| 00:45 | <hober> | MikeSmith: I guess I should try tmux at some point. Does it have an equivalent of screen's "nethack on"? |
| 00:46 | <hober> | MikeSmith: because you will pry "You cannot escape from window 0!" out from under my cold, dead, bloated Emacs pinky finger. |
| 00:50 | <MikeSmith> | hober: dunno what nethack on does but I suspect tmux has an equivalent. tmux also handle Unicode like 🍺 that screen fails to display |
| 00:53 | <MikeSmith> | hober: http://www.reddit.com/r/commandline/comments/1y91lz/tmux_vs_screen/ |
| 00:53 | <MikeSmith> | seems it doesn't have it :-( |
| 00:59 | <MikeSmith> | man how do people read reddit on mobile |
| 01:01 | <MikeSmith> | hmm looks way better in chrome than in Firefox Nightly |
| 01:13 | <sgalineau> | man how do people read reddit at all |
| 01:31 | <zewt> | MikeSmith: ircing on couch on ios also showing glyph that I'm certain desktop irc defaulted on |
| 01:33 | <SamB> | what, this one? 🍺 |
| 02:14 | <zewt> | SamB: called it http://i.imgur.com/P0oMgis.png |
| 02:32 | <zewt> | speaking of use cases for high-precision audio, https://twitter.com/ID_AA_Carmack/status/456620100758429697 |
| 02:37 | <zewt> | http://i.imgur.com/yKIqZdS.png theonion with some more font failure |
| 02:38 | <zewt> | you can probably fingerprint the developers of websites based on which characters they assume works, if they aren't competent enough to test on more than one platform |
| 02:38 | <zewt> | (also plurals fails) |
| 02:40 | <zewt> | holy shit, youtube just used an alert() |
| 02:41 | <zewt> | (shock triggered by the fact that chrome regressed alerts and removed tab-modal alerts, putting it back into the dark ages, so I'm there clicking tabs and nothing is happening because the tab has an alert open) |
| 02:41 | <zewt> | (chrome's catastrophic regression being incomprehensibly worse than youtube's lazy alert) |
| 02:43 | <JonathanNeal> | What do any of you think about hashes like ##some+text auto-anchoring to parts of the page matching the text, ala http://sandbox.thewikies.com/autoanchor/ and http://sandbox.thewikies.com/autoanchor/##protocols |
| 02:47 | <MikeSmith> | JonathanNeal: would be nice to have some kind of way to address abitrary text or ranges in a page |
| 02:47 | <MikeSmith> | "##" is some suggested convention for doing that? |
| 02:48 | <JonathanNeal> | KevinMarks2: |
| 02:50 | <JonathanNeal> | MikeSmith: I believe so. Kevin suggested it (and Timothy Cole, in that document). I pinged him if he has thoughts to add. |
| 02:51 | <MikeSmith> | JonathanNeal: I guess you know that shepazu is following all this stuff closely |
| 02:52 | <JonathanNeal> | Honestly, I just stumbled upon it tonight. It just seemed like a great idea. |
| 02:55 | <JonathanNeal> | Like, something I would have expected on the web by now. Hey, shepazu, you’re an html5 homie, right? |
| 03:06 | <SamB> | how about them xpaths |
| 03:08 | <zewt> | xpath was cool until querySelector |
| 03:10 | <zewt> | at which point it settled to a cool level equivalent to the macarana, blue LEDs and java |
| 03:29 | <SamB> | xpath can do more the CSS selectors ... |
| 03:30 | <SamB> | s/the/than/ |
| 04:12 | <JonathanNeal> | btw, https://github.com/chapmanu/fragmentions |
| 05:14 | <MikeSmith> | JonathanNeal: interesting |
| 05:16 | <MikeSmith> | getElementByText |
| 05:17 | <MikeSmith> | JonathanNeal: shepazu can tell you about some more sophisticated ways that existing sites that use annotations are handling this kind of addressing |
