] I don't think that's possible [06:43:18.0000] not in the syntax, no [06:43:35.0000] but i can require a dancing unicorn in prose if i like [06:44:08.0000] amusing when people try to puff themselves up by saying "open letter" when they mean "blog post" [06:44:28.0000] "open blog post"? [06:44:32.0000] zcorpan: that just makes me sad that you haven't sent me a dancing unicorn [06:44:39.0000] i keep my blog posts to myself [06:44:51.0000] annevk - PR 9 should be a somewhat improved version of PR8 [06:45:00.0000] do you open each post with "dear diary" [06:46:05.0000] darobin: it hasn't arrived yet? maybe it died trying to swim across the sea :-( [06:46:31.0000] zcorpan: why is it not fontSize? [06:46:45.0000] zcorpan: sadness prevailed? [06:46:51.0000] annevk: because /topic [06:47:37.0000] I thought CSS had that bit right [06:47:46.0000] /me needs to go [06:48:05.0000] annevk: apparently trident/webkit/blink support both fontSize and font-size [06:48:41.0000] that does sound like sadness has prevailed [06:49:44.0000] hallvors: so don't you think we should do that in Fetch [06:49:54.0000] hallvors: e.g. should not do the fragment thing either [06:50:15.0000] hallvors: I think even old Fetch (which XHR references) requires this [06:53:17.0000] I certainly think we should do it in Fetch ;) [06:53:36.0000] so you can add it there if you want [06:54:08.0000] - just that my main concern still is shipping the XHR spec in a usable state [06:54:39.0000] right, you realize shipping XHR also references fetch? [06:55:47.0000] I'm not sure if we'll be "allowed" to - what are those pub rules again? Isn't Fetch too immature? [06:56:41.0000] anyway, it's up to you. By all means add it in Fetch if you would rather do that [06:56:49.0000] there's Fetch and HTML fetch [06:56:52.0000] the lowercase was intentional [06:56:59.0000] XHR has referenced the latter for an eternity [06:57:01.0000] you can't remove that [06:57:11.0000] you should really know this if you're editing XHR... [07:01:30.0000] Right - it's already covered by HTML5 fetch text. Good. [07:52:04.0000] hello [08:12:33.0000] tobie, for http://www.w3.org/wiki/Testing/Infra/Notification_Hell , can I propose critic as an alternative solution? [08:14:18.0000] Ms2ger: jgraham brought that up on #testing earlier. [08:17:17.0000] And if you guys like dinos, Mozilla has a nice one ;) [08:17:54.0000] Ah, zcorpan [08:18:16.0000] /me continues reading backlog [08:18:17.0000] Nvm :) [08:19:12.0000] "To JavaScript programmers, it's the only thing that makes sense." [08:19:15.0000] /me stops reading [08:36:37.0000] is the w3c html page down? [08:37:14.0000] which one, reyre? [08:37:33.0000] JonathanNeal: http://www.w3.org/html/wg/drafts/html/master/ [08:39:37.0000] whatwg.org/html is up [08:41:08.0000] Ms2ger: awesome, i'll use that :) [08:41:11.0000] thanks [08:41:18.0000] Np [08:41:52.0000] Ms2ger: is this less up to date then w3c site? [08:42:16.0000] More [08:44:57.0000] Ms2ger: agh :( that means the spec has changed again for WEBVTT ... [08:45:01.0000] sigh [08:58:58.0000] reyre: I think they're divergent forks, so different people will have different opinions in which is more up to date [09:01:02.0000] But he works on Mozilla, so for his purposes, the WHATWG version is most up to date :) [09:10:10.0000] (is that also true for WebVTT? I'm not sure what happened there) [09:12:52.0000] jgraham: iirc silvia is maintaining the spec at http://dev.w3.org/html5/webvtt/ [09:13:19.0000] I don't know about the whatwg side [09:23:18.0000] the format is W3C-only [09:23:22.0000] or CG-only [09:23:35.0000] and such are part of HTML still [10:27:47.0000] Ms2ger: depends on which folks at Mozilla for which spec to use [10:28:00.0000] Track [10:28:55.0000] for acc implementations the W3C html spec is generally more up to date and followed [10:29:42.0000] Ah, we're implementing longdesc? [10:30:18.0000] Ms2ger: nut talking about longdesc [10:30:34.0000] Ms2ger: thats not in the HTML spec [10:31:18.0000] Ms2ger: but alas you are [10:31:32.0000] I wonder how that was decided [10:32:30.0000] the acc support for longdesc has been implemented in Firefox for donkeys years, only browser that implemented AFAIK [10:32:57.0000] Here is the bug: https://bugzilla.mozilla.org/show_bug.cgi?id=854848 [11:10:45.0000] "While we've considered longdesc in the past, the landscape has changed and there are reasons to support it now. For one, the W3C validator now supports it, making it more likely to be supported across browsers." [11:10:50.0000] sigh [11:11:26.0000] I didn't add support for it to the validator as an endorsement. [11:11:48.0000] I added it mostly because of aggressive lobbying. [11:12:19.0000] Similar to the kind of aggressive lobbying you can see in the comments in that bug. [11:13:41.0000] Where the word "aggressive" is an understatement. [11:13:56.0000] There are a lot of other words that kind of thing could be called. [11:14:03.0000] Seems like an effective way to do lobbying - you increase the pressure until just one of your opponents cracks, and then you can use them as leverage against your remaining opponents, until you win [11:14:14.0000] MikeSmith, apparently, "successful"? [11:14:44.0000] tantek: