| 09:38 | <tobie> | annevk: what does "Meta:" stand for in your commit message? |
| 09:42 | <annevk> | tobie: a commit that does not affect the standard itself |
| 09:45 | <annevk> | tobie: https://github.com/whatwg/html/pull/2043 |
| 09:46 | <tobie> | annevk: ty |
| 09:54 | <robertkowalski> | terinjokes: Domenic: do i miss something obvious here: https://github.com/whatwg/console/issues/35#issuecomment-260292521 |
| 10:03 | <Ms2ger> | annevk, is the first change in https://github.com/servo/servo/pull/14126/files#diff-d582538526f2d8ebdedf40a39f238c35L346 correct? |
| 10:04 | <annevk> | Ms2ger: I don't think so |
| 10:05 | <annevk> | Ms2ger: https://github.com/w3c/web-platform-tests/commit/a6d3d9bfb6b4453e92e179058b33d92d2623efd3 is the last change to that file |
| 10:05 | <annevk> | Ms2ger: where is this change originating? |
| 10:05 | <Ms2ger> | That's in the servo copy, but it conflict with your change |
| 10:06 | <annevk> | Yeah, not sure what's going on, nox? |
| 10:07 | <annevk> | tobie: btw, it'd be easier to review if you create your branches directly on heycam/webidl |
| 10:07 | <tobie> | annevk: how so? |
| 10:07 | <nox> | annevk: What? |
| 10:08 | <annevk> | nox: see above |
| 10:08 | <tobie> | annevk: I think it's a bad habit. Creates a culture of those with commit rights to the repo and those without. |
| 10:08 | <annevk> | tobie: can then check them out more easily |
| 10:08 | <annevk> | tobie: I wish GitHub would solve that somehow |
| 10:09 | <annevk> | with fake branches or whatever |
| 10:09 | <nox> | annevk: Sure but I don't understand what you are unsure about. |
| 10:09 | <tobie> | annevk: adding a remote branch isn't super difficult |
| 10:09 | <annevk> | nox: well, the change Ms2ger pointed to |
| 10:09 | <annevk> | nox: https://github.com/servo/servo/pull/14126/files doesn't match the spec change |
| 10:09 | <nox> | annevk: I know, but you said you don't think it's correct. |
| 10:09 | <nox> | annevk: Uh really? |
| 10:10 | <nox> | annevk: Which part? |
| 10:10 | <annevk> | nox: the first test |
| 10:10 | <annevk> | nox: see the change I made |
| 10:10 | <nox> | How does it not follow the spec? |
| 10:10 | <annevk> | nox: https://github.com/w3c/web-platform-tests/commit/a6d3d9bfb6b4453e92e179058b33d92d2623efd3 |
| 10:11 | <annevk> | nox: the host setter doesn't reset port, basically |
| 10:13 | <nox> | annevk: Ah. |
| 13:14 | <zcorpan> | MikeSmith: http://stackoverflow.com/questions/29788868/w3c-validation-of-srcset - would it be a good idea to link to https://ericportis.com/posts/2014/srcset-sizes/ in the "all image candidate strings must specify a width" error message? |
| 14:11 | <zcorpan> | with no styles, no scripts, no external resources, STP loads single-page in about 1.5s |
| 14:13 | <zcorpan> | chrome about 12s, firefox about 15s |
| 14:24 | <zcorpan> | according to devtools safari doesn't spend any time at all on layout |
| 14:29 | <Ms2ger> | Do foo.com:1234 and foo.com:4321 share cookies? |
| 14:33 | <Ms2ger> | I'm told they do |
| 14:37 | <annevk> | Ms2ger: yes |
| 14:38 | <annevk> | Ms2ger: cookies don't care about origins as much |
| 15:07 | zcorpan | documented perf findings in https://bugs.chromium.org/p/chromium/issues/detail?id=312527#c34 |
| 15:36 | <MikeSmith> | zcorpan: yeah I think linking to https://ericportis.com/posts/2014/srcset-sizes/ from that error message would be good. Please file a validator bug for it. |
| 15:37 | <MikeSmith> | I’m gonna try to work on validator code next week, get through the open bugs |
| 15:38 | <zcorpan> | filed |
| 15:38 | <zcorpan> | MikeSmith: awesome |
| 15:40 | <MikeSmith> | thanks |
| 15:40 | <MikeSmith> | annevk: when you have time can you maybe please respond to https://stackoverflow.com/questions/40580913/fetch-api-custom-request-headers-cors-and-cross-origin-redirects |
| 15:41 | <MikeSmith> | (or anybody else here who has time to answer that) |
| 15:41 | <MikeSmith> | I have tried to read up on it myself but not confident enough to answer |
| 15:42 | <MikeSmith> | in particular the “Why does this restriction exist? What security issue is prevented by not following and running pre-flight on the followed URL?” (for redirects on requests with custom headers) |