| 05:19 | <JonathanNeal> | interesting good,bad,lazy,just-interesting? Looks like we’ll just throw a prototype demo up and get feedback. |
| 05:21 | <JonathanNeal> | Related, <a href="##foo"> appears to be invalid markup “Illegal character in fragment component” adding “ Characters should be represented in NFC and spaces should be escaped as %20” — is there something more to this? |
| 05:21 | <MikeSmith> | ah |
| 05:21 | <MikeSmith> | yeah the URL spec doesn't allow fragments to contain a hash sign |
| 05:24 | <JonathanNeal> | That’s good and bad, because one concern regards sites already using ##. Googling various terms related to double hashes did not yield any results on the subject. |
| 06:04 | <zcorpan_> | MikeSmith: it appears you should also make some more links in platform.whatwg.org point to whatwg so that it doesn't look like we're not involved in web standards :-P |
| 06:22 | <TabAtkins> | SamB: XPath can do more than Selectors, but Selectors can do more than XPath. The two are roughly equivalent, but not strict subsets. |
| 06:25 | <MikeSmith> | zcorpan_: yeah I had already been meaning to change most of those |
| 06:26 | <MikeSmith> | lemme change some right now |
| 06:27 | <zcorpan_> | MikeSmith: (q.v. blink-dev; i wasn't serious but if there are links that should actually point to whatwg, go ahead) |
| 06:31 | <MikeSmith> | zcorpan_: ah yeah I haven't read up on that thread. that guy seemed a bit confused |
| 06:32 | <MikeSmith> | still I want to make the right stuff more prominent |
| 07:33 | <MikeSmith> | zcorpan: http://platform.html5.org/ |
| 07:36 | <zcorpan> | MikeSmith: why are the whatwg items bigger font? |
| 07:37 | <MikeSmith> | zcorpan: to make them bigger |
| 07:37 | <MikeSmith> | and to match more closely with the size of the icon |
| 07:37 | <MikeSmith> | should I make them the same size? |
| 07:38 | <zcorpan> | yeah i think it looks a bit silly now :-) |
| 07:38 | <MikeSmith> | k |
| 07:39 | <zcorpan> | might be nice with icons for everything? |
| 07:39 | <MikeSmith> | zcorpan: should I make the whatg icon smaller to match the text? |
| 07:39 | <MikeSmith> | icons for everything? |
| 07:39 | <MikeSmith> | zcorpan: like, add a w3c icon too? |
| 07:39 | <zcorpan> | yeah |
| 07:39 | <MikeSmith> | OK |
| 07:39 | <MikeSmith> | will add it later |
| 07:40 | <zcorpan> | as for smaller icons, maybe that makes it ugly on lowres screens? |
| 07:40 | <MikeSmith> | yeah |
| 07:40 | <MikeSmith> | so I'll leave them as-is |
| 07:40 | <zcorpan> | unless you use the svg version but maybe there's no svg version for w3c/kronos/etc |
| 07:46 | <MikeSmith> | zcorpan: there is for w3c at least |
| 07:46 | <MikeSmith> | but I'll just do the bitmaps for now |
| 07:47 | <MikeSmith> | I wonder what's the latest URL for Promises? |
| 07:48 | <MikeSmith> | https://people.mozilla.org/~jorendorff/es6-draft.html#sec-promise-objects I guess |
| 07:48 | <zcorpan> | yeah |
| 08:45 | <MikeSmith> | zcorpan: OK take another look now |
| 08:46 | <zcorpan> | MikeSmith: A+ |
| 08:54 | <MikeSmith> | yay |
| 08:54 | <MikeSmith> | so I'll stop for now |
| 09:05 | <annevk> | MikeSmith: maybe add a link to http://wiki.whatwg.org/wiki/DOM_XPath |
| 09:09 | <MikeSmith> | annevk: ok, added |
| 09:12 | <annevk> | MikeSmith: would be even cooler btw if the specs with their own logo... |
| 09:14 | <MikeSmith> | annevk: yeah I'd need to switch them over the svg ones I guess |