not sure it's successful for users [11:15:12.0000] certainly yeah it's successful for the longdesc partisans [11:15:20.0000] MikeSmith: you could just add snippets of whatever bits you found most obnoxious to the validator output whenever it finds a longdesc [11:15:22.0000] MikeSmith - indeed, or for the web, or for accessibility overall. [11:15:42.0000] Philip`: sadly yeah I guess so [11:17:45.0000] jacobolus: Yeah I should have the validator emit a warning, "Despite some people wanting you to think otherwise, there is no consensus among accessibility professionals that the longdesc attribute is a win for accessibility." [11:20:35.0000] better to let it be, there are other things to waste ones time on [11:21:05.0000] yeah [11:21:28.0000] we don't always reach consensus, but when we do, it's by exhaustion [11:21:55.0000] heh [11:35:33.0000] way too mild. name names! step it up! [11:35:33.0000] "X, Y, and Z companies pressured the validator to add 'longdesc', but everyone else on the web thinks it's a dumb idea" [11:35:53.0000] :p [12:14:59.0000] So, ! [12:18:53.0000] jacobolus: companies or individuals? [12:19:21.0000] zcorpan: I dunno. I'm just joking :) [12:27:25.0000] jacobolus: You should be warned that one of the Commandments of The Lobby is "Thou Shalt Not Joke About Longdesc" [12:28:30.0000] jgraham_: maybe browsers can just put in a picture of longcat every time someone uses a longdesc? [12:28:46.0000] (completely serious recommendation) [12:28:57.0000] jacobolus, but how will the blind people see it? [12:29:11.0000] long meows? [13:11:40.0000] woah, crashed gecko [13:11:43.0000] it's been a while [13:11:49.0000] http://software.hixie.ch/utilities/js/live-dom-viewer/saved/2276 [13:12:04.0000] Ms2ger: ping https://www.w3.org/Bugs/Public/show_bug.cgi?id=21736 [13:12:55.0000] Hixie_, which Gecko? [13:13:47.0000] Mozilla/5.0 (Macintosh; Intel Mac OS X 10.8; rv:23.0) Gecko/20130508 Firefox/23.0 [13:15:24.0000] /me frowns [13:17:17.0000] That's pretty out of date? [13:17:45.0000] I can find no proof in the code that a named getter would ever have existed on select [13:18:56.0000] Hixie_, huh. [13:19:03.0000] "pretty out of date"! that's barely a month old! and i've been gone for most of that month! :-P [13:19:12.0000] gotta give the poor browser time to download its updates [13:19:39.0000] i'm actually surprised it's not older, since i was afk from 0504! :-) [13:19:55.0000] The crash with namedItem is probably https://bugzilla.mozilla.org/show_bug.cgi?id=877910 [13:20:28.0000] But I'm quite confused where I got that [13:20:29.0000] any idea what made you file the bug? :-) [13:21:48.0000] I presume something [13:23:11.0000] I thought that "more than 24 hours" was "pretty out of date" for a browser these days :p [13:23:28.0000] Although actually some of mine are rather more out of date than that [13:23:43.0000] Opera 12.16 for example... [13:24:14.0000] /me looks at the bugs he touched around that time [13:25:11.0000] https://bugzilla.mozilla.org/show_bug.cgi?id=862084 presumably... [13:26:14.0000] Hixie_, so I probably just got confused by the namedItem [13:26:21.0000] k [13:27:57.0000] Hixie_, while you're in the area, https://www.w3.org/Bugs/Public/show_bug.cgi?id=22225 [13:29:46.0000] i've long left the area, but i'll be back. :-) [13:30:03.0000] That fixes the crash you hit ;) [13:30:16.0000] ...how? o_O [13:30:30.0000] resource selection algorithm, 9, , 10: Failed: Queue a task, using the DOM manipulation task source, to fire a simple event named error at the candidate element, in the context of the fetching process that was used to try to obtain candidate's corresponding media resource in the resource fetch algorithm. [13:30:34.0000] Because JSAPI sucks :) [13:30:52.0000] anyone have any idea wtf i was talking about when i wrote "in the context of..." above? [13:31:33.0000] And in particular, we were creating the returned JS object manually, and needed the JS object for the options collection to exist for that [13:32:49.0000] Hixie_: Not a clue. [13:33:13.0000] ok i'm removing that text [13:33:16.0000] doesn't seem to mean anything [13:33:22.0000] and i've no idea what i was trying to say [13:33:42.0000] Ms2ger: that sounds... unrelated to the idl? :-) [13:34:00.0000] Hixie_, well, our IDL said 'object' [13:34:16.0000] And if it says that, you need to use JSObjects [13:34:20.0000] It sucks [13:34:24.0000] k [13:34:28.0000] i guess [13:34:29.0000] :-) [13:36:07.0000] Hixie_, fun stuff! :) [13:36:37.0000] Can your IDL not express A or B [13:36:39.0000] ? [13:36:42.0000] Yes [13:36:46.0000] But not for return values [13:36:55.0000] Oh [13:37:06.0000] Lifetime management there is somewhat annoying, aiui [13:38:01.0000] It'll come eventually, I hope [13:48:47.0000] hmm... the crash report comments for that crash are really angry [13:49:06.0000] and/or uninformed [13:51:39.0000] Better than uniformed, which is what i read first [13:54:18.0000] so... why do we allow flow content in
and [14:14:01.0000] interesting that view-source:http://svn.whatwg.org/webapps/source is horribly slow in browsers. except presto is fast [14:15:40.0000] does 
 make webidl-check skip validation?

