| 01:54 | <Hixie_> | anyone got any suggestions for how ATs should handle exposing headings on pages like www.terminix.com ? |
| 01:55 | <Hixie_> | looks like they just changed <div> for <section> for no apparent reason |
| 01:56 | <SamB> | (the solution may involve a large trout?) |
| 07:33 | <zcorpan> | jgraham: ping https://critic.hoppipolla.co.uk/r/1464 |
| 07:51 | <foolip_> | nessy: are Mondays still your WebVTT day? |
| 07:51 | <foolip> | now that I'm back in Sweden the time overlap is a lot smaller... |
| 08:20 | <annevk> | http://xkcd.com/1367/ <3 |
| 08:21 | annevk | is back on the +1 timezone too |
| 08:28 | <foolip> | annevk: nice xkcd :) |
| 08:29 | <foolip> | annevk: which is the least objectionable w3.org for the DOM fork? https://github.com/silviapfeiffer/webvtt-spec/commit/512ed724dd26e78fdc6c24329b2f70077a7e8fb7 |
| 08:29 | <foolip> | +URL |
| 08:29 | <annevk> | They're all pretty objectionable |
| 08:30 | <JakeA> | hah |
| 08:30 | <annevk> | But if I was forced to pick, http://w3c.github.io/dom/ maybe? |
| 08:31 | <zcorpan> | foolip: can't you reference the whatwg one? |
| 08:31 | <foolip> | zcorpan: I'd love to |
| 08:31 | <foolip> | I'll just ask nessy then |
| 08:31 | <jgraham> | zcorpan: Done |
| 08:31 | <zcorpan> | thx jgraham |
| 08:39 | <JakeA> | annevk: Any idea how "ready" ES6 maps & iterators are? |
| 08:39 | <annevk> | JakeA: ES6 maps are done |
| 08:40 | <JakeA> | annevk: Done like promises were, or actually done? :D |
| 08:40 | <annevk> | JakeA: iterators are more or less done, though there's a discussion on generators going on still on es-discuss I think |
| 08:40 | <annevk> | JakeA: Map is shipped by multiple implementations |
| 08:40 | <JakeA> | Gotcha |
| 08:41 | <JakeA> | annevk: Worried about iterators and ServiceWorker. Don't want them to hold us up. |
| 08:41 | <annevk> | JakeA: they want to sign off on ES6 within a couple of months, so they better not change much |
| 08:41 | <annevk> | JakeA: the AsyncMap idea does not make much sense as Domenic_ explained I think |
| 08:42 | <annevk> | JakeA: and returning a promise for an iterator does not make much sense either, what you really want is an asynchronous iterator, which is ES7-maybe material |
| 08:44 | <JakeA> | annevk: Having an async way to get stuff from a cache is a must & we can't wait for ES7 |
| 08:44 | <annevk> | JakeA: is iteration a must for v1? |
| 08:45 | <annevk> | JakeA: we're discussing two things here; 1) importance of AsyncMap (does not seem important) 2) importance of iteration and whether that should be async-per-item or async-for-all-items-at-once |
| 08:46 | <JakeA> | annevk: I'm at-best indifferent to AsyncMap |
| 08:47 | <JakeA> | annevk: In terms of iteration, I'm happy to drop forEach, but need to keep values/keys |
| 08:48 | <annevk> | JakeA: what I mean by async-iterator btw is that you get a promise of sorts for each value, meaning you can get the initial value much quicker |
| 08:48 | <annevk> | JakeA: the current approach requires getting all the keys/values |
| 08:49 | <annevk> | JakeA: unfortunately there's no primitive for that yet |
| 08:49 | <annevk> | JakeA: but I think if you had an AsyncMap, it'd be more like that |
| 08:50 | <annevk> | JakeA: my suggestion would be to not bother with AsyncMap as it cannot exist without an implementation and just create a minimal cache API, explaining the constraints |
| 08:52 | <JakeA> | annevk: I don't have many use-cases for getting multiple items out of the Cache/CacheList. It's mostly for cleanup in onactivate. |
| 08:53 | <annevk> | JakeA: we could consider offering cleanup methods without exposing the primitives they are built on until we have asynchronous iteration |
