| 00:09 | <MikeSmith> | bravo sleevi http://lists.w3.org/Archives/Public/public-webrtc/2015Jan/0013.html |
| 01:32 | <MikeSmith> | https://medium.com/@dmitriid/w3c-and-whatwg-should-die-a-quick-and-horrible-death-a7cff5e1da8f is subtle and understated |
| 01:33 | <MikeSmith> | I think that guy needs to try to state things more emphatically |
| 01:38 | <MikeSmith> | btw apparently WHATWG has been abbreviated to just "WG" |
| 01:39 | <MikeSmith> | take a break for the holidays and come back to find out all kinds of things have happened |
| 01:39 | <MikeSmith> | anyway, I support this rebranding |
| 01:39 | <MikeSmith> | WG++ |
| 01:39 | <terinjokes> | WG-ng++ |
| 01:40 | <terinjokes> | the next generation WG, for C++ |
| 01:42 | <caitp> | what does the "WHAT" part stand for anyway |
| 01:43 | <caitp> | and can it be condensed into "WUTWG" |
| 01:43 | <terinjokes> | slightly serious question: how does one pronounce WHATWG? |
| 01:43 | <MikeSmith> | what wee gee |
| 01:43 | <terinjokes> | i say "WHAT-wig", but everyone looks at me funny |
| 01:43 | <Hixie> | i pronounce it "whatwuhjee" |
| 01:44 | <MikeSmith> | Hixie pronounces it funny |
| 01:44 | <MikeSmith> | everybody else says wee |
| 01:44 | <MikeSmith> | because weeee |
| 01:45 | <MikeSmith> | hey this prodigy has an actual concrete proposal |
| 01:45 | <MikeSmith> | "Get people from Apple to implement all font-related specs." |
| 01:45 | <MikeSmith> | +1 |
| 01:46 | <MikeSmith> | but I think he means something different than what that sentence actually say |
| 01:46 | <MikeSmith> | though "Dig W3C and WHATWG out of their graves, kill and bury them again." is somewhat concrete too |
| 01:46 | <jamesr__> | what double uwe gee |
| 01:50 | <MikeSmith> | the WHAT WAI gee |
| 01:50 | <MikeSmith> | good brainstorming |
| 01:51 | <MikeSmith> | "Both W3C and WHATWG should die a quick and horrible death, and the development of web standards should be given to a lean and mean group of people" |
| 01:51 | <MikeSmith> | we're plenty mean enough at least |
| 01:51 | <Hixie> | yeah i wonder where he's gonna find these people |
| 01:51 | <MikeSmith> | yeah |
| 01:52 | <MikeSmith> | finding the mean people is easy |
| 01:52 | <gsnedders> | MikeSmith: the not having reference implementations bit is almost concrete |
| 01:52 | <gsnedders> | MikeSmith: fuck you, that's totally the hardest part! |
| 01:52 | <MikeSmith> | hey gsnedders! |
| 01:52 | <MikeSmith> | perfect timing! |
| 01:52 | <MikeSmith> | you're lean |
| 01:52 | <MikeSmith> | but you're not mean |
| 01:52 | <MikeSmith> | damn |
| 01:52 | <gsnedders> | I can be mean! |
| 01:52 | <MikeSmith> | well let's join together, combine efforts |
| 01:53 | <MikeSmith> | no gsnedders you do the lean part, I'll do the mean part |
| 01:53 | <MikeSmith> | problem solved |
| 01:54 | <caitp> | when you're having trouble reaching consensus, take them bowling |
| 01:54 | <gsnedders> | I'm randomly either really good at bowling or really shit. |
| 01:55 | <gsnedders> | I'm never in between. |
| 01:55 | MikeSmith | adds this stuff to his notes |
| 01:55 | <gsnedders> | I either go out and get strikes every second go, or come last. |
| 01:56 | <MikeSmith> | forget bowling I think the only way we're going to resolve this is with a three-way dance-off among this guy, Robin, and Lachlan |
| 01:56 | <MikeSmith> | he Lachlan is both lean and somewhat mean |
| 01:56 | <MikeSmith> | Robin in unfortunately neither mean nor super lean |
| 01:56 | <gsnedders> | I'm choosing Girls Just Want To Have Fun for the dance-off, yo. ;P |
| 01:56 | <MikeSmith> | heh |
| 01:58 | <MikeSmith> | either that or Super Freak |
| 01:58 | <Hixie> | man medium is annoying |
| 01:59 | <Hixie> | also how the hell do i have 1214 medium followers |