[14:17:18.0000] 
Hixie_: for 
 i think the use case was something like Q&A. for  i guess things like forms?

[14:20:17.0000] 
"idl extract" is to make the idl checker skip the block, yeah

[14:20:31.0000] 
those use cases kind of make sense i guess

[14:24:41.0000] 
thanks

[14:27:33.0000] 
Hixie_: allowing flow content there... wasn't that a zcorpan thing?

[14:27:50.0000] 
Hixie_: convoluted examples of complex markup in header cells

[14:29:51.0000] 
annevk: don't think so :-)

[14:31:42.0000] 
i recall 2006-ish when we pondered about content models i argued for phrasing-only in dt/th. but recently someone came up with the above use cases and the spec was changed to allow flow content. (also in figcaption)

[14:33:41.0000] 
http://lists.whatwg.org/pipermail/commit-watchers-whatwg.org/2011/005776.html

[14:33:52.0000] 
http://lists.whatwg.org/pipermail/commit-watchers-whatwg.org/2010/004216.html

[14:34:05.0000] 
good morning, Whatwg!

[14:35:46.0000] 
https://www.w3.org/Bugs/Public/show_bug.cgi?id=13174#c6 already mentioned my plans for  in  :-)

[14:37:10.0000] 
those examples seem to have changed :-(

[14:39:34.0000] 
maybe we should start attaching snapshots of pages to bugs

[14:40:49.0000] 
well comment 2 has all you really need

[14:46:29.0000] 
/me notices dglazkov's mornings getting later and later

[14:58:22.0000] 
anyone got IE10? IE9 is acting odd on http://software.hixie.ch/utilities/js/live-dom-viewer/saved/2280

[15:00:09.0000] 
sicking: mounir pinged you on https://www.w3.org/Bugs/Public/show_bug.cgi?id=20580

[15:00:58.0000] 
Hixie_: I get "error: Invalid argument. on line 4" in IE10

[15:01:14.0000] 
well at least the line number makes more sense

[15:01:16.0000] 
following a single "log:"

[15:01:18.0000] 
IE9 was saying line 128

[15:01:29.0000] 
but that still makes no sense

[15:01:32.0000] 
thanks though!

[15:01:36.0000] 
anyone understand https://www.w3.org/Bugs/Public/show_bug.cgi?id=21800 ?

[15:05:23.0000] 
I don't even know what s/b is.

[15:07:06.0000] 
the google says maybe "should be"?

[15:08:57.0000] 
Are  tags liberal?

[15:09:32.0000] 
Howdy Hixie_, Did you and MikeSmith end up emailing folks about the meta name property?

[15:10:33.0000] 
JonathanNeal: it's on my list of things to deal with in 2014 Q1

[15:10:41.0000] 
then again, it was on my list for 2013 Q1

[15:11:06.0000] 
https://twitter.com/antimattur/status/342036107585806336 beautiful

[15:11:39.0000] 
oh god, that's an actual quote http://www.altova.com/list/xml-dev/201306/msg1000330505.html

[15:11:45.0000] 
lol interwebs

[15:15:53.0000] 
he does go on to say "it is virtually forgotten" so it's not clear what he meant...

[15:16:26.0000] 
xforms wasn't a thorn in the side, it was the catalyst :-)

[15:17:21.0000] 
bbiab.