| 09:16 | <darobin> | mmmm, w3c-test.org down? |
| 09:18 | <annevk> | darobin: glad the DOM thing was a misunderstanding |
| 09:18 | <darobin> | annevk: what DOM thing? |
| 09:19 | <annevk> | darobin: ownerElement |
| 09:19 | <darobin> | oh yeah, I fucked that one up |
| 09:20 | <darobin> | I was sure it was in after the discussion in www-dom so when it didn't show up in the snapshot after the copy I assumed I'd screwed up syncing |
| 09:21 | <darobin> | I should've given myself more credit, cut and pasting is something I'm actually becoming good at |
| 09:24 | <Ms2ger> | darobin, w3c-test.org does appear down from here |
| 09:24 | <darobin> | Ms2ger: yeah, I'm looking into it |
| 09:24 | <darobin> | the server is running |
| 09:24 | <Ms2ger> | Ah, doing something useful, excellent ;) |
| 09:25 | <darobin> | mmmm, but using 8321m of memory and 50% CPU, that doesn't sound right |
| 09:26 | <darobin> | errr, MikeSmith are you on the server too? |
| 09:27 | <MikeSmith> | darobin: yeah |
| 09:27 | <darobin> | heh |
| 09:27 | <MikeSmith> | I just restarted wptserve |
| 09:27 | <Ms2ger> | MikeSmith, the phantom of the server |
| 09:27 | <darobin> | we were doing the same thing :) |
| 09:27 | <MikeSmith> | darobin: hah |
| 09:27 | <MikeSmith> | blood brothers |
| 09:27 | <darobin> | got me confused when I killall python and it matches nothing |
| 09:27 | <MikeSmith> | I beat you to the kill this time |
| 09:28 | <darobin> | lol |
| 09:28 | <darobin> | I wonder if we shouldn't have a cronjob restart the server daily |
| 09:28 | <darobin> | clearly it can get into a bad state |
| 09:29 | <darobin> | or maybe the systeam has a nice watchdog we could use |
| 09:29 | <Ms2ger> | Would be interesting to figure out what gets it into that state |
| 09:30 | <darobin> | Ms2ger: yeah, I managed to get it in a bad state on my local machine but I couldn't figure out what it was unhappy about |
| 09:30 | <darobin> | it was something about trying to write to a closed socket |
| 09:30 | <darobin> | Ms2ger: I can repro by reloading lots and lots of times brutally |
| 09:30 | <MikeSmith> | maybe it has something to do with it also writing binary data to its log |
| 09:31 | <MikeSmith> | or emitting it to the console -- stderr I guess |
| 09:31 | <darobin> | but the problem took me deep inside the python libs and I'm not proficient enough to debug that well |
| 09:31 | <MikeSmith> | we should get the webdriver folks to look at it |
| 09:32 | <darobin> | why them? |
| 09:32 | <MikeSmith> | David Burns if he has time |
| 09:32 | <darobin> | because pythonistas? |
| 09:32 | <MikeSmith> | darobin: because they have the python skills |
| 09:32 | <MikeSmith> | yeah |
| 09:32 | <darobin> | good point |
| 09:32 | <darobin> | though I thought we had Ms2ger and jgraham_ as python deities :) |
| 09:32 | <Ms2ger> | Nope |
| 09:32 | <Ms2ger> | Just jgraham |
| 09:33 | jgraham | isn't sure about deitiy |
| 09:33 | <jgraham> | But I could certainly have a go at debugging the problem if I could reproduce it |
| 09:35 | <jgraham> | Anyway, I need to catch a train |
| 09:37 | <MikeSmith> | we need some rr for wptserve |
| 09:39 | <darobin> | we could use something like https://github.com/tsenart/vegeta |
| 10:21 | <Ms2ger> | "if the W3C group truly only C&P'ed WHATWG specs, no one would be in the WGs." |
| 10:21 | <Ms2ger> | Then why are they in the WGs now? |
| 10:26 | <jgraham> | I don't think anyone *is* in the HTMLWG, at least |