| 01:59 | <Hixie> | since i only created my account about 15ms ago |
| 02:00 | <Hixie> | wait this is dumb |
| 02:00 | <Hixie> | it's not showing my response on the page |
| 02:02 | <MikeSmith> | sgalineau_: please make some magic with https://medium.com/@dmitriid/w3c-and-whatwg-should-die-a-quick-and-horrible-death-a7cff5e1da8f |
| 02:03 | <Hixie> | my reply was https://medium.com/@Hixie/i-wish-things-were-that-simple-df0d2a786ecc fwiw |
| 02:03 | <caitp> | that guy seems a bit upset |
| 02:10 | <gsnedders> | wait you can post replies on medium? |
| 02:10 | <gsnedders> | weird |
| 02:39 | <MikeSmith> | sgalineau_: btw I don't know if Reginald was being facetious in https://twitter.com/raganwald/status/551808605763014656 but if not I hope he realizes how wrong he is |
| 02:42 | <caitp> | he's actually spot on there |
| 02:43 | <caitp> | except it doesn't work well on the web |
| 02:43 | <MikeSmith> | I guess I'll have to disagree with you about that today |
| 02:43 | <MikeSmith> | though I'd disagree pretty much any other day as well |
| 02:44 | <caitp> | think about it mike, you're a speaker, it's your job to analyze your audience |
| 02:44 | <MikeSmith> | heh |
| 02:44 | <caitp> | and speak within the constraints that they impose |
| 02:44 | <MikeSmith> | I hate public speaking |
| 02:44 | <MikeSmith> | public speaking is a con job |
| 02:44 | <MikeSmith> | it's an acit |
| 02:44 | <MikeSmith> | *act |
| 02:45 | <MikeSmith> | it's entertainment |
| 02:45 | <caitp> | you'd probably tell a different joke to your mom than you would to your best friend steve from middleschool |
| 02:45 | <tantek> | meh, it's your responsibility, but "being offended" is in the mind of the offendee. |
| 02:45 | <caitp> | you tailor things you say for different people, it's just natural |
| 02:45 | <MikeSmith> | and anyway when I do go out a speak at events I say all kinds of offensive stuff. I make a point of it |
| 02:45 | <caitp> | people do it |
| 02:45 | <caitp> | sometimes we just do a bad job of it |
| 02:46 | <caitp> | people that run around saying "well if you get upset, its YOUR fault for being STUPID" are generally not very pleasant to be around :p |
| 02:47 | <MikeSmith> | yeah I try to not say Christ or fucking Christ or holy fucking Christ so much around my mom, given that she's pretty deeply religious |
| 02:47 | <MikeSmith> | right now I kind of feel like not going out of my way to do that so much |
| 02:47 | <MikeSmith> | but I guess I still will avoid it |
| 02:48 | <tantek> | caitp - sure, can't absolve responsbility for it, on either side. |
| 02:48 | <tantek> | it's a false dichotomy to try to frame as it one person's "fault" or the others |
| 02:48 | <tantek> | they're both at fault :P |
| 02:48 | <MikeSmith> | but I'll avoid saying it mostly just because there's no humor value in it, in that context |
| 02:49 | <MikeSmith> | nobody's ever at fault for taking an opportunity to say something that's genuinely funny and deeply irreverant at the same time |
| 02:49 | <MikeSmith> | you don't get opportunities like that very often |
| 02:49 | <MikeSmith> | so you take them when you get them |
| 02:49 | <MikeSmith> | like inspiration from God |
| 02:50 | <MikeSmith> | the only thing I fault people for is not being funny |
| 02:50 | <MikeSmith> | being offensive without being funny is .. unrefined |
| 02:51 | <caitp> | but that's the thing, funny is pretty subjective, and it depends on a lot of things |
| 02:51 | <MikeSmith> | people who make really good humor are to be treasured |
| 02:51 | <caitp> | peoples moods, the atmosphere of a place, how tipsy everyone is |
| 02:52 | <MikeSmith> | right |
| 02:52 | <caitp> | professional comedians figured out they can't do "objectively funny", so they do the next best thing |