[15:25:05.0000] 
omg, Boeing is using SVG

[15:27:32.0000] 
not the best choice for circuit diagraming

[16:28:12.0000] 
https://www.w3.org/Bugs/Public/show_bug.cgi?id=21791 stupid browsers :-|

[16:41:59.0000] 
Hixie_, here

[16:42:00.0000] 
?

[16:42:13.0000] 
if any html editor here please I have one question thanks

[16:47:54.0000] 
dekiss: don't ask to ask.

[16:48:15.0000] 
it's hard to tell if anyone knows the answer to your question if you don't state it.


2013-06-05
[18:07:20.0000] 
JonathanNeal: I e-mailed the mailing list to invite some feedback but I don't plan on mailing anybody else

[18:07:36.0000] 
who do you think we should e-mail>

[18:07:42.0000] 
?

[18:07:43.0000] 
?

[18:12:26.0000] 
I dunno. I tweeted it for some feedback.

[18:16:10.0000] 
rillian, sry now I saw

[18:16:21.0000] 
rillian, who edit html spec?

[18:16:28.0000] 
who is authorized?

[18:29:48.0000] 
https://twitter.com/TVRaman/status/342047516025552896

[18:30:00.0000] 
Xforms in the air today I guess

[18:30:08.0000] 
even got Raman tweeting

[18:30:23.0000] 
I didn't even know he was twitter

[18:34:49.0000] 
MikeSmith, you guys get feedback in emails and here and edit hmtl specification based on the feedback?

[18:38:58.0000] 
dekiss: Hixie is the sole editor of the whatwg HTML spec. The W3C publishes a version of the spec also, based mostly only the whatwg spec but with some additions and changes. There are several editors of the W3C HTML spec but probably the most active one is SteveF, who's on this channel often and who focuses on accessibility features.

[18:41:02.0000] 
nice

[18:41:36.0000] 
I will look for him to ask him if accessibility technology devices can interpret dom and javascript

[18:43:17.0000] 
dekiss: the answer is that the DOM and the results of JS operations on the DOM do get exposed to accessibility software

[18:43:37.0000] 
though in a somewhat different form

[18:44:42.0000] 
MikeSmith,  thanks I will download and/or buy some devices to check it myself

[18:58:08.0000] 
Did you guys change the validator to stop allowing  from wrapping block content?

[18:58:59.0000] 
JonathanNeal: it doesn't allow it now

[18:59:08.0000] 
there's been no change

[18:59:29.0000] 
that causes an error to be reported

[19:35:05.0000] 
JonathanNeal - see, I was pretty sure you couldn't 


[22:54:54.0000] 
Hixie: are you referencing level 4 of Selectors in HTML because something in the spec specifically requires something in level 4, or rather just because it's the latest?

[23:20:40.0000] 
MikeSmith: i believe i'm not referencing the editor's draft, actually

[23:20:44.0000] 
er

[23:20:44.0000] 
now

[23:20:45.0000] 
not not

[23:21:21.0000] 
yeah, i'm just referencing the editor's draft

[23:21:32.0000] 
(closest they have to a "living standard")

[23:32:30.0000] 
Hixie_: ok

[23:47:57.0000] 
OH: "Good ideas never die ... XForms..."

[00:08:12.0000] 
Ms2ger: "proves that bad ideas never die, either"?

[00:17:18.0000] 
 http://w3cmemes.tumblr.com/post/34634329920/csswg-resolves-to-use-less-magic

[00:17:19.0000] 
 dbaron: Could introduce some magic

[00:59:01.0000] 
it's really not *that* magical

[01:11:00.0000] 
tobie: Are you going to rename the ttwf-docs repo?

[01:11:17.0000] 
yeah, I absolutely should.

[01:12:07.0000] 
also the docs suffix is idiotic

[01:12:15.0000] 
as we'll have more in there than just docs

[01:12:20.0000] 
including the main page, etc.

[01:12:29.0000] 
… the events

[01:13:27.0000] 
testtwf-website?

[01:14:01.0000] 
WFM

[01:14:07.0000] 
deal

[02:11:24.0000] 
tobie: in general I take the "docs" abbreviation to just mean (Web) "documents" and not necessarily documentation

[02:11:33.0000] 
e.g., "/www/docs" and such

[02:11:46.0000] 
oh, you're old school, man!

[02:11:51.0000] 
heh

[02:11:57.0000] 
I am that, I guess

[02:12:13.0000] 
anyway, -website is definitely clearer

[02:16:41.0000] 
bah

[02:16:53.0000] 
it should be testtwf-var-www-sites

[02:17:35.0000] 
heh

[02:18:08.0000] 
Hmm, we would attract more users if we went all OSX. Library-WebServer-Documents-TestTWF ftw

[02:37:02.0000] 
good brainstorming

[03:04:13.0000] 
lol at "let's proceed only on es-discuss. This is […] not a browser issue"

[03:05:32.0000] 
In other news, skepticism remains about how much TC39 really want to work with the rest of the community.

