| 01:45 | <terinjokes> | is there a recommended way to check for Symbol support? |
| 01:50 | <terinjokes> | !!(global.Symbol && (typeof new Symbol() === 'symbol')) |
| 01:53 | <caitp> | works in v8 :D |
| 01:53 | <caitp> | did they ship that in jsc yet? |
| 02:25 | <erlehmann> | i made file2datauri better! http://news.dieweltistgarnichtso.net/bin/file2datauri.html |
| 02:25 | <erlehmann> | it now percent encodes text or us-ascii content if that makes it shorter |
| 02:34 | <caitp> | I stand correct, symbol is not a constructor :< |
| 02:34 | <caitp> | corrected* |
| 02:34 | <caitp> | that appears to be incorrect |
| 05:30 | <kazi> | Hi guys! I was trying to implement CSS 'background-size' property in my project. |
| 05:30 | <kazi> | But that produced "unknown vendor extension" in a validator |
| 05:32 | <kazi> | I used "-moz-" and "-webkit-". Is there any suggestion to pass the validation process however.? |
| 05:33 | <kazi> | I need to have "background-size: cover;" in my CSS |
| 05:43 | <zewt> | not worth worrying about |
| 05:46 | <hemanth__> | has chrome implemented the Stream API? |
| 05:51 | <hemanth__> | I find a window.Stream not sure which Stream is that. [Didn't find any API docs for the same] |
| 06:28 | <cabanier> | TabAtkins: how can I link to a constructor with bikeshed? It doesn't seem to be working for me |
| 06:37 | <TabAtkins> | cabanier: Should just be linkable like any other function (though it has type "constructor", not " method "), but sometimes it generates weird titles; I need to fix that. |
| 06:38 | <cabanier> | TabAtkins: OK. I will experiment a bit more. It doesn't seem to link with a constructor in the idl |
| 06:39 | <TabAtkins> | If you force generation, see what title attr it gets. |
| 07:46 | <MikeSmith> | extensible web summit 2.0 happening now |
| 07:46 | <MikeSmith> | JakeA talking |
| 08:00 | <JakeA> | (I stopped talking btw) |
| 08:03 | <darobin> | that never lasts long |
| 08:04 | <annevk> | MikeSmith: is this the backchannel or is there one? |
| 08:04 | <annevk> | MikeSmith: would love to see session notes |
| 08:04 | <annevk> | mounir: morning |
| 08:07 | <MikeSmith> | annevk: there's an etherpad .. I'm looking for the URL now |
| 08:08 | <Domenic> | http://oksoclap.com/p/ews-berlin |
| 08:08 | <MikeSmith> | thanks |
| 08:08 | <hemanth__> | Domenic: I noticed window.Stream in chrome, is that the Stream API? |
| 08:08 | <Domenic> | hemanth__: no |
| 08:09 | <hemanth__> | Domenic: what is that thinge? didn't find an API docs for it... |
| 08:09 | <Domenic> | hemanth__: no idea. do you have experimental web platform features on? |
| 08:09 | <hemanth__> | Domenic: yup |
| 08:09 | <Domenic> | hemanth__: probably an experiment then |
| 08:10 | <MikeSmith> | annevk: made #ews-berlin |
| 08:12 | <hemanth__> | Domenic: tweeted at you ;) |
| 08:17 | <annevk> | LimeChat is not great with invites |
| 08:18 | <annevk> | Oh my, those emails from Glenn |
| 08:18 | <annevk> | "My opinion is based on 20 years of experience with the W3C and 40 years of experience with standards bodies." |
| 08:18 | <annevk> | this needs the before your race was born meme |
| 08:18 | <annevk> | hober: ^^ |
| 08:18 | <annevk> | Since before your race was born, I have been involved with standards bodies |
| 08:24 | <hemanth__> | is anyone else noticing an error at http://oksoclap.com/p/ews-berlin ? |
| 08:25 | <MikeSmith> | hemanth__: nope |
| 08:26 | <hemanth__> | An error occured The error was reported with the following id: 'ZM5zq6TVFOL7ztvDQQSz' |
| 08:28 | <Ms2ger> | annevk, sounds more like "I really don't have anything to back this up" |
| 08:29 | <annevk> | Ms2ger: well yeah, typical argument from authority fallacy |
| 08:34 | <MikeSmith> | when the list of standards you were the main force for starts with TTML I guess people have a right to question how much authority that merits you |