| 02:52 | <MikeSmith> | some people suck at being funny, I'll concede |
| 02:52 | <MikeSmith> | sometimes we all suck at being funny |
| 02:52 | <MikeSmith> | but that means we just need to try harder |
| 02:53 | <MikeSmith> | geez |
| 02:53 | <MikeSmith> | "objectively funny" is complete nonsense |
| 02:54 | <MikeSmith> | if you say|draw|meme really funny, maybe 75% percent of the people who read it are going to be offended |
| 02:54 | <MikeSmith> | maybe 75% of them deserve to be offended |
| 02:54 | <MikeSmith> | they need to check their offense |
| 02:55 | <MikeSmith> | that's the point |
| 02:55 | <caitp> | well this is part of why I say it doesn't work very well on the web |
| 02:55 | <MikeSmith> | that's often the message: check your offense |
| 02:55 | <caitp> | nobody really gives a shit about being an effective communicator on the web |
| 02:55 | <MikeSmith> | I think it does work pretty well, if it makes me laugh and it makes other people laugh, then it's worked |
| 02:57 | <MikeSmith> | of course qualified by, if it's not hateful or trying to cause real harm or pain to somebody |
| 02:58 | <MikeSmith> | people aren't caused real pain by, e.g., seeing rude depictions of their holy figures, or hearing crude jokes about them |
| 02:58 | <MikeSmith> | or by hearing me say "fucking christ" or whatever |
| 06:49 | <MikeSmith> | hsivonen: https://github.com/validator/validator/issues/22#issuecomment-69128876 |
| 06:51 | <MikeSmith> | seems the TLS handshake with the vnu backend fails if site has TLS 1.0 disabled (that is only TLS 1.1+ enabled) |
| 06:55 | <MikeSmith> | hsivonen: I wonder if upgrading the jetty version might fix it |
| 06:55 | MikeSmith | will try that locally |
| 07:05 | <MikeSmith> | hsivonen: stack traces https://gist.github.com/sideshowbarker/21fc8a22ad3cad41eb74 https://gist.github.com/sideshowbarker/7fd83a3791f4702c2402 |
| 08:16 | <annevk> | Is @dimitriid asking for HTML5 Super Friends to come save the day? |
| 08:48 | <Ms2ger> | gsnedders, verboten? (re twitter) |
| 09:03 | <MikeSmith> | annevk: only after the first phase of his plan, which is to put Apple and "guys who created SASS" in charge of enforcing new standards and implementing everything and to "work on specs for a year" |
| 09:03 | <MikeSmith> | then webdevs just step in reap all the obvious benefits from the outcome of that |
| 10:17 | <hsivonen> | MikeSmith: commented on github |
| 10:19 | <MikeSmith> | hsivonen: thanks |
| 10:20 | <hsivonen> | It's kinda amazing that the Apache HttpClient developers have enough time and enthusiasm to keep redesigning and rewriting their project as much as they do |
| 10:21 | <hsivonen> | and then they resisted fixing the important stuff like SNI |
| 10:21 | <hsivonen> | (now fixed, but the resistance to fixing it was sad) |
| 10:28 | <annevk> | hsivonen: resistance? |
| 10:33 | <hsivonen> | annevk: https://issues.apache.org/jira/browse/HTTPCLIENT-1119 |
| 10:35 | <MikeSmith> | hope it's not a pattern across Apache projects. Reminiscent of the years-long Apache web server project resistance to fixing the default content-type brokenness |
| 12:02 | <hemanth> | meow |
| 12:48 | <MikeSmith> | annevk: a webdev friend asks, "why has the w3c to reinvent js APIs that already exist in Node, only to make them worse (eg Text{En,De}coder vs Buffer) ?" |
| 12:48 | <annevk> | MikeSmith: don't think W3C had much to do with it |
| 12:48 | <MikeSmith> | where by "w3c" I think he actually means the Encoding spec |
| 12:48 | <MikeSmith> | yeah |
| 12:49 | <MikeSmith> | what can I say to help enlighten him? |
| 12:51 | <annevk> | MikeSmith: looks like Buffer is some kind of byte structure, I guess you could ask him why Node reinvented ArrayBuffer... |