[03:07:15.0000] 
ouch

[03:07:38.0000] 
:/

[03:35:32.0000] 
I don't know why he insists on trolling all the time

[03:35:41.0000] 
it's annoying

[03:43:37.0000] 
Habit?

[03:44:19.0000] 
You would know

[03:50:40.0000] 
Indeed so

[03:58:08.0000] 
annevk: this text misunderstands the meaning of CORS-same-origin, right: https://dvcs.w3.org/hg/html-media/rev/69dbfb8baca8#l1.18 ?

[03:59:12.0000] 
hsivonen: that reads very weird

[03:59:20.0000] 
annevk: that is, the CORS-same-origin concept covers the case where CORS has been used, so the "or use the crossorigin..." bit is confused

[04:00:00.0000] 
hsivonen: yes

[04:00:34.0000] 
also, xsl:template o_O

[04:10:21.0000] 
annevk: my bank exposed an expired certificate in production this morning and my reaction was "also, ISO-8859-15 o_O"

[04:11:48.0000] 
need a bigger O

[04:11:49.0000] 
I wonder if cobol outputs ISO-8859-15

[04:19:12.0000] 
Ms2ger: ISO-8859-1 is way too new for Cobol

[04:19:16.0000] 
argh

[04:19:21.0000] 
-15

[06:09:42.0000] 
hsivonen: http://i.imgur.com/4J7Il0m.jpg is  beautiful

[06:13:59.0000] 
annevk: btw philipj suggested maybe Fetch should cover the fetching story that apply to media elements

[06:14:27.0000] 
I know there's a sad story for  elements. There's one for media elements too?

[06:14:39.0000] 
what's about 
I should write these stories down somewhere so I don't forget about them.

[06:15:09.0000] 
well media elements support http range requests

[06:15:25.0000] 
zcorpan: per Document a URL is only fetched once and then simply retrieved from cache, irrespective of HTTP cache headers on the fetch.

[06:15:58.0000] 
I'll be back in 45min, need to ensure I have dry clothes before traveling Friday :)

[06:16:22.0000] 
Eh, dry clothes are overrated

[06:16:33.0000] 
loading an .ogg resource means doing a range request from 0 to infinity, then if the server supports range requests and the ogg resource doesn't have an index, you seek to the end (i.e. abort the first request and do a new request) of the file to find out the duration, then you do a third request from where you left of from the first request

[06:17:27.0000] 
and there's a different cache for media resources

[06:18:29.0000] 
we reverse engineered gecko in this area :-)

[06:19:16.0000] 
also, there are unsubmitted tests in this area :-/

[06:21:42.0000] 
zcorpan: Yeah, I noticed that the other day when Ms2ger was asking

[06:22:29.0000] 
Making Gecko run tests on their own reverse-engineered behaviour should be reason enough to release them ;)

[06:26:15.0000] 
the media cache tests are ugly php hacks

[06:26:15.0000] 
with state management being text files on the server, iirc

[06:26:50.0000] 
Nice

[06:27:11.0000] 
zcorpan: seems really wasteful to send a request expecting to abort it, since the server will probably push out a ton of data that would get discarded

[06:27:38.0000] 
surprised it doesn't just leave the first request running, and run a second request at the end in parallel

[06:27:42.0000] 
zewt: yeah well

[06:28:29.0000] 
the server probably can't tell the difference (it might not even be able to detect the abort until after the second request starts anyway), and it'd be a lot more efficient

[06:28:55.0000] 
yeah i've pondered about that as a way to do it as well

[06:28:56.0000] 
but it was hard to argue that we should spend time implementing such a tweak when there were lots of more important things to implement

[06:29:11.0000] 
I presume a certain amount of real world experience went into this design

[06:29:17.0000] 
Although I might be wrong ofc

[06:29:59.0000] 
i'm wary about making that assumption these days :)

[06:30:31.0000] 
i don't know if the parallel strategy was considered before things got implemented in gecko and presto

[06:31:02.0000] 
Well it seems kind of obvious

[06:31:45.0000] 
So I favour the explaination that it was considered and rejected over the explaination that no one even thought of it

[06:32:34.0000] 
i thought of it, but not until after we had implemented the aborting strategy

[06:33:17.0000] 
maybe the gecko people thought of it too and rejected it because it wouldn't work with some proxies or whatever

[06:33:21.0000] 
also means that if the server doesn't support range, you get fewer round-trips (since the main request doesn't have to be a Range and you don't have to redo it if it doesn't work), not that broken servers is a big optimization point