| 10:27 | <jgraham> | Well I guess Microsoft are |
| 10:27 | <Ms2ger> | I think Mozilla still is |
| 10:29 | <Ms2ger> | Do you / zcorpan know if the XHR test suite is comprehensive? |
| 10:29 | <jgraham> | Well I mean people are in the group |
| 10:30 | <jgraham> | But I don't think anyone believes that the HTMLWG is a useful mechanism for getting changes into HTML |
| 10:34 | tobie_ | wonders whether the overhead of discussing the WHATWG vs. W3C situation is <=, >= or == than dealing with the W3C overhead. |
| 10:34 | tobie_ | ducks. |
| 10:34 | <jgraham> | quack |
| 10:36 | <jgraham> | (and I think the fact that people have just stopped participating in the W3C process should be a significant data point in answering that question) |
| 10:37 | <tobie_> | So my intent wasn't to start a discussion on that topic and add some meta-overhead. |
| 10:37 | <tobie_> | :) |
| 10:42 | <zcorpan> | Ms2ger: i'm not that familiar with the xhr testsuite but i have a feeling it has various gaps in coverage |
| 10:44 | <jgraham> | zcorpan: Do you store that response locally as "%(name)s: I'm not familiar with the %(suite_name) testsuite but I have a feeling it has various gaps in coverage"? |
| 10:46 | <zcorpan> | jgraham: i'm not that familiar with the response testsuite but i have a feeling it has various gaps in coverage |
| 10:49 | <jgraham> | darobin-- for mixing big whitespace changes with semantic changes |
| 10:50 | <darobin> | ah shit, sorry jgraham |
| 10:50 | <darobin> | I blame aryeh for using tabs in the first place :) |
| 10:50 | <jgraham> | :) |
| 11:14 | <darobin> | Ms2ger: do you know why detach() was turned into a noop? |
| 11:14 | <darobin> | I ask because it's clearly not a noop in at least Blink and WebKit (apparently not in IE either), and that's causing an awful lot of test failures |
| 11:17 | <darobin> | annevk: ^ |
| 11:17 | <Ms2ger> | darobin, because it wasn't useful |
| 11:18 | <darobin> | Ms2ger: mmmmm |
| 11:18 | <Ms2ger> | I'm surprised nobody else killed it yet |
| 11:19 | <darobin> | if they did they'd pass most if not all of the range tests |
| 11:19 | <Ms2ger> | zcorpan, I should bug... Hallvors, I guess? |
| 11:30 | <Ms2ger> | "name of illusion causing proponents of a concept to see most things as that concept? e.g. webintents,annotations,cards" |
| 11:30 | <Ms2ger> | tantek, going to coin "monadism" |
| 11:35 | <jgraham> | That seems a bit like an error of the form "They laughed at Columbus, they laughed at Fulton, they laughed at the Wright brothers. But they also laughed at Bozo the Clown." |
| 11:36 | <annevk> | darobin: what do you mean by clearly not a noop? |
| 11:36 | <annevk> | darobin: if you mean they set a flag and threw for some methods, that doesn't necessarily make it useful |
| 11:36 | <darobin> | annevk: after you call it, trying to access various attributes throws |
| 11:36 | <darobin> | (as per DOM3) |
| 11:37 | <darobin> | yeah, I didn't say it was useful, just that it seemed to be implemented as per DOM3 |
| 11:37 | <annevk> | we removed throwing because it could just be GC'd if you didn't have references |
| 11:37 | <annevk> | detach() wasn't needed for that |
| 11:37 | <jgraham> | darobin: Here's your chance to write a patch, fix Blink, get a whole load of tests to pass, and get DOM closer to Rec! |
| 11:37 | <darobin> | I wonder if other implementations might need it? |