| 12:51 | <annevk> | MikeSmith: not really clear to me what to say here |
| 12:52 | <MikeSmith> | yeah I don't really know what Buffer is in node |
| 12:52 | <MikeSmith> | annevk: ok |
| 12:52 | <annevk> | I'm guessing it's http://nodejs.org/api/buffer.html |
| 12:53 | <jgraham> | Often when people say "reinvented X but worse", they mean "did X but different" |
| 12:53 | <annevk> | TextEncoder/TextDecoder originates from jsbell and some WHATWG discussion. I don't know if Buffer was around back then or why people didn't look at it... |
| 12:55 | <annevk> | I believe some Node.js people have aversion to participating in standards so that might explain why it's not always used as a source of inspiration |
| 12:57 | <Domenic> | Node's Buffer was created around the same time or a little before ArrayBuffer started shipping |
| 13:01 | <ondras> | I would say node's Buffer is similar to CommonJS's Binary/F: wiki.commonjs.org/wiki/Binary/F ? |
| 13:02 | <annevk> | ArrayBuffer goes back to at least Feb 2011 |
| 13:02 | <Ms2ger> | MikeSmith, ask the webdev friend to make his suggestion before we ship? |
| 13:03 | <ondras> | ArrayBuffer is commonjs's Binary/B, right? |
| 13:05 | <annevk> | Domenic: lifecycle callbacks, do you know why they are transfered? |
| 13:05 | <annevk> | Domenic: and they are essentially statics on the class, correct? |
| 13:06 | <ondras> | http://wiki.commonjs.org/wiki/Binary |
| 13:06 | <ondras> | ah |
| 13:06 | <ondras> | loaded, finally |
| 13:07 | <annevk> | Domenic: also, can you reply to that thread with bz about Element's constructor, seems kind of relevant to figure out how we want to design that |
| 13:07 | <Domenic> | annevk: I am putting off all of this stuff until TAG F2F winds down and I can write reasoned responses |
| 13:07 | <annevk> | Domenic: when is that? |
| 13:07 | <Domenic> | tonight, plus maybe some recovery sleep pushing us in to tomorrow |
| 13:08 | <Domenic> | as for lifecycle callbacks, I don't remember (haven't had time to reboot all of this in to my mind) but one thing I remember is that it avoids calling into potentially-different user code every time. |
| 13:08 | <Domenic> | also it allows you to delete the lifecycle callbacks after they're transferred: |
| 13:08 | <Domenic> | https://github.com/domenic/html-as-custom-elements/blob/master/src/register-element.js#L23-L27 |
| 13:09 | <Domenic> | I am excited to get in to designing Element constructor though |
| 13:10 | <Domenic> | bz and I have quite different assumptions/priorities in our respective strawmen and I want to sit down and think hard about which ones are valuable to keep, which ones are in conflict, etc. |
| 13:11 | <annevk> | Domenic: it seems weird to delete them like that, can't we pass them in as arguments then? |
| 13:11 | <Domenic> | also want to see if there's any solution for the upgrading thing... that's an annoying problem |
| 13:12 | <annevk> | Domenic: and at the very least they should be symbols if we keep them on the classes... |
| 13:12 | <Domenic> | annevk: yeah I would like that. But the argument is that they're good, especially for user classes, so you can do e.g. `super.createdCallback()` in a custom element deriving from a custom element. |
| 13:13 | <annevk> | super doesn't refer to an instance? |
| 13:13 | <annevk> | I guess that makes sense |
| 13:13 | <Domenic> | not sure what refer to an instance means |
| 13:14 | <Domenic> | but you can imagine e.g. even if you do `class X extends HTMLInputElement {}` (or something less complicated) being able to call `super.attributeChangedCallback()` would be nice. If we specified HTMLInputElement.prototype.attributeChangedCallback. |
| 13:14 | <annevk> | Domenic: so you're destroying that benefit by deleting them, no? |