| 08:41 | <Ms2ger> | The refusal to let you find the children of a commit must be one of the most infuriating misfeatures of git |
| 08:59 | <jgraham> | Ms2ger: Not that surprising given the data model (but also possible with a little effort) |
| 09:05 | <Ms2ger> | jgraham, critic handles it fine |
| 09:08 | <jgraham> | Ms2ger: Right but critic presumably caches it locally |
| 09:08 | <jgraham> | (e.g. in the database) |
| 09:17 | <Philip`> | Ms2ger: I think whenever I want to do that, I do something like "git branch -a --contains $REV" and choose which branch is the one I care about, then "git log $BRANCH" and search for $REV |
| 09:42 | <hsivonen> | what particularly evil SVG/MathML innerHTML cases should I test? |
| 09:45 | <Ms2ger> | !summon zcorpan |
| 09:47 | <ondras> | Domenic: ? |
| 09:47 | <ondras> | Domenic: I am interested in https://github.com/promises-aplus/promises-tests/blob/master/lib/tests/2.2.2.js#L35 |
| 09:48 | <ondras> | Domenic: the resolution (which implies calling the onFulfilled callback defined above) is supposed to be async? |
| 09:48 | <ondras> | Domenic: because you set the isFulfilled flag *after* the .resolve() ends.. |
| 09:48 | <mounir> | annevk: morning |
| 10:07 | <annevk> | mounir: hey |
| 10:09 | <annevk> | Hixie_: are you no longer using html5.org to split the specification? |
| 10:10 | <MikeSmith> | http://oksoclap.com/p/ews-berlin-spec-utopia なう |
| 10:10 | <annevk> | looks like it |
| 10:35 | <hsivonen> | "Bruce Lawson questions what the term Extensible Web really means, and whether it's just a vehicle for Google [and others] to promote the set of technologies they want to push" |
| 10:35 | <hsivonen> | http://oksoclap.com/p/ews-berlin |
| 10:37 | <jgraham> | hsivonen: Off the top of my head I can't think of anything more than the obvious cases to test |
| 10:40 | <hsivonen> | jgraham: ok |
| 10:41 | <hsivonen> | it looks like Chrome gets innerHTML on <annotation-xml> wrong |
| 10:41 | <hsivonen> | boo |
| 10:41 | <hsivonen> | unless someone wants to argue why it's right per spec |
| 10:44 | <Ms2ger> | jgraham, r? https://critic.hoppipolla.co.uk/r/2542 |
| 10:45 | <hsivonen> | hmm. the spec defines the annotation-xml stuff in terms of the start tag token |
| 10:45 | <hsivonen> | and a context node arguable doesn't have one |
| 10:47 | <annevk> | I wish DreamHost offered a way to disable passwords for SSH |
| 10:49 | <hsivonen> | is "HTML - <script>alert('LOL')</script>" as a bugzilla component under WHATWG in the W3C Bugzilla an unauthorized script injection or an admin's test? |
| 10:49 | <Ms2ger> | MikeSmith, ^ |
| 10:51 | <jgraham> | hsivonen: I think that zcorpan may have had something to do with that |
| 10:51 | <jgraham> | Or I saw him mention it |
| 10:52 | <jgraham> | Ms2ger: Done |
| 10:54 | <Ms2ger> | Takk |
| 10:58 | <hsivonen> | annevk: If you want to bikeshed annotation-xml: https://www.w3.org/Bugs/Public/show_bug.cgi?id=26783 |
| 11:27 | roc | gnashes teeth |
| 11:30 | <roc> | Before "extensible Web" was cool, I (and others) argued for Web Audio to be low-level primitives we could build on with JS. Google pushed to standardize and ship what they'd already implemented --- the built-in functionality they knew audio professionals wanted. |
| 11:30 | <roc> | Google people including ... Chris Wilson. |
| 11:39 | <hsivonen> | jgraham: https://github.com/html5lib/html5lib-tests/pull/50 I have very little clue if I'm using this github thing right. |
| 11:40 | <jgraham> | roc: People rewriting history to make their own position seem like it has been logically consistent all along is a type of cognitive bias. They don't even necessarily realise that they're doing it, or to what extent. |
| 11:41 | <jgraham> | hsivonen: It seems like you are |