[06:33:43.0000] 
(also are)

[06:34:18.0000] 
though I guess you want to know that (in the "don't need to read from the end of the file" case) anyway, to know whether to enable seeking

[06:34:56.0000] 
zewt: if the server doesn't support range, the first request is carried on as if it wasn't a range request

[06:38:03.0000] 
though I guess you want to know that (in the "don't need to read from the end of the file" case) anyway, to know whether to enable seeking

[06:38:06.0000] 
zewt: if the server doesn't support range, the first request is carried on as if it wasn't a range request

[06:40:41.0000] 
now i don't follow :-)

[07:07:15.0000] 
ie10 still doesn't put spaces between console.log parameters? :|

[07:09:50.0000] 
jgraham: I'd favour asking

[07:11:05.0000] 
jgraham: I'd favour asking

[07:12:32.0000] 
ie10 still doesn't put spaces between console.log parameters? :|

[07:14:52.0000] 
annevk: Yes, of course, I wouldn't suggest we blindly assume everything must have been done for a good reason. But I also wouldn't go in with the attitude that things that look suboptimal don't have any reasoning behind them.

[07:19:59.0000] 
annevk: So... when do we start on that common definitions spec?

[07:21:00.0000] 
GPHemsley: do you have spare time?

[07:21:12.0000] 
ain't that a loaded question

[07:21:44.0000] 
/me really wishes he got paid for his spare time

[07:21:59.0000] 
heh

[07:22:12.0000] 
Too much, clearly ;)

[07:22:51.0000] 
You get paid for it in time

[07:23:06.0000] 
Which is, I have been led to believe, money

[07:23:40.0000] 
aha

[07:27:36.0000] 
heh

[07:31:25.0000] 
GPHemsley: I'm not sure we're quite there yet. Might need some more incremental evolution first.

[07:31:49.0000] 
incremental evolution where? because we may start to have divergent definitions

[07:33:15.0000] 
Well talking to each other helps with that :)

[07:33:52.0000] 
jgraham: I'm probably more worried about the cases where we don't even realize we're using/defining the same terms.

[07:33:56.0000] 
At some level you always have people not specs

[07:34:22.0000] 
Well if people aren't reading each other's specs, we have a bigger problem

[07:36:05.0000] 
Why is that? There are plenty of specs (that annevk writes) that have nothing to do with what I'm doing.

[07:37:12.0000] 
Ms2ger: Where in the code will I need to look to fix this? https://bitbucket.org/ms2ger/anolis/issue/8/should-override

[07:37:20.0000] 
GPHemsley: I think as a specification editor it's part of your job to be familiar with a large set of specifications. That's the only way you avoid needlessly introducing new concepts and terminology into the world.

[07:37:30.0000] 
GPHemsley: I didn't say that everyone had to read everyone else's specs. But if you are writing a spec and it isn't being read by anyone that is writing/reading any other specs, that's not good

[07:37:50.0000] 
Also, as annevk says, the reverse is true

[07:37:51.0000] 
GPHemsley: It'll also help understand where we might need to abstract or share concepts.

[07:38:46.0000] 
GPHemsley, either xref.py or xspecxref.py in anolislib/processes

[07:39:33.0000] 
tobie: Any chance of giving me whatever the permissions are on testtwf-website so that I can add hooks? (admin?)

[07:40:14.0000] 
annevk: By that logic, then, you should already be familiar with all the stuff I'm doing in mimesniff, right?

[07:40:57.0000] 
Ms2ger: Oh, and I heard there were tests I can run?

[07:41:28.0000] 
Yeah

[07:41:36.0000] 
GPHemsley: I don't think that follows directly from what I said and I haven't checked mimesniff lately

[07:42:24.0000] 
GPHemsley, python runtests.py in the top dir

[07:42:36.0000] 
Ms2ger: Ah, k, thanks

[07:42:55.0000] 
GPHemsley, adding a test for your case when you fix it is appreciated ;)

[07:43:04.0000] 
k

[07:50:36.0000] 
argh, I forget how hg works

[07:56:06.0000] 
GPHemsley - that happens to me all the time. I forget how (insert command line tool here) works.

[07:56:18.0000] 
heh

[07:56:28.0000] 
Well hg in particular is bad

[07:56:32.0000] 
well this is specifically because I've been spending so much time with git lately

[07:57:40.0000] 
So when learning any new one, I end up making "xyz for dummies" wiki pages for each such command line tool, because I can predict fairly accurately that my future self will be a dummy with regards to said xyz command line tool.