| 13:14 | <Domenic> | yep, in favor of spec-compliance at the moment. But easy to restore. |
| 13:15 | <annevk> | Yeah I thought we should have something like Node.clonedSymbol etc. |
| 13:15 | <hemanth> | Domenic, I'm interested in contributing to CustomHTMLMediaElementImpl of html-as-custom-element, is there any TODO list for the same :) ? |
| 13:15 | <annevk> | to allow for such a future |
| 13:15 | <annevk> | hemanth: PR https://github.com/domenic/html-as-custom-elements/ I guess |
| 13:15 | <Domenic> | hemanth: heh, it's such a huge project, I got scared off. I can try to write up the missing pieces, but even just getting CustomHTMLMediaElementImpl working is huge, before even starting on CustomHTMLVideoElement |
| 13:16 | <hemanth> | annevk, thanks to was very helpful ;) |
| 13:17 | <Domenic> | https://github.com/domenic/html-as-custom-elements/blob/master/src/elements/CustomHTMLMediaElement-impl.js is all just stubs. Maybe there are tests we could find and start passing easy ones like playbackRate or something. |
| 13:17 | <hemanth> | Domenic, agree, but there must be some strategy to bring down this beast... |
| 13:17 | <hemanth> | +1 |
| 13:18 | <Ms2ger> | Domenic, guess what, there's a repository full of tests for the web platform |
| 13:18 | <zcorpan> | just change the spec to HTMLMediaElement : HTMLElement {}; |
| 13:18 | <annevk> | hemanth: could start with determining all the internal state |
| 13:18 | <hemanth> | annevk, sure, starting with? |
| 13:19 | <annevk> | hemanth: for HTMLMediaElement, then fill in the various algorithms to manipulate said state, etc. |
| 13:20 | <Domenic> | Ms2ger: yeah, with a whole 2 tests for HTMLSelectElement/HTMLOptionElement; I have lost faith since finding that. Perhaps unfairly. |
| 13:21 | <Domenic> | annevk: one thing that would be nice is separate classes for every tag name, as you alluded to. If we can pull that off a lot of things get easier I think. |
| 13:22 | <Ms2ger> | Domenic, I think I wrote them all :) |
| 13:22 | <Ms2ger> | Domenic, but video/audio is much better |
| 13:22 | <Domenic> | that is good news :) |
| 13:22 | <annevk> | Domenic: either way things would work |
| 13:23 | <annevk> | Domenic: but first I need to know about your plan vs bz' plan, then we can figure out how things trickle down |
| 13:23 | <hemanth> | annevk, idl will have the clue right? |
| 13:23 | <Ms2ger> | Domenic, and there's 8 for select+option now ;) |
| 13:24 | <hemanth> | Domenic, I really liked that tool that converts idls to js classes <3 :) |
| 13:24 | <annevk> | hemanth: IDL does a lot, but not that much |
| 13:25 | <hemanth> | annevk, need to dig the spec then... |
| 13:26 | hemanth | rubs his eyes reading https://github.com/WebKit/webkit/blob/master/Source/WebCore/html/HTMLMediaElement.cpp ^_^ |
| 13:30 | <hemanth> | annevk, what method would you suggest to determine the internal states? |
| 13:32 | <annevk> | hemanth: I'd read the spec |
| 13:32 | <Domenic> | they usually say things like "has an associated foo" |
| 13:33 | <Domenic> | https://html.spec.whatwg.org/multipage/embedded-content.html#media-element |
| 13:33 | <Domenic> | E.g. "All media elements have an associated error status" is saying that there is an internal slot [[AssociatedError]] on all HTMLMediaElement instances |
| 13:34 | <Domenic> | "The error attribute, on getting, must return the MediaError object created for this last error, or null if there has not been an error." is saying that you define `get error() { this@[[AssociatedError]] === undefined ? null : this@[[AssociatedError]]; }` |
| 13:34 | <Domenic> | where this@[[AssociatedError]] in my ES code ends up being something more like p(this).AssociatedError, if I recall |
| 13:34 | <Ms2ger> | What's the obsession with [[]] anyway |