| 08:54 | <annevk> | JakeA: I recommend looping in Domenic_ when he wakes up in a couple of hours as he prolly has some insight |
| 08:54 | <JakeA> | yeah |
| 08:55 | <JakeA> | annevk: Don't want to make too many assumptions around use-cases, want it open to devs as much as possible |
| 08:55 | <JakeA> | annevk: Will continue this when Domenic_ is up. Maybe slightlyoff too, he's got bigger asyncmap opinions than I do |
| 10:41 | <smaug____> | do any algorithm in specs use functions in the pseudocode |
| 10:42 | <annevk> | smaug____: what do you mean? |
| 10:42 | <smaug____> | perhaps I just need to define two different things |
| 10:42 | <smaug____> | where one uses the other one |
| 10:43 | <annevk> | smaug____: you're writing a spec? |
| 10:43 | <smaug____> | annevk: effectively "pass arguments FOO and BAR to function FOOBAR and store the return value in RETVAL" |
| 10:43 | <smaug____> | annevk: just an algorithm |
| 10:44 | <annevk> | smaug____: you can do that, but you need to be very clear the function is the initial value of some property at the start of the lifetime of the global scope |
| 10:44 | <annevk> | smaug____: so it can't be overwrriten |
| 10:45 | <smaug____> | oh, I mean function would be in pseudo-code level |
| 10:45 | <smaug____> | not a JS thing or anything |
| 10:45 | <annevk> | smaug____: and usually style is Let RETVAL be the result of invoking FOOBAR with FOO and BAR. |
| 10:45 | <annevk> | smaug____: oh, that happens all the time, see e.g. how the URL parser is invoked at various places |
| 10:47 | <smaug____> | "the result of ..." is perhaps the keyword I need |
| 10:55 | <smaug____> | but no real functions |
| 10:55 | <smaug____> | one needs to describe what happens |
| 10:55 | <smaug____> | in prose |
| 10:55 | <smaug____> | and perhaps link to somewhere |
| 11:08 | <smaug____> | ah, shadow dom event path makes events to skip window object |
| 11:24 | <nessy> | foolip_: yes, Mondays I work on WebVTT and other spec stuff |
| 11:25 | <nessy> | foolip, zcorpan: we did end up referencing the WHATWG DOM spec |
| 11:25 | <zcorpan> | nessy: cool |
| 11:25 | <Ms2ger> | Makes sense |
| 11:43 | <zcorpan> | what should i use to create/edit animated gifs? |
| 12:06 | <darobin> | zcorpan: the short answer is: you shouldn't |
| 12:07 | <darobin> | zcorpan: more helpfully, I guess it depends on what your comfortable environment is |
| 12:07 | <darobin> | on the CLI, imagemagick can create animated GIFs |
| 12:07 | <darobin> | IIRC there are also some export plugins for Photoshop |
| 12:07 | <darobin> | and a bunch of small, simple UI tools |
| 12:08 | <darobin> | last I did this was with imagemagick, but it's been a rather long while |
| 12:08 | <darobin> | crufty but it works |
| 12:08 | <zcorpan> | i don't have photoshop. i can check out imagemagick, thanks |
| 12:08 | <darobin> | zcorpan: this seems to have the right docs: http://www.imagemagick.org/Usage/anim_basics/ |
| 12:09 | <zcorpan> | darobin: perfect |
| 12:32 | <annevk> | JakeA: I msised the GitHub issue, seems Domenic_ made identical points to me |
| 13:40 | <odinho> | zcorpan: Gimp has an animation extension that I know many people has used. But again, depends on what you need to do/create it from. |
| 13:44 | <zcorpan> | Ms2ger: thanks for the chocolate! |
| 13:44 | <Ms2ger> | :) |
| 13:44 | <newbie|2> | hello ... I would like to create smooth transactions between web pages? you have any advice? some indication? |
| 13:52 | <darobin> | odinho: I'm not sure that friends let friends use Gimp :) |
| 13:57 | <jgraham> | Eh |
| 13:57 | <jgraham> | It's not so bad |
| 13:57 | <jgraham> | I mean it's hardly like telling someone to use vim |
| 13:58 | <odinho> | I quite like Gimp. :) Not as much as Inkscape, Darktable and Scribus (for different things though), but I'm quite effective in it if I need to do some manipulating :) |