| 11:43 | <roc> | jgraham: yeah. And, people are allowed to learn and change their minds. |
| 11:43 | <roc> | it's just annoying. |
| 11:44 | <jgraham> | Well I don't know about this "letting people learn" theory. It's not something we look for in politicians! |
| 11:45 | <roc> | hehe |
| 11:45 | roc | voted in New Zealand's general election today |
| 12:10 | <MikeSmith> | hsivonen: admin test I forgot to delete |
| 12:10 | <MikeSmith> | hsivonen: will delete it now |
| 12:11 | <MikeSmith> | ok gone now |
| 12:14 | <Domenic> | ondras: yes. See http://promisesaplus.com/#point-34 + footnote |
| 12:15 | <JonathanNeal> | Good morning! |
| 12:16 | <ondras> | Domenic: I see. Do I assume correctly that this behavior changed/was-not-specified-before when comparing to 1.0 ? |
| 12:16 | <Domenic> | ondras: nope, it was in 1.0. |
| 12:16 | <ondras> | interesting. |
| 12:17 | <ondras> | my 1.0-compliant impl now fails due to sync fulfillment |
| 12:17 | <ondras> | well, time to fix the code. |
| 12:17 | <Domenic> | probably just better test coverage, sorry :P |
| 12:17 | <ondras> | yeah, np |
| 12:18 | ondras | is actually quite happy that there is about a bazillion tests now |
| 12:34 | <cwilso___> | roc: I know you did. And in some ways, the media streams concept was quite right. In others, it was not. I still believe that offering ONLY those low-level primitives - i.e. expecting every developer to come up with their own convolution.js, dynamics.js, biquad.js - would have been a bad idea for the utility of audio on the web. However, I will point |
| 12:34 | <cwilso___> | out... |
| 12:34 | <ondras> | Domenic: testing the number of getter calls for returnedValue.then seems a bit strict to me, though |
| 12:34 | <ondras> | Domenic: imagine a code "if (value.then) value.then(...)" which fails. is that really a mistake? |
| 12:35 | <Domenic> | ondras: yes |
| 12:35 | <cwilso___> | 1) my first face-to-face contact with Chris Rogers was arguing with him to make worker-based script processors. It took me another year or so to fully understand how broken the current script processors are. |
| 12:35 | <ondras> | Domenic: any usecase that will explain this particular scenario to me? |
| 12:36 | <Domenic> | ondras: interop in the face of poorly-behaved then methods |
| 12:36 | <Domenic> | we could either spec "access .then twice" or "access it once"; we chose the latter. |
| 12:36 | <ondras> | hmm. |
| 12:36 | <ondras> | what about "accessing .then is idempotent"... :-) |
| 12:37 | <cwilso___> | 2) I was not rewriting history at all (presuming this is in ref to EWS) - I made it clear this journey of layering and attempt to architecturally layer Web Audio was in response to the TAG's review of about a year ago, not that it was that way all along. If that did not come out in the notes, or if I was not clear enough about it, I apologize. |
| 12:40 | <darobin> | jgraham: how far do you reckon you are from landing your TestTWF docs improvements? |
| 12:42 | <cwilso___> | I hope that I've been clear, in my occasional nose-poking in the media list, that I do think we need to build that lower-level primitive set - for example, exposing multiple interfaces/devices. Yes, in some ways that is backing into what you attempted to push several years ago. I think there is more to it than that; but at any rate: |
| 12:43 | <darobin> | (people here at the EWS want to join but we don't want them to read the wrong crap) |
| 12:44 | <jgraham> | darobin: Umm, I think the update is basically done, but to actually land it I might need to either have access to the main repo or get someone who has access to do the landing (because some travis setup is needed) |
| 12:44 | <jgraham> | (also we need a new repo with the w3c account) |
| 12:44 | <MikeSmith> | jgraham: new repo? |