| 15:43 | <MikeSmith> | as far as relevant spec citations, I found http://stackoverflow.com/questions/16386148/why-browser-do-not-follow-redirects-using-xmlhttprequest-and-cors/20854800#20854800 but that is to the obsolete CORS spec, and I am not sure what part of Fetch to cite that corresponds to that |
| 15:44 | <MikeSmith> | also found http://stackoverflow.com/questions/18539403/chrome-cancels-cors-xhr-upon-http-302-redirect/38810391#38810391 which seems to cite the same part of the old CORS spec |
| 15:45 | <MikeSmith> | once I understand this better myself I will try to write up something to add to MDN explaining it |
| 15:45 | <annevk> | MikeSmith: done |
| 15:45 | <MikeSmith> | thanks |
| 15:49 | <MikeSmith> | annevk: ah yeah I see this is from https://github.com/whatwg/fetch/issues/204 and I remember I asked you about it after coming across another SO question not so long ago |
| 15:49 | <MikeSmith> | eventually I guess it will actually sink in and I’ll remember |
| 15:51 | <annevk> | MikeSmith: maybe SO will get smarter about duplicates too |
| 15:55 | <MikeSmith> | annevk: yeah |
| 15:56 | <MikeSmith> | though in practice now I guess the current mechanism for making SO smarter about duplicates is for people to get smarter enough (and remember enough) to recognize something as a duplicate, and reference what it duplicates |
| 15:58 | <MikeSmith> | anyway it seems like people are still going to run into the problem and not know why or how to get around it |
| 15:58 | <MikeSmith> | at least until it gets fixed in Chrome (since it has already been fixed in Firefox) |
| 15:59 | <MikeSmith> | hope to find a way to |
| 16:00 | <MikeSmith> | hope to find a way to keep others from wasting time trying to figure it out, and getting frustrated |
| 16:00 | <MikeSmith> | seems like getting the info into SO is the best way to do that (even if it means repeating it) |
| 16:06 | <annevk> | Yeah, I don't mind, it's not a lot of work |
| 17:05 | <tobie> | Domenic, rbyers: think one of you could check if there's anything I could do to make it easier for Chrome folks to triage WebIDL bugs (see: https://bugs.chromium.org/p/chromium/issues/detail?id=662379#c3) |
| 17:16 | <Domenic> | tobie: hmm rbyers might know more but that looks good to me, it's already gotten the attention of the relevant folks. Not sure why that person came in and added the needs triage label... |
| 17:16 | <tobie> | Domenic: OK, cool. Not sure either. Hence the question. :) |
| 19:05 | <KiChjang> | why is WHATWG not publishing standards for CSSOM? |
| 19:32 | <TabAtkins> | Because the CSSWG is working reasonably well for it? |
| 19:38 | <KiChjang> | fair enough, i'm just wondering whether they play nice to WHATWG standards |
| 20:35 | <younder> | I'm a newbie to this group, but from what past history suggests WHATWG does influence W3C. I mean there never was a XHTML2 standard. Instead we have HTML5. |
| 20:36 | <KiChjang> | W3C still likes to do their own thing and waste their time |
| 20:36 | <younder> | Als Håkon Lie the original creator of CSS works for Opera which was one of the original founders of WHATWG. |
| 20:37 | <KiChjang> | http://w3c.github.io/charter-html/group-charter.html |
| 20:37 | <KiChjang> | tell me how this is different than WHATWG's html standard? |
| 20:37 | <botie> | will do |
| 20:48 | <gsnedders> | KiChjang: also the fact that nothing happened with CSSOM for a long time wasn't politics: it was having nobody willing to work on it. the CSS WG as a venue was never the limiting factor. |
| 20:49 | <KiChjang> | huh, that's weird |
| 20:50 | <KiChjang> | i'm guessing web devs aren't giving it as much attention? |
| 22:14 | <zcorpan> | interesting, in safari the multipage link is rendered sooner than the rest |
| 22:16 | <zcorpan> | unclear to me why the rest of the "above the fold" is not rendered at the same time |
| 22:37 | <Domenic> | layers created by transform? |
| 22:39 | <mounir> | someone here knows the story behind "stopImmediatePropagation"? (ie. why was it added?) |
| 22:58 | <zcorpan> | Domenic: yeah. header { transform: rotateZ(0deg); } makes the full header appear early (and the toc a bit later) |
| 23:03 | <zcorpan> | though chrome renders everything early without a transform hack |