| 06:33 | <hsivonen> | MikeSmith: thanks |
| 07:33 | <zcorpan> | mkwst: is http://web-platform.test:8000/html/semantics/embedded-content/the-iframe-element/iframe_sandbox_popups_nonescaping-3.html a valid test? why is the child frame allowed to navigate its popup? |
| 09:36 | <nox> | annevk: Did I make sense? |
| 09:37 | <annevk> | Yeah, but there's some communication mismatch |
| 09:37 | <hsivonen> | zcorpan: yeah, I think we should make <menuitem> parse like <span>. And the same thing for other new elements that browsers already parse like <span> |
| 09:40 | <nox> | annevk: I just noticed someone played the business interest card, |
| 09:40 | <nox> | I should not even have bothered. |
| 09:41 | hsivonen | wonders what the context is |
| 09:41 | <nox> | hsivonen: https://github.com/whatwg/url/issues/118#issuecomment-276072218 |
| 09:42 | <nox> | hsivonen: This happened yesterday btw: https://github.com/servo/servo/pull/15303 |
| 09:42 | <nox> | Way too broken for now, but baby steps. |
| 09:44 | <hsivonen> | nox: nice (for the latter URL) |
| 09:44 | <nox> | "the successful HTML coup" |
| 10:10 | <zcorpan> | hsivonen: ok, thanks |
| 10:20 | <zcorpan> | hsivonen: to be clear, it should parse like an unknown element, not like <span>, yes? <span> breaks out of <svg> |
| 10:21 | <hsivonen> | zcorpan: yeah. good point. |
| 10:22 | <hsivonen> | zcorpan: has the initiative to make <video> magic in <svg> been buried yet? |
| 10:22 | <zcorpan> | hsivonen: i think the issue is still open, but i haven't seen any implementor interest |
| 10:26 | <hsivonen> | zcorpan: do you happen to have the issue URL at hand? |
| 10:27 | <zcorpan> | https://github.com/whatwg/html/issues/919 |
| 10:35 | <zcorpan> | hsivonen: sorry about driving <menuitem> parsing in a direction y'all didn't like, but i didn't really get clear messages earlier. |
| 10:42 | <hsivonen> | zcorpan: sorry. I should have been more awake about HTML parsing stuff while working on other stuff |
| 11:31 | <nox> | All these random people participating in that issue... |
| 11:35 | <Ms2ger> | zcorpan, can you file a chrome bug about its failure in http://w3c-test.org/html/semantics/tabular-data/the-table-element/caption-methods.html ? |
| 11:35 | <zcorpan> | Ms2ger: sure |
| 11:44 | <hsivonen> | People want BNF. Give people BNF and people are unhappy with BNF. |
| 11:46 | <zcorpan> | Ms2ger: https://bugs.chromium.org/p/chromium/issues/detail?id=687116 |
| 11:49 | <annevk> | jgraham: how do I use _raw_headers? |
| 11:52 | <zcorpan> | 118 keeps on giving. but on the plus side, i think there's examples in there now we can put in the spec to close the issue |
| 11:54 | <annevk> | jgraham: it seems I can str() it to get it to output something |
| 12:13 | <annevk> | jgraham: I filed https://github.com/w3c/wptserve/issues/112 on making it an official API |
| 14:57 | <annevk> | Domenic: are you around? |
| 14:57 | <annevk> | Domenic: <script type=module src=/x> <script type=module src=/x> yields one module fetch, right? Does the same go for module worker fetches? |
| 15:00 | <annevk> | Domenic: or is that only if /x is imported? |
| 15:16 | <annevk> | Domenic: see https://github.com/WebAssembly/design/issues/972 if you want more fun times |
| 15:17 | <annevk> | Domenic: I've decided I know the answers to the above questions to an okayish extent |
| 15:31 | <smaug> | foolip: SharedWorkerStart in https://www.chromestatus.com/metrics/feature/timeline/popularity/5 means calling the start() method ? |
| 15:39 | <foolip> | smaug: no constructor actually (maybe the counter location was moved) |
| 15:39 | <smaug> | foolip: but ok, basically usage |
| 15:39 | <foolip> | smaug: yeah, a fair estimate of real usage (counting is before exception handling) |
| 15:41 | <annevk> | foolip: would that also count adding a onmessage handler which implicitly calls start? |
| 15:43 | <foolip> | annevk: that's MessagePort#start(), right? |
| 15:43 | <annevk> | foolip: yeah |
| 15:43 | <foolip> | let me see |
| 15:43 | <annevk> | foolip: what other start() is there though? |
| 15:45 | <foolip> | Probably none, just checking that I was looking at the right thing. At least setting onmessage does call start(), I can confirm, but it doesn't look like start() has any connection to the SharedWorker constructor? |
| 15:46 | <foolip> | You have to use the constructor to get to any of these APIs, right? |
| 15:46 | <foolip> | (Sorry, I've never tried using SharedWorker) |
| 15:47 | <annevk> | SharedWorker has a port which would have start() on it |
| 15:52 | <foolip> | right, ok. so the use counter is only for the SharedWorker constructor. https://www.chromestatus.com/metrics/feature/timeline/popularity/782 is for MessagePort#start(), but only for explicit calls. The implicit call by setting onmessage does not seem measured. |
| 15:52 | <foolip> | smaug: what are you trying to gague the risk of changing? |
| 15:54 | <smaug> | not really. there was just an email thread about SharedWorker usage in Gecko. That measure is per browsing session, I think |
| 15:55 | <smaug> | didn't Safari remove SharedWorkers |
| 15:55 | <smaug> | yet usage seems to be going up |
| 16:06 | <foolip> | They did? |
| 16:06 | foolip | tested and yes they did |
| 21:44 | <zcorpan> | something i find annoying with standard.css: i use f.lux, so in the evenings, the unvisited links become almost indistinguishable from surrounding text |
| 21:44 | <zcorpan> | i suppose the same thing happens for some classes of color blindness |
| 21:45 | <jgraham> | (doesn't the latest OSX have some built in thing like f.lux but less buggy?) |
| 21:45 | <zcorpan> | iOS does, dunno about OSX |
| 21:46 | <jgraham> | Ah 10.12.4 |
| 21:46 | <jgraham> | So probably next update |
| 21:46 | <zcorpan> | nice |
| 21:47 | <zcorpan> | anyway, any great ideas for fixing link styling? |
| 21:47 | <jgraham> | Oh, you want helpful suggestions too? |
| 21:48 | <jgraham> | I think the short answer is no; I haven't got the colour temperature variation, so it's hard to tell what the problem is |
| 21:50 | <zcorpan> | the problem is that the amount of blue light emitted is much less than usual, so blue text becomes dark blue |
| 21:54 | zcorpan | files an issue with pictures |
| 21:54 | <jgraham> | Make all the links green to match the rest of the styles? |
| 22:24 | <zcorpan> | https://github.com/whatwg/resources.whatwg.org/issues/44 |
| 22:26 | <zcorpan> | jgraham: i suppose that would help me but would still be annoying for people with green-weak or green-blind color blindness :-) also might not work so well in notes with a green background and surrounding text is green |
| 23:55 | <Domenic> | annevk: small ping on https://github.com/whatwg/url/issues/97 just to make sure you didn't forget about it |