| 13:35 | <Ms2ger> | Is that because Word can't do links? |
| 13:35 | <Domenic> | O_O it has nothing to do with links |
| 13:36 | <Ms2ger> | Then why? |
| 13:36 | <Domenic> | It is a notation for denoting internal slots as opposed to regular properties |
| 13:36 | <Ms2ger> | Regular props are already in quotes |
| 13:37 | <Domenic> | Not in ES they aren't |
| 13:37 | <Domenic> | Nor in most specs it seems... |
| 13:37 | <Ms2ger> | Really |
| 13:37 | <Domenic> | "The error attribute, on getting" no quotes |
| 13:37 | <Ms2ger> | Most specs don't access "regular props" |
| 13:37 | <Domenic> | they talk about them though |
| 13:37 | <hemanth> | Cool, thanks for the heads up Domenic, will paw more at it... |
| 13:38 | <annevk> | In a world of markup I don't think we need to use [[ and ]] (could be stylesheet that adds them if you really insist) but having something is useful |
| 13:39 | <Domenic> | Yeah I mean it's shorter to type than <span class="internal-slot">AssociatedError</span>, certainly. |
| 13:39 | <Domenic> | annevk: btw you may find https://github.com/domenic/html-as-custom-elements/blob/master/src/elements/URLUtils-impl.js interesting |
| 13:39 | <Ms2ger> | Maybe use [[Foo]] in the bikeshed source |
| 14:35 | <Ms2ger> | MikeSmith, what's "physical web" supposed to mean? Like a spider web? |
| 14:36 | <MikeSmith> | buff |
| 14:38 | <MikeSmith> | it's about taking the Web outside to play sports, or at least get some direct sunshine once in a while |
| 14:38 | <MikeSmith> | or climbing a mountain with annevk |
| 14:39 | <Domenic> | NO DAD I WANT TO STAY INSIDE AND PROGRAM MY COMPUTERS OK!? |
| 14:40 | <MikeSmith> | no son today we go out and chop some wood |
| 14:41 | <MikeSmith> | I thought the page at https://google.github.io/physical-web/ used to say a lot more than it does now |
| 14:44 | <MikeSmith> | "the Physical Web is a discovery service: a smart object broadcasts relevant URLs that any nearby device can receive" https://github.com/google/physical-web#the-physical-web |
| 14:44 | <MikeSmith> | Ms2ger: 👆 |
| 14:45 | <caitp> | what could possibly go wrong |
| 14:46 | <MikeSmith> | heh |
| 14:47 | <MikeSmith> | good times await |
| 14:47 | <MikeSmith> | caitp: sorry for ranting yesterday |
| 14:47 | <MikeSmith> | I was a little wound up |
| 14:47 | <caitp> | it's all good |
| 15:17 | <annevk> | "An Update on ES6 classes in V8" it's interesting that what I was afraid of happened, except it happened so fast that nobody got hurt |
| 15:18 | <Ms2ger> | annevk, ? |
| 15:19 | <annevk> | Ms2ger: Chrome was planning on shipping a conservative subset of ES6 classes; I was afraid that once we got to subclassing builtins and DOM that subset might have to change; people got together and changed the subset |
| 15:22 | <caitp> | it was a christmas miracle |
| 18:33 | <jamesr__> | wow https support in python is worse than i thought |
| 18:34 | <jamesr__> | at least if you actually care about making sure the connection is to what you expect |
| 19:16 | <jgraham> | jamesr__: Yes, I have heard that. Is requests any better? |
| 19:28 | <jamesr__> | it's supposed to be, but deploying it to everyone who wants to run the script i have to worry about is tricky |
| 21:01 | <jamesr__> | oh hey, IDB is a W3C rec |
| 21:39 | <Hixie> | https://medium.com/@dmitriid/ok-w3c-and-whatwg-dont-die-but-7952221fcbe4 |
| 21:39 | <Hixie> | (in particular, note "I also have much less gripe with WHATWG, than with W3C") |
| 21:44 | <Domenic> | Wow how did the same person suddenly become much more reasonable. |
| 21:45 | <Domenic> | Haha annevk "And what a surprise, there’s actual evidence that the author of the spec actually was<https://annevankesteren.nl/about>; a web-developer (hence the useful spec, I presume)." |