[07:58:45.0000] 
e.g. for hg, I didn't make this one, but I've contributed to it: http://wiki.csswg.org/tools/hg

[07:59:06.0000] 
you may find it of some use, and feel free to add any problems / questions you encounter

[08:02:25.0000] 
jgraham, it's much better than git, fortunately

[08:06:36.0000] 
Yes, because applications where the first step in using them is "fiddle about with a configureation file to enable essential features" are so easy to use

[08:08:57.0000] 
jgraham, "I forgot how to set it up" is a lot less of a problem than "I forgot how to use it"

[08:10:44.0000] 
I rarely know how to use things that I have to Google to even set up correctly.

[08:12:59.0000] 
jgraham - LOL: "fiddle about with a configureation file to enable essential features" are so easy to use

[08:13:07.0000] 
yeah, that.

[08:16:04.0000] 
Although it does look like if you have new enough mercurial, you get histedit with only a config option and not a config option + downloading a random script from the internet

[08:16:09.0000] 
So I guess that's progress

[08:17:02.0000] 
Ms2ger: Are the tests themselves documented? Because I've got one failing without telling me why or what it's for.

[08:17:13.0000] 
Still would be nice if someone would go back in time and convinve the hg people that cheap, temporary, local branches are an essential feature

[08:17:24.0000] 
Unfortunately, they don't believe in editing history\

[08:18:06.0000] 
GPHemsley, which one fails? They all pass for me

[08:18:17.0000] 
toc-basic.src.html

[08:18:35.0000] 
apparently x != y

[08:18:41.0000] 
but the difference is subtle and hard to see

[08:19:44.0000] 
id=baz? vs id=baz

[08:19:59.0000] 
Interesting

[08:20:53.0000] 
Ignore it for now and/or file a bug, I guess

[08:22:28.0000] 
Ms2ger: How is it that you're not seeing it?

[08:22:37.0000] 
(Maybe a different version of html5lib?)

[08:22:48.0000] 
Yeah, that's the most likely

[08:26:59.0000] 
0.95 vs 1.0b?

[08:27:19.0000] 
Would be interested in knowing if it's that

[08:27:37.0000] 
Doesn't seem to make a difference

[08:28:54.0000] 
Ms2ger: Is there a fake xref file I can use?

[08:29:41.0000] 
GPHemsley, I don't think so

[08:29:55.0000] 
You'll probably need an .options file too

[08:32:24.0000] 
Ms2ger: How do I add the xspecxref process?

[08:32:28.0000] 
(to the options)

[08:32:55.0000] 
/me reads some code

[08:32:59.0000] 
heh

[08:33:33.0000] 
{ 'processes': 'xspecxref' }

[08:33:51.0000] 
"TypeError: buildReferences() takes at least 3 arguments (2 given)"

[08:33:52.0000] 
Hrm

[08:34:04.0000] 
Or ['xspecxref']

[08:34:36.0000] 
Yeah, needs to be an array

[08:34:37.0000] 
yeah, right, already did that :)

[08:34:48.0000] 
the issue may be with not having an xref file

[08:35:17.0000] 
You may need to point at it...

[08:36:07.0000] 
'xref': 'data' or whatever you called the dir

[08:51:02.0000] 
Ms2ger: https://bitbucket.org/ms2ger/anolis/pull-request/8/fix-8-should-override/diff

[08:51:53.0000] 
Thanks, going to look in a bit

[09:38:17.0000] 
odinho: I think I suggested the shutdown. I'm only still there to turn out the lights :)

[09:42:44.0000] 
jgraham: ARE YOU IN OSLO

[09:43:45.0000] 
/me covers his ears

[09:50:24.0000] 
nimbu: NO. YOU KNOW THAT THE INTERNET LETS YOU TALK TO PEOPLE A LONG WAY AWAY WITHOUT SHOUTING?

[09:50:54.0000] 
THATS WHY IT IS GOOD

[09:51:04.0000] 
So we can probably talk at a normal volume

[10:07:42.0000] 
BUT THE TUBES

[10:08:50.0000] 
have to shout over the noise of splits

[10:21:54.0000] 
miketaylr: when you say that, I think you don't appreciate the elegance of minimal spanning trees

[10:22:46.0000] 
:)

[10:22:48.0000] 
/me throws a brick at rillian 

[10:23:09.0000] 
/me thinks he's done having classes about MSTs now

[10:44:11.0000] 
hello

[10:57:34.0000] 
Feedback welcome: http://wiki.whatwg.org/wiki/Contexts

[10:57:46.0000] 
especially from Hixie_ (HTML) and annevk (Fetch) ^^

[10:58:09.0000] 
and any CSS folks who care to weigh in (TabAtkins, SimonSapin, etc.)

[10:58:19.0000] 
GPHemsley: for bonus points, plot that against CSP fetch types

[10:58:54.0000] 
GPHemsley: CSS has these img-src, media-src, etc. thingies we need to integrate into the rest of the stack somehow

[10:59:00.0000] 
not CSS, CSP

[10:59:02.0000] 
grmbl

[10:59:08.0000] 
ah, k

[10:59:12.0000] 
do you have a link handy/

[10:59:13.0000] 
?