| 11:38 | <darobin> | jgraham: yeah! or I could just write something to the effect that 97% of failures come from the fact that browsers support detach() when they shouldn't and call it a day :) |
| 11:39 | <annevk> | darobin: need it for what? |
| 11:39 | <darobin> | that said, I guess a patch removing stuff from Blink would actually be quite easy to get through the process :) |
| 11:39 | <darobin> | annevk: I'm presuming this was initially introduced due to some implementation that needed to be told what to do; I was wondering if it were possible that someone still implemented ranges in a way that made detach() useful |
| 11:40 | <darobin> | (which isn't a reason to keep it, I'm just wondering why it hasn't been killed yet) |
| 11:40 | <jgraham> | darobin: Well that might work if your only care about the spec going to Rec :p |
| 11:40 | <annevk> | darobin: Java prolly needs detach() |
| 11:40 | <darobin> | annevk: that's what I was thinking |
| 11:40 | <annevk> | we only care about JavaScript |
| 11:40 | <darobin> | jgraham: I care about many other things, but patching Blink isn't really one of them |
| 11:40 | <darobin> | apparently it's not even implemented in JS! |
| 11:42 | <annevk> | OMG IDL |
| 11:42 | <annevk> | better call Domenic_ |
| 11:44 | darobin | puts on the Ghostbusters theme music |
| 11:45 | <jgraham> | annevk: Pretty sure no one uses OMG IDL these days :p |
| 11:50 | <zcorpan> | OMG OMG IDL |
| 12:01 | <jgraham> | It always confuses me when posting a branch to GitHub fails to create a review :( |
| 12:02 | <jgraham> | Maybe I should make a "git review" command |
| 12:17 | <zcorpan> | jgraham: maybe test_obj should have a .timeout() function? |
| 12:18 | <jgraham> | zcorpan: Yeah, I was thinking .force_timeout() or something |
| 12:20 | <zcorpan> | sounds good |
| 12:30 | <darobin> | jgraham: actually, I recall seeing some OMG IDL not very long ago, I forget where |
| 12:30 | <darobin> | I think it was a payment system defined with it |
| 12:30 | <darobin> | apparently it's still a thing |
| 12:31 | <darobin> | indeed, not only is the OMG still a thing, but they've also discovered animated GIFs! http://www.omg.org/ |
| 12:33 | <jgraham> | I was expecting the executives looking at the laptop to start poking their own eyes out |
| 12:34 | <jgraham> | Disappointed to find it was only balloons |
| 12:37 | <zcorpan> | i like the scrolling thing |
| 12:41 | <jgraham> | I'm a big fan of the fact that the top quote is "Personally I have made many friends at the OMG" |
| 12:41 | <jgraham> | That's a strategy the W3C should adopt |
| 12:42 | <jgraham> | "W3C: Because joining us is a bit like having a social life" |
| 12:42 | <jgraham> | Maybe s/joining us/being in a working group/ |
| 12:43 | jgraham | applies for the position of brand consultant |
| 12:57 | <tobie_> | Can't believe this is a thing. |
| 13:00 | <jgraham> | OMG? |
| 13:01 | <jgraham> | How did you think they got their name? |
| 13:05 | <zcorpan> | jgraham: need to leave now, assume that i've LGTMed some future change involving force_timeout() if you don't want to wait until tuesday |
| 13:29 | <MikeSmith> | fyi UK peeps: https://twitter.com/w3c/status/456772456372400128 |
| 13:40 | <jgraham> | MikeSmith: Uh |
| 13:41 | <jgraham> | That appears to be "GCHQ have just joined W3C" |
| 13:41 | <jgraham> | I see no possible way this could go wrong |
| 13:51 | <MikeSmith> | keep your friends close |
| 13:51 | <MikeSmith> | as the saying goes |
