| 07:26 | <sideshowbarker> | freddy: About the new SRI FWPD, I guess there aren’t any plans to include a list of any substantive/non-editorial changes (if there have been any) since the Rec publication? |
| 07:31 | <Meghan Denny> | should https://html.spec.whatwg.org/#the-details-and-summary-elements be using :first-child to align with the language of context in https://html.spec.whatwg.org/#the-summary-element ? |
| 07:35 | <annevk> | Where should it use that? |
| 07:36 | <Meghan Denny> | making the selectors details > summary:first-child and details[open] > summary:first-child respectively |
| 07:38 | <annevk> | I see, no, this is intentional. |
| 07:42 | <Meghan Denny> | i see, the details shadow tree description clarifies below |
| 07:51 | <vrafaeli> | So WHATWG does not deal with the Manifest, or does it? |
| 08:04 | <annevk> | vrafaeli: not really, no. |
| 10:13 | <akaster> | Any encoding spec editor thoughts on https://github.com/whatwg/encoding/issues/324#issuecomment-2816666810 ? Seems that the big 3 have aligned on 0-byte read/write for detached array buffers and encodeInto(). Not handling this case was causing a crash in ladybird, which we fixed by doing what everyone else does instead of the proposed throwing behavior |
| 10:40 | <annevk> | akaster: I was trying to think it through this morning, but got distracted. Looking at it now I think the question is what destination's byte length should return. And we should probably accept what is implemented. Perhaps Web IDL should make that return 0 for the detached case? |
| 10:45 | <akaster> | That would address the underlying issue where every spec dealing with ArrayBuffer objects needs updating for detached objects right? Though personally I would prefer if "invalid operation for this object" was actually invalid instead of a no-op but 🤷♂️ |
| 10:55 | <annevk> | Whether it's invalid depends on who you ask I suspect. The JS API also returns 0, iirc. |
| 10:55 | <annevk> | I suppose we could sprinkle detached checks all over too. As long as we're consistent that would also work for me. |
| 11:26 | <annevk> | akaster: I think https://webidl.spec.whatwg.org/#buffersource-byte-length might be broken? [[ArrayBufferByteLength]] should already be 0 for detached buffers, so that should work fine. But [[ByteLength]] on views is not even defined for DataView?! And can also contain the value AUTO? It seems we'd hit all kinds of complications on step 2. |
| 11:34 | <annevk> | Oh no, I think it works. But not entirely sure about the detached view case. |
| 11:34 | <annevk> | Though I suspect AUTO is used for growable things, which means it doesn't work for those. |
| 15:05 | <vrafaeli> | How do you guys decide what is under the domain of W3C and what under WHATWG? |
| 15:09 | <annevk> | akaster: I updated the relevant issues. At this point I think fixing https://github.com/whatwg/webidl/issues/1385 is the way to go. |
| 15:10 | <akaster> | thanks! WebIDL issues are always fun, because every time I suggest fixing one I get warnings that "if you fix that, you'll end up the WebIDL guy" |
| 15:11 | <annevk> | akaster: could be true, but multiple people have thus far successfully evaded that outcome too |
| 15:12 | <annevk> | vrafaeli: https://www.dictionary.com/e/you-guys/ and "it depends"; https://github.com/w3c/whatwg-coord has some historical context |
| 15:12 | <akaster> | surely someone will fail the dodge eventually |
| 15:13 | <annevk> | that's the hope |
| 15:13 | <akaster> | ♨️🥔🤾♂️ |
| 16:00 | <annevk> | Noam Rosenthal: I was assuming you'd still address the remaining comments on https://github.com/whatwg/fetch/pull/1647. Is that still on your radar? |
| 18:03 | <Dominic Farolino> | Who all here, if anyone, is going to Web Engines Hackfest this year (June 2-4 in Spain)? |
| 18:07 | <bkardell> | ✋️ |
| 18:09 | <smaug> | I should decide rather soon |
| 20:40 | <jmdyck> | https://webengineshackfest.org/ |
| 21:48 | <jarhar> | Anyone have feedback for this? https://github.com/whatwg/html/pull/11186 @keithamus:matrix.org |