| 12:45 | <jgraham> | MikeSmith: The travis-based solution I have uses two repos, one for source and one for generated content rather than two unrelated branches in the same repo |
| 12:45 | <jgraham> | Seemed to work better with Travis and kind of seems cleaner |
| 12:46 | <MikeSmith> | ah yeah I remember now |
| 12:46 | <darobin> | wait, which repo do you need access to? |
| 12:46 | <cwilso___> | yes you deserve credit for presciently pushing for low-level primitives. I do still think providing the higher-level functionality audio professionals (and non-professionals) want is (and was) important; but that should not be the only level. No, I do not claim to have always known all of this, nor am I attempting to say I've always felt precisely as I do |
| 12:46 | <cwilso___> | now. |
| 12:48 | <jgraham> | darobin: Let me think… |
| 12:56 | <jgraham> | darobin: So probably what I need is 1) to push the updates to the contributing file 2) For someone to create a testtwf-website-src repo (or to use the existing repo as the src repo and create a testtwf-website-generated repo and reconfigure the DNS stuff to use that) 3) Someone to generate an encrypted token to use for deployment on travis with the GIT_NAME, GIT_EMAIL and GH_TOKEN variables. This involves first deciding on a user to use (or creating |
| 12:56 | <jgraham> | ... OAuth token for travis to use |
| 12:57 | <jgraham> | travis encrypt 'GIT_NAME="Your Committer Name" GIT_EMAIL=committer⊙ec GH_TOKEN=oAuthToken |
| 12:58 | <cwilso___> | roc/jgraham: in short, I'll stand by the "we needed to enable audio magic (cf https://twitter.com/brucel/status/510007267522256896), but yes, we should have looked more closely at the primitives." |
| 13:08 | <annevk> | I doubt roc was opposed to doing higher-level at some point, just opposed until some library work was done, which would be roughly in line with the extweb party line |
| 13:09 | jgraham | wonders if there is a scedule for a higher level API on top of IndexedDB… |
| 13:18 | <roc> | cwilso___: I wasn't accusing you of rewriting history , that was jgraham :-). |
| 13:21 | <roc> | Your position has obviously evolved, and that's cool. |
| 13:25 | <jgraham> | I'm not sure that I accused anyone of anything more than being human |
| 13:31 | <cwilso___> | My panties are not in a bunch here. Just explaining the context of what I said, since I wasn't speaking at great length there, and I didn't want to be the source of your teeth-gnashing. At least, not for that reason. |
| 13:32 | <roc> | ta |
| 13:33 | <cwilso___> | annevk: yes, if extweb had been a thing then, it might have swayed me to push Chris Rogers harder. If I'd deeply understood the web audio architecture, at least. |
| 14:23 | <Ms2ger> | "We need to stop being assholes" @chaals |
| 14:24 | <Ms2ger> | May I suggest you start, chaals? |
| 14:32 | <Ms2ger> | @robinberjon:"If anyone acts like a jerk we kick them out" |
| 14:32 | <Ms2ger> | Bwahahahahahahahaha |
| 14:33 | astearns | /kick Ms2ger |
| 14:34 | <annevk> | hsivonen: http://annevankesteren.nl/2014/09/tls-first-steps |
| 14:36 | <darobin> | jgraham: so, I'm being too distracted by the summit at the same time in order to properly process your needs here; but it sounds like something I could do. Do you mind sending me an email or some such? |
| 14:38 | <Ms2ger> | darobin, do you have any examples of kicking out jerks? |
| 14:38 | <darobin> | Ms2ger: from where? |
| 14:38 | <darobin> | Ms2ger: I'm not exactly sure what your question is |
| 14:39 | <darobin> | Ms2ger: there's ample precedent for conferences for instance |
| 14:39 | <jgraham> | darobin: Sure |
| 14:39 | <Ms2ger> | I saw the quote "If anyone acts like a jerk we kick them out" on twitter |
| 14:39 | <darobin> | jgraham: ta |