| 13:57 | <Ms2ger> | jgraham, to join the Intelligence Agencies Business Group? |
| 13:59 | <beverloo> | that'd be a great forum to discuss how to not flag security issues |
| 14:16 | <jgraham> | Ms2ger: Want do review the merge in https://critic.hoppipolla.co.uk/8241e648?review=1151 ? |
| 14:19 | <Ms2ger> | r+ |
| 14:20 | <jgraham> | Ms2ger: Thanks |
| 14:20 | Ms2ger | was going to do some work this afternoon... *sigh* |
| 15:01 | <Ms2ger> | �The language does not prevent you from deeply nesting classes, but good taste should. [...] Nesting more than two levels invites a readability disaster and should probably never be attempted.� |
| 15:02 | <jgraham> | ? |
| 15:05 | <SamB__> | Ms2ger: which one? |
| 15:06 | <Ms2ger> | Java |
| 15:06 | <Ms2ger> | Oh, I also can't read |
| 15:07 | <SamB> | is that talking about actual closures or just some boring namespace/protection thing? |
| 15:15 | <tantek> | Ms2ger interesting, I'll have to read up on monads and monadology. |
| 15:15 | <tantek> | there are more examples too, e.g. "triples" ;) |
| 15:15 | <tantek> | but it's particularly concerning that this illusion pattern appears to repeat itself just in web technology circles so often |
| 15:17 | <tantek> | also, there seems to be some consensus that it's a form of confirmation bias, in particular a type of Maslow's hammer. See replies here: https://twitter.com/t/status/456552583499243521 |
| 15:29 | <annevk> | Ms2ger: where is that from? |
| 15:29 | <Domenic_> | MikeSmith: platform.html5.org is really excellent now. I like how it links to EDs for the W3C things too. |
| 15:30 | <annevk> | MikeSmith: http://simonsapin.github.io/data-urls/ is the latest we have on data URLs |
| 15:31 | <SimonSapin> | It’s more a collection of open issues than anything else |
| 15:45 | <tantek> | also, this article from that thread has a lot of applicability in many web platform discussions: http://www.joelonsoftware.com/articles/fog0000000018.html - so many architecture astronauts |
| 16:15 | <MikeSmith> | Domenic_: Thanks |
| 16:16 | <MikeSmith> | annevk: didn't know about that one |
| 16:16 | MikeSmith | looks |
| 16:52 | <Domenic_> | hehehe https://twitter.com/antimattur/status/456835404596252673 |
| 16:59 | <JonathanNeal> | MikeSmith, ready ready for some in the wild testing of fragmentions? |
| 17:05 | <MikeSmith> | JonathanNeal: you win |
| 17:06 | MikeSmith | and JonathanNeal are playing "Best spring-break pickup lines" |
| 17:07 | <JonathanNeal> | Oh we are? I plan on observing Good Friday. If you’re free, let’s make it a GREAT Friday! |
| 17:18 | JonathanNeal | listens to the crickets, but still thinks it was clever. |
| 17:19 | <jsbell> | that feeling when you haven't settled into a new workflow and you feel like a t-rex in a china shop |
| 17:25 | <JonathanNeal> | jsbell: that does sound kind of awesome though. Like, those plates don’t stand a chance. |
| 17:25 | <jsbell> | Rawr! Whoops, sorry. Let me pick that up. Oh, wait, useless arms. |
| 17:29 | <JonathanNeal> | Tea, Rex? |
| 17:31 | <jgraham> | jsbell: Do you want help with something, or are you just happily complaining? |
| 17:31 | <jsbell> | jgraham: Just happily complaining. :) |
| 17:32 | <Domenic_> | annevk: FYI I think I'm going to do a "how to make subclassable APIs" guide. I feel like we've talked about related topics a lot. https://github.com/w3ctag/subclassable-apis-guide/issues |
| 17:35 | <annevk> | Domenic_: one thing that'd be interested is considering the impact on the current APIs |
