| 07:30 | <ondras> | SamB: my plan was to actually aim for jpegs with large EXIFs; EXIFs padded with arbitrary data to match an exact target file size |
| 07:40 | <zcorpan> | https://www.w3.org/Bugs/Public/show_bug.cgi?id=25136 -> http://www.xkcd.com/1172/ |
| 08:02 | <zcorpan> | jgraham: hmm, maybe t.add_cleanup(function() { throw 1 }) should just silently ignore the exception |
| 08:46 | <Ms2ger> | jsbell, if you read the logs: https://bugzilla.mozilla.org/show_bug.cgi?id=916644 |
| 08:50 | <ondras> | just out of curiosity, why shall be these ctors disabled? |
| 08:50 | <Ms2ger> | No good reason afaict |
| 08:50 | <ondras> | 'cause there are those "new" shiny Event constructors, right? |
| 08:51 | <Ms2ger> | It's "new Event()" vs "Event()" |
| 08:51 | <ondras> | aha |
| 08:51 | <ondras> | so it is not document.createElement vs. new HTMLDivElement ? |
| 08:52 | <Ms2ger> | No |
| 08:52 | <ondras> | okay then |
| 11:12 | <jgraham> | zcorpan: You win the "I crashed jgraham's computer" award for crashing my computer (kinda) |
| 11:13 | <zcorpan> | jgraham: how did i do that? |
| 11:13 | <zcorpan> | jgraham: evil test again? |
| 11:14 | <jgraham> | (one of your tests has while(true) {postMessage(1)} in a worker and when I edited the test but accidentially forgot to call worker.terminate, everything froze until the OOM killer started taking out random processes) |
| 11:14 | <jgraham> | (I think) |
| 11:14 | <zcorpan> | interesting |
| 11:15 | <jgraham> | (I didn't actually check that the OOM killer was involved but otherwise I don't know why Chrome and the networking stack both died) |
| 11:15 | <zcorpan> | i think i didn't quite get those tests to do what i wanted them to do, not sure if everyone has a bug or they're fundamentally bogus |
| 11:16 | <jgraham> | Is the idea that worker.terminate is supposed to be sync so that after you call that no more messages are processed even if they are already in the queue? |
| 11:17 | jgraham | isn't sure he is suppsoed to guess what the spec says by reading the tests ;) |
| 11:18 | <jgraham> | It kinda seems like steps 2 and 3 in the terminate a worker algorithm should be swapped around |
| 11:33 | <gsnedders> | Having glaziers replacing window is mostly unsurprisingly full of loud bangs. |
| 11:51 | <jgraham> | gsnedders: What about having glazou replace Windows? |
| 11:52 | <gsnedders> | :) |
| 11:56 | <zcorpan> | jgraham: i don't recall how that stuff works exactly |
| 12:28 | <annevk> | Hixie: on http://www.whatwg.org/specs/ "the relevant FAQ entry" is a dead link |
| 12:28 | <annevk> | Hixie: should we put Twitter accounts and GitHub repositories there too? |
| 13:59 | <jgraham> | !summon Ms2ger |
| 13:59 | <jgraham> | As Ms2ger would say |
| 14:01 | <ondras> | nice, does he grant wishes once summoned? |
| 14:02 | <jgraham> | ondras: That's what I'm hoping at least |
| 14:02 | <ondras> | :) |
| 14:47 | <annevk> | Hixie: I recommend skimming through https://speakerdeck.com/dherman/status-report-es6-modules |
| 15:11 | <annevk> | mathiasbynens: suggestion for http://dev.opera.com/articles/view/showmodaldialog/ |
| 15:12 | <annevk> | mathiasbynens: add a bit on how something in a standard can still be deemed broken and removed at a future date of compatibility analysis changes |
| 15:12 | <annevk> | mathiasbynens: now on the standards side we should probably better identify those cases |
| 15:22 | <mathiasbynens> | annevk: sounds good. will do |
| 15:35 | <SamB> | ondras: annevk I assume s/date of/date if/ ? |
| 15:35 | <SamB> | er. |
| 15:35 | <SamB> | darn it |
| 15:35 | <SamB> | that was for annevk |
| 15:35 | <annevk> | yes |
| 15:35 | <annevk> | btw, did we sort out the logo situation? |
| 15:36 | <annevk> | SamB: should I just commit a CONTRIBUTORS and LICENSE file as per that bug? |
| 15:37 | <SamB> | well, Wikimedia Commons thinks that the WHATWG logo itself (and rasters of it) is too simple to be copyrighted |
| 15:37 | <TabAtkins> | SamB: That's... probably incorrect. |
| 15:37 | <SamB> | which would basically leave the only potentially copyrighted thing in logo.svg being the path for the question mark |
| 15:38 | <SamB> | TabAtkins: oh? |
| 15:39 | <TabAtkins> | Actually, I'm not sure about copyright. It's certainly fine to be trademarked. |
| 15:40 | <SamB> | sure, sure |
| 15:40 | <TabAtkins> | But copyright can apply to very small pieces of text. |
| 15:40 | <TabAtkins> | And I don't think that the art application would exclude something like the WHATWG logo, which has specific design/color/etc choices that would distinguish it from other similar things. |
| 15:43 | <SamB> | that sounds trademarky to me |
| 15:46 | <TabAtkins> | It's possible. I'm not a copyright lawyer. Just sounds fishy to me, is all. |
| 15:47 | <SamB> | anyway, Matthew Raymond would presumably not get to claim copyright based on the choice of color, and the current logo doesn't even use the same question mark, does it? |
| 15:48 | <SamB> | (Hixie is the one who said that it should be that color) |
| 15:48 | <TabAtkins> | No clue. |
| 15:49 | SamB | wants an "onionshell" tool for any two image URLs ... |
| 15:53 | <zcorpan> | does anyone know what the implications of http://www.whatwg.org/specs/web-apps/current-work/multipage/rendering.html#bidi-rendering are to review https://critic.hoppipolla.co.uk/4a7a86f8?review=886 ? |
| 15:54 | <annevk> | SamB: added a README.md with the necessary info |
| 15:57 | <SamB> | TabAtkins: anyway, I uploaded the original PNG to https://commons.wikimedia.org/wiki/File:WHATWG_logo_(Matthew_Raymond).png and then proposed that it be deleted; https://commons.wikimedia.org/wiki/Commons:Deletion_requests/File:WHATWG_logo_(Matthew_Raymond).png is the resulting discussion |
| 16:00 | <annevk> | SamB: btw, I used http://www.fileformat.info/convert/image/svg2raster.htm to create the PNG images, it is apparently terrible at it |
| 16:01 | <SamB> | annevk: at least it parsed logo-xhr.svg correctly |
| 16:07 | <SamB> | annevk: unlike inkscape and rsvg both |
| 16:07 | <annevk> | yeah that was weird |
| 16:11 | <SamB> | ah, so that uses batik? |
| 16:12 | SamB | goes to compare that with inkscape ... |
| 16:14 | <SamB> | hmm, does that not have a CLI tool? |
| 16:20 | <dglazkov> | good morning, Whatwg! |
| 16:32 | <SamB> | so it looks like batik does ship a command-line rasterizer as /usr/bin/rasterizer :-( |
| 23:32 | <MikeSmith> | https://twitter.com/pcwalton/status/449299846873108480 👈 pretty cool |
.png){kind=link}
.png){kind=link}