| 14:39 | <darobin> | Ms2ger: right, yes, the idea is that people who act like jerks, after fair warning and no matter how good they are at the tech, can get blocked |
| 14:40 | <darobin> | also, the idea is that you can tell someone "please stop being a jerk" and give them a chance to retract whatever |
| 14:40 | <Ms2ger> | In, say, HTMLWG or WebApps? |
| 14:40 | <darobin> | Ms2ger: oh, I wasn't talking about W3C (though I'd hope it'd move there too) |
| 14:40 | <darobin> | I meant in places like Discourse and the stuff we're looking at building around it |
| 14:41 | <darobin> | e.g. in GH comments on specs |
| 14:41 | <Ms2ger> | I see |
| 14:41 | <darobin> | Ms2ger: do you have a specific concern? |
| 14:41 | <Ms2ger> | I was thinking of one particular jerk |
| 14:41 | <darobin> | I won't venture a guess :) |
| 14:41 | <Ms2ger> | I'll just narrow it to someone with 40 years standards experience |
| 14:42 | SteveF | everybody has their own favourite jerk to kick |
| 14:42 | <darobin> | Ms2ger: somehow I was thinking you had that in mind :) |
| 14:42 | <darobin> | Ms2ger: he's certainly one of the use cases I have in mind |
| 14:43 | <Ms2ger> | Would be nice to have such a policy at W3C, of course |
| 14:43 | <darobin> | Ms2ger: https://github.com/webspecs/specs.webplatform.org/blob/master/src/render/policy/anti-jerk.html |
| 14:44 | <darobin> | Ms2ger: yes. My hope is that if we build something successful then the policy can spread |
| 14:44 | <darobin> | that's how it happened with conferences |
| 14:44 | <darobin> | a few cool conferences started having anti-jerk policies, and now most do (in our world) |
| 14:45 | <jgraham> | That policy is probably too "fun" to be useful. |
| 14:45 | <jgraham> | Most of the instances of disruptive behaviour I have witnessed haven't been harassment, which is the only thing really called out |
| 14:48 | <darobin> | jgraham: I am happy to refine it over time, this was just my first shot |
| 14:48 | <darobin> | that said, I don't want to make it a processy thing; it's a spirit thing |
| 14:49 | <darobin> | you can call out things the way you think they are already though, e.g. http://lists.w3.org/Archives/Public/public-webapps/2014JulSep/0516.html |
| 14:50 | <Ms2ger> | darobin, ++, but I don't believe that'll change his behaviour |
| 14:51 | <darobin> | Ms2ger: yeah, I know — but if people start ignoring that sort of statement instead of starting useless threads from it (which I think can be a change of behaviour) then it works too |
| 14:51 | <darobin> | we can implement some form of social hellbanning |
| 14:51 | <darobin> | which is, after all, what "don't feed the troll" used to be for |
| 14:52 | darobin | wonders if Discourse supports hellbanning, needs checking |
| 14:53 | <jgraham> | Yeah, so the problem with a "spirit" thing rather than a "clear guidelines" thing is that people who are determined will ignore the spirit and then when they are called out on their behaviour will complain that they are being silenced and argue that they didn't break any actual rules. Or there will implicitly be one set of rules for people that the list moderators know and like and a different set for everyone else. |
| 14:54 | <darobin> | jgraham: I know, but the alternative is bureaucracy |
| 14:54 | <darobin> | that said, I will absolutely take improvements |
| 14:55 | <SteveF> | > Or there will implicitly be one set of rules for people that the list moderators know and like and a different set for everyone else. - happens all the time |
| 14:56 | <darobin> | SteveF: that only happens if you formally have moderators |
| 14:56 | <darobin> | that said, I'd get more pissed off at someone I know who's supposed to know better than to troll, than at a newcomer who might not be familiar with the stuff |