| 17:35 | <annevk> | interesting* |
| 19:06 | <IZh> | Hi. |
| 19:27 | <KevinMarks> | The difference in what can be in a URL fragment and what can be in an id is interesting |
| 19:29 | <annevk> | KevinMarks: you mean given that a fragment is a sequence of bytes and an ID is a string? |
| 19:30 | <KevinMarks> | A fragment can encode anything with percent encoding, but an id can't have spaces |
| 19:32 | <KevinMarks> | So the fragmention idea could just be "if the fragment contains whitespace, search body text rather than ids" |
| 19:33 | KevinMarks | has the terminology wrong there. |
| 19:34 | <annevk> | data:text/html,<style id="x x">head,:target{display:block;height:20em}:target{background:purple}</style>#x x |
| 19:34 | <annevk> | does seem to work |
| 19:35 | <annevk> | restrictions on IDs are mostly so they can appear space-separated elsewhere I suppose |
| 20:47 | <JonathanNeal> | annevk, is this related to fragmentions? |
| 21:58 | <benschwarz> | Hixie: yt? |
| 22:17 | <jgraham> | jsbell: I'm not sure why you created a new review, but I think we're nearly there |
| 22:18 | <jsbell> | jgraham: heh, was just gonna ping. since I wasn't in a branch a merge got in there and critic stopped tracking (and wouldn't start again when I poked it) |
| 22:20 | <jgraham> | jsbell: Oh, the merge shouldn't have upset it (although it is unorthodox), but a rebase will (although it is more normal; there are just a few manual steps) |
| 22:21 | <benschwarz> | annevk: someone left an issue on developers.whatwg.org asking if they could translate it to chinese… |
| 22:22 | <jsbell> | jgraham: I'm sure I poked the wrong button somehow. Then flattened it down to a diff and reapplied. And obviously mucked that up. *sigh* Anyway... updated. |
| 22:23 | <annevk> | benschwarz: sounds good, no? |
| 22:23 | <jgraham> | jsbell: And we're done! Thanks for the patch |
| 22:23 | <annevk> | benschwarz: although might be a bit tricky keeping it up to date |
| 22:24 | <jsbell> | Also, no longer filtering mail from critic into /dev/null :P |
| 22:25 | <jsbell> | jgraham: If I was gonna add some encoding API tests, any preference for where in web-platform-tests ? |
| 22:25 | <jgraham> | jsbell: Well that's a seperate spec so a new top level directory |
| 22:25 | <jgraham> | /encoding/ or something I guess |
| 22:26 | <jsbell> | jgraham: In blink we have encoding/ for general (e.g. fetch/form/document) encoding tests and encoding/api for the API specifically. How's that sound? |
| 22:29 | <jgraham> | jsbell: So the normal convention is to have the subdirectories (if any) match the spec sections so that we can easilly tell which sections have some tests |
| 22:29 | <jgraham> | (specifically to have the name match the id of the heading element) |
| 22:30 | <jgraham> | so enconding/api for API tests looks right |
| 22:30 | <jsbell> | jgraham: groovy. |
| 22:37 | <annevk> | Domenic_++ |
| 22:48 | <benschwarz> | annevk: do you know of any other efforts for translation that have occurred? |
| 22:48 | <benschwarz> | it'd be an epic job |
| 22:48 | <annevk> | benschwarz: could ask @myakura |
| 22:49 | <annevk> | benschwarz: he has done some work on that years ago iirc |
| 22:56 | <zewt> | annevk: heh, the #x x thing puts "#x x" in <body> in chrome but not in firefox (but treats it as a hash in both) |
| 22:57 | <zewt> | different parsing of data:? |
| 22:57 | <annevk> | Chrome has a bug |