| 13:58 | <darobin> | okay, in fairness I haven't tried it in years |
| 13:59 | <darobin> | but the switch from Photoshop was pretty... brutal |
| 14:03 | <jgraham> | Darktable is pretty brutal if you have tried Aperture (if I remember which one that is) |
| 14:03 | <odinho> | I'm sure, but I haven't. |
| 14:03 | <jgraham> | It was like "we almost copied the interface, but only got half the functionality. However we made it four times as complicated so you probably won't notice.: |
| 14:04 | <jgraham> | "(but will instead give up in frustration)" |
| 14:10 | <darobin> | that's a bit how Gimp felt |
| 14:11 | <odinho> | I think it has much to do with which one you do first as well. |
| 14:11 | <darobin> | "We copied pretty much all the functionality, but instead of putting it behind convenient key bindings or UI buttons we have this unique approach of a context menu seven levels deep where you can easily find everything" |
| 14:12 | <odinho> | Darktable is one of the most effective programs I've worked with for my pictures. So it can't all be bad. I like it a lot. |
| 14:13 | <jgraham> | odinho: I might like it if I hadn't previously used Lightroom. Which is what I previously meant when I said "Aperture". |
| 14:14 | <odinho> | Yes. I continue to not be amazed that software houses with $millions and many developers, photographers, ux people and the like are able to make good products. |
| 14:16 | <jgraham> | I'm not sure I was asking you to be amazed |
| 14:16 | <odinho> | You were not, I was saying something about my expectations for different software. |
| 14:17 | <odinho> | I have started to use unfree Lightworks for video editing, though. But I still dabble with the free Linux video editors now and then. But unlike every other field, the free linux video editors are just not good enough. |
| 14:18 | <jgraham> | Well that evidently depends on how much you value free / low expectations vs quality |
| 14:18 | <jgraham> | I would *love* there to be a guenuine alternative to Lightroom that was open source |
| 14:19 | <jgraham> | Maybe Darktable will become that thing |
| 14:19 | <jgraham> | Then I wouldn't have to reboot into Windows just to edit photos |
| 14:19 | <odinho> | Well, -- I seem not to know what I'm missing from Lightroom, which is just as well for me. |
| 14:19 | <jgraham> | But at the moment the quality difference is high enough that I do have to :( |
| 14:19 | <odinho> | I'm super happy with Darktable. Gave me much more than any other software I've used before did. :) |
| 14:20 | <Ms2ger> | Emscripten lightroom? :) |
| 14:20 | <odinho> | Ms2ger: Kinda hard for proprietary software, no? |
| 14:20 | <jgraham> | Ms2ger: Yeah, supposedly Adobe like the web today |
| 14:21 | <jgraham> | Although I guess they would have to emscripten Lua too, which could be interesting |
| 14:27 | <zewt> | gimp just seemed stuck in a mid-90s UI, tons of floaty windows |
| 14:27 | <jgraham> | I think they fixed that |
| 14:27 | <ondras> | recent builds have the "one window" mode |
| 14:27 | <zewt> | been quite a while since i've looked at it (nothing but photoshop is actually an option in the wild, typically) |
| 14:27 | <ondras> | OTOH some people prefer that floating approach |
| 14:28 | <ondras> | still, the Gimp is very hard to use properly. |
| 14:28 | <zewt> | photoshop's learning curve is a bit steep, but everyone has already passed it |
| 14:29 | <odinho> | I never understood the "nothing but PS is an option". I've several times asked people to give before/after pictures they've done, and allow me to do the same thing in Gimp. |
| 14:30 | <ondras> | odinho: well the shadow/highlight feature is hard to do properly in gimp |