[10:59:22.0000] 
https://dvcs.w3.org/hg/content-security-policy/raw-file/tip/csp-1.0-specification.html

[10:59:34.0000] 
I think there might be a 1.1 too somewhere

[10:59:43.0000] 
it's not very well organized over on w3.org :/

[10:59:55.0000] 
MDN says https://dvcs.w3.org/hg/content-security-policy/raw-file/tip/csp-specification.dev.html

[11:00:03.0000] 
annevk, perhaps it could be better organized on w3.org/wiki ;)

[11:00:08.0000] 
GPHemsley: nice

[11:02:20.0000] 
incidentally, I think there may also need to be an 'embed' or 'plugin' context, but I haven't quite figured that out yet

[11:06:21.0000] 
yeah, HTML defines a bunch of that itself

[11:06:43.0000] 
getting these things extracted is good work

[11:08:01.0000] 
HTML currently only uses "browsing context" by name

[11:08:07.0000] 
the rest are currently implicit

[11:08:07.0000] 
the rest are currently implicit

[11:08:34.0000] 
but I needed to figure them out so that I could have separate sniffing algorithms for each

[11:18:12.0000] 
annevk: any idea why i can’t seem to ssh into html5.org anymore?

[11:18:19.0000] 
getting “Connection to html5.org closed by remote host. Connection to html5.org closed.”

[11:18:20.0000] 
annevk: OK, updated with CSP directives and 'plugin' context type

[11:19:44.0000] 
oh, hmm, maybe I should move that column

[11:23:50.0000] 
Ms2ger: I'm actually a little surprised no one's tried to fix the MSP thing. It's not like the network doesn't have to run the same daemon on all servers, so it should be easy to extend the protocol on that side to do multiple routing and de-dup

[11:24:37.0000] 
?

[11:26:32.0000] 
Updated: http://wiki.whatwg.org/wiki/Contexts

[11:26:46.0000] 
(Help removing questions marks appreciated.)

[11:29:19.0000] 
Ms2ger: MST, not MSP. It's going to be one of those typo days

[11:29:23.0000] 
/me considers tea

[11:30:09.0000] 
Michigan State Police?

[11:30:14.0000] 
Ministerio de Salud P�blica?

[11:30:29.0000] 
how are those different?

[11:30:30.0000] 
Measurements of Student Progress?

[11:30:31.0000] 
^_^

[11:31:41.0000] 
/me wonders if there's any website with reasonable UX in English that covers most European rail

[11:35:54.0000] 
I thought everyone used the german rail website for tat

[11:35:56.0000] 
*that

[11:36:05.0000] 
But I don't know about "reasonable UX"

[11:37:01.0000] 
Although if search is anything to go by, my information might be outdated

[11:38:28.0000] 
hmm this is getting linked to from a popular-this-week stackoverflow answer; someone (annevk?) might want to update it... https://developer.mozilla.org/en-US/docs/DOM_Levels

[11:39:18.0000] 
Ms2ger: Minimal Spanning Tree. I was referring to your comments of an hour past

[11:39:22.0000] 
Domenic_: Which SO? Or don't you know?

[11:41:03.0000] 
jgraham: I got it in my weekly Programmers newsletter. http://programmers.stackexchange.com/questions/200247/a-few-clarifications-about-the-dom?newsletter=1&nlcode=31316|4c54

[11:42:05.0000] 
Domenic_: heh, that seems quite out of date

[11:42:14.0000] 
jgraham, yeah, that does look a lot better than what I ended up with before, thanks

[11:42:18.0000] 
matjas: no, email me?

[11:42:28.0000] 
matjas: can sort out tomorrow hopefully

[11:44:56.0000] 
Hmm, someone with a SO account might like to provide a better answer

[11:45:10.0000] 
(It's Dutch translation doesn't appear to be entirely accurate, but hey, there is one)

[12:23:00.0000] 
Heh, CSS Eleven

[12:23:07.0000] 
wat

[12:25:16.0000] 
http://my.opera.com/dstorey/blog/css-eleven seems to be a contemporary blog post about them

[12:28:26.0000] 
is this like html5 superfriends?

[12:28:31.0000] 
Heh, CSS Eleven

[12:28:33.0000] 
wat

[12:28:55.0000] 
http://my.opera.com/dstorey/blog/css-eleven seems to be a contemporary blog post about them

[12:28:56.0000] 
is this like html5 superfriends?

[12:29:32.0000] 
Mm, forgot about those too

[12:29:51.0000] 
I guess those turned out just as useless

[12:35:24.0000] 
Ms2ger - nah - html5superfriends had an impact on the spec

[12:35:44.0000] 
e.g.