| 14:57 | <jgraham> | I don't think it only happens if you formally have moderators at all |
| 14:57 | <jgraham> | It's just the halo effect |
| 15:56 | <zcorpan> | Ms2ger: pong |
| 15:56 | <Ms2ger> | zcorpan, I don't remember what I needed you for |
| 15:57 | <zcorpan> | Ms2ger: a sandwich? |
| 15:58 | <Ms2ger> | No thanks, my spaghetti is nearly ready :) |
| 16:00 | <zcorpan> | hsivonen: the script component MikeSmith added on my request. although it failed to run the script because innerHTML doesn't run scripts, i assume something like <img onerror> works |
| 16:04 | <zcorpan> | hsivonen: i noticed because the "HTML - <img>" component read "HTML -" |
| 16:04 | <zcorpan> | on the search page or so |
| 16:21 | <annevk> | zcorpan: perhaps file a bug on Bugzilla? |
| 16:22 | <zcorpan> | annevk: need to check if it happens in a newer bugzilla |
| 16:57 | <jgraham> | Anyone know if "bundle" is a thing that gets installed with ruby gems, or something you have to install seperately? |
| 16:57 | <jgraham> | Looks like it's seperate |
| 17:00 | <MikeSmith> | zcorpan: I think the only newer bugzilla you're likely to find is https://bugzilla.mozilla.org |
| 17:01 | <MikeSmith> | hmm or I guess 4.4.5 is the more recent than the w3c instance but I don't know who might be running that or anything in between 4.4.2 and that |
| 17:09 | <annevk> | It's kind of annoying that with e10s all TLS dialogs are broken |
| 17:13 | <jgraham> | annevk: Is there a bug? |
| 17:14 | <annevk> | jgraham: good question. I suspect there is, but have not looked |
| 17:16 | <annevk> | I cannot find anything, will file a bug |
| 17:21 | <annevk> | jgraham: checked with cpeterson, he's filing a bug |
| 17:26 | <jgraham> | ap: Great |
| 17:26 | <jgraham> | uh |
| 17:26 | <jgraham> | annevk: Great |
| 17:27 | <annevk> | TabAtkins: could you please reply to the matches() / closest() thread? |
| 17:28 | <annevk> | TabAtkins: apparently we want to land our implementation but it's blocking on a decision on :scope in closest() |
| 17:29 | <annevk> | "As a W3T member, I would think it your duty to stay out of the fray. Best listen to your duty." |
| 17:29 | <annevk> | Glenn's arguments are getting better by the minute |
| 17:30 | <annevk> | jgraham: https://bugzilla.mozilla.org/show_bug.cgi?id=1066181 |
| 17:57 | <annevk> | Philip`: I have a DNS entry "*.philip A 92.243.11.39" |
| 17:59 | <annevk> | Philip`: that might cause problems with HSTS |
| 17:59 | <annevk> | odinho: same goes for you |
| 17:59 | <annevk> | odinho: "odinho CNAME html5.s0.no." |
| 18:08 | <mathiasbynens> | annevk: how so? |
| 18:08 | <mathiasbynens> | http://odinho.html5.org/ would get SSL + HSTS; HSTS would not affect http://html5.s0.no/ |
| 18:10 | <mathiasbynens> | annevk: btw, *.html5.org is on a shared host, so we have to use SNI… which means no support for IE on WinXP |
| 18:11 | <mathiasbynens> | (i don’t really care, but it may affect people wishing to run *.html5.org-hosted tests on oldIE — they’d have to click through a few warnings) |
| 18:14 | <annevk> | mathiasbynens: I know, see my blog |
| 18:14 | mathiasbynens | had missed it! |
| 18:15 | mathiasbynens | still wishes for an @annevkbot twitter account |
| 18:15 | <annevk> | mathiasbynens: hmm, I guess I can offer those people something else |
| 18:15 | <annevk> | heh |
| 18:15 | <annevk> | mathiasbynens: it does seem bad to CNAME another server that does not use TLS and then claim stuff is secure |
| 18:18 | <mathiasbynens> | hmm |
| 18:19 | <Philip`> | annevk: I have no idea what HSTS is, so I don't know if that matters |
| 18:19 | <mathiasbynens> | Philip`: https://www.owasp.org/index.php/HTTP_Strict_Transport_Security |
| 18:19 | <annevk> | Philip`: perfect configuration requires all subdomains to use TLS |
| 18:19 | <Philip`> | annevk: It looks like the only subdomain that's still occasionally used is fonts.philip.html5.org |