| 14:30 | <ondras> | as an example. |
| 14:30 | <ondras> | also, things trivial in PS - text outline/shadow for instance - is complicated in gimp and not adjustable once generated |
| 14:31 | <zewt> | i get UI mockups from our artist as PSDs that I have to be able to use (and sometimes edit and send back) |
| 14:31 | <zewt> | pretty common |
| 14:31 | <zewt> | (not to say file format lock-in is a good thing, but it's there) |
| 14:32 | <odinho> | Yeah, I use Inkscape for text and vector stuff. Only use Gimp for actualy image manipulation. For adjustment Darktable is better, for vector-stuff Inkscape is better. For layout Scribus is better. One tool for each job. |
| 14:32 | <ondras> | well PSD as a web developer's input format is also mid-90s if you ask me |
| 14:32 | <zewt> | we use PSDs for all UI, works well for us |
| 14:33 | <zewt> | (not just for web, for iOS, Android, etc) |
| 14:33 | <ondras> | http://shelby.tv/video/youtube/e7-ZnHMDZto/julie-ann-horvath-because-f-k-photoshop-jsconf-eu-2012 |
| 14:34 | <jgraham> | IIRC the main missing piece in gimp is adjustment layers |
| 14:34 | <jgraham> | Which have been coming RSN for like a decade |
| 14:34 | <odinho> | jgraham: Yeah, agree. That is sorely missing. |
| 14:35 | <ondras> | yeah, those are nice |
| 14:36 | <zewt> | iirc it was missing record/playback last time i used it |
| 15:45 | <dglazkov> | good morning, Whatwg! |
| 16:06 | <JonathanNeal> | Anyone here at Event Apart San Diego? |
| 17:42 | <Hixie_> | what is @@unscopables? |
| 17:44 | <Ms2ger> | I think the thing to make names not show up in the weird event listener scope pollution thingy |
| 17:45 | <miketaylr> | or in with statements? |
| 17:45 | <Ms2ger> | Go and wash your mouth with soap |
| 17:46 | <Hixie_> | oh hey a Ms2ger |
| 17:46 | <Hixie_> | did you get my mail the other day? |
| 17:46 | <Hixie_> | i dunno if i have your right address |
| 17:46 | <Ms2ger> | Yeah, I got it |
| 17:47 | <Ms2ger> | The reason I didn't reply yet is the answer to your question: lack of time :/ |
| 17:51 | MikeSmit1 | ponders what novel mileage the wags might be able to get out of http://lists.w3.org/Archives/Public/public-html-commits/2014May/0009.html |
| 17:52 | <MikeSmith> | "maybe we need a spec that defines 'super well-formed'" |
| 17:53 | <MikeSmith> | extra well-formed, exceptionally well-formed, etc |
| 17:53 | <MikeSmith> | pathologically well-formed? |
| 17:54 | <Domenic> | what ... what is that... |
| 17:56 | <MikeSmith> | I think the polyglot document has now reached the point where it's become sentient but also insane and is now trolling itself |
| 18:03 | <Hixie_> | anyone around who can review the current createImageBitmap() spec's use of Promises and tell me how to update it to the most recent Promises spec? |
| 18:03 | <Hixie_> | it was written back when Promises were in DOM |
| 18:04 | <Domenic> | Hixie_: sure |
| 18:13 | <Domenic> | Hixie_: https://www.w3.org/Bugs/Public/show_bug.cgi?id=25662 |
| 18:14 | <Hixie_> | you rock, thanks! |
| 18:16 | <Domenic> | Yay ^_^. Going to lunch, but lmk if anything's unclear. |
| 20:08 | <zewt> | new URL("foo://bar.com/").host is ""? grr |
| 20:09 | <zewt> | (in Chrome, at least) |
| 20:09 | <annevk> | Hixie_: see ES6 for @@unscopables |
| 20:10 | <annevk> | Hixie_: it's a way to hide things from the with statement, and should work for event listeners too |
| 20:11 | <zewt> | sounds like a shitty group of superheroes |
| 21:45 | <Domenic> | oh dear, I thought navigator.onLine got renamed to navigator.online... |
| 22:06 | <Hixie_> | annevk: that sounds very confusing |
| 22:06 | <Hixie_> | annevk: having things sometimes work and other times not even though similar things work in different cases |