| 18:19 | <Philip`> | annevk: (and that's an obsolete service anyway) |
| 19:51 | <odinho> | annevk: You can nuke mine, a few links will break and all, -- and it's a rather nice hostname, but I don't use it enough to hold off on your perfect rating :] |
| 20:31 | <kamome_> | Hi, on schema.org, for openinghours, there is an example withe the time element: itemprop="openingHours" datetime="Tu,Th 16:00-20:00" |
| 20:32 | <kamome_> | But that doesn't validate (that datetime is not valid) - what would be the right way to mark this up? |
| 20:32 | <kamome_> | If there is such a way ... |
| 20:33 | <kamome_> | In fact, I'm looking for a solutionfor datetime="Mo-Su 07:00-22:00", if that makes a difference. |
| 20:36 | <Hixie_> | kamome_: use a <data itemprop="openingHourse" value="Tu,..."> instead of <time> |
| 21:30 | <kamome_> | <Hixie_>: Thanks, will try that. |
| 21:42 | <annevk> | odinho: Philip`: ok, I guess we'll see how it goes |
| 21:43 | <annevk> | odinho: Philip`: if those services break, you have been warned, if we can keep them working, so much the better |
| 21:44 | <annevk> | http://lists.w3.org/Archives/Public/www-tag/2014Sep/0034.html wut |
| 21:48 | <thin> | Hello. |
| 21:48 | <thin> | Does anyone remember a proposal by a member of the Chrome team about |
| 21:48 | <thin> | using most semantic class names in your markup |
| 21:48 | <thin> | and then having a new kind of script halfway between js and css |
| 21:49 | <thin> | that you would use to select elements based on their semantic class name, and then on the fly add the truck load of OOCSS class names that are so popular lately |
| 21:49 | <thin> | and then after that the real css and js files are applied. |
| 21:49 | <thin> | Anyone remember? Have a link? |
| 21:51 | <jgraham> | annevk: It seems to me that sort of message is only important insofar as it stands in the way of whatever your actual goals are. If it doesn't it seems like the best thing is to ignore it |
| 21:52 | <Hixie_> | thin: i don't recall such a proposal, but fwiw the HTML spec a few years ago had something based on "semantic class names" and it didn't go down well |
| 21:54 | <thin> | I don’t think it was the kind of proposal to show up on the radar of the offical whatwg or w3c stuff but it was a seriously written blog post by someone who I think worked at google on chrome. |
| 21:56 | <thin> | I basically said. Avoid using classnames like .blue in your actual markup. Use .branding. Then in this new scripting language sandwiched after dom ready but before css or js is applied, you’d select dom nodes based on their sensible class names and add to them all the mess OOCSS class names you want. And then the css and js happen. |
| 21:57 | <thin> | Still doesn’t ring a bell? |
| 21:58 | <Hixie_> | thin: maybe try #chromium or #blink? |
| 21:58 | <jgraham> | thin: I don't know what OOCSS is, or what behaviour the "semantic" classnames would have beyond whatever the scripts/stylesheets applied |
| 22:01 | <thin> | jgraham: it’s an approach to css styling that promises flexibility through inheritance but to me it just looks like class=“blue-gradient-background white-text” all over again. |
| 22:03 | <jgraham> | thin: Well it sounds like what you're talking about is something a library would implement, not a browser |
| 23:04 | <Hixie_> | anyone got IE who can test http://software.hixie.ch/utilities/js/live-dom-viewer/?saved=3099 for me? |
| 23:05 | <Hixie_> | i just need the output in the "Rendered view:" bit. This is for https://www.w3.org/Bugs/Public/show_bug.cgi?id=25929 |
| 23:09 | <kamome_> | Hixie: Thanks again, data for openingHours works for validator.w3.org as well as for www.google.com/webmasters/tools/richsnippets |
| 23:09 | <Hixie_> | coolio |