| 00:07 | <AryehGregor> | Blah. Should I test parsing and serialization at all? CSSOM is such a mess . . . |
| 00:11 | <gsnedders> | It's so far from standardized there's no point. |
| 02:24 | <jarek> | Hi |
| 02:24 | <jarek> | what's the status of this proposal? http://lists.w3.org/Archives/Public/www-dom/2011OctDec/0020.html |
| 02:26 | <jarek> | I think it would be most elegant if could use CSS selectors for element creation |
| 02:27 | <jarek> | e.g. var buttonElement = createElement("a#edit-button.button[href='http://test.com']";) |
| 02:28 | <jarek> | s/if could use/if we could |
| 02:32 | <jarek> | creates <a id="edit-button" class="button" href="http://test.com"></a>; |
| 03:49 | <Hixie> | jarek: createElement(":not(:first-child)") ? :-) |
| 03:49 | <jarek> | Hixie: of course I meant a subset of CSS... |
| 03:51 | <jarek> | CSS selectors already made its way into DOM with querySelector() and querySelectorAll() |
| 03:51 | <Hixie> | jarek: :-P |
| 03:51 | <jarek> | from my reading of the mailing list it looks like there is no consensus on this |
| 03:59 | <Hixie> | you could say that |
| 04:00 | <Hixie> | i kinda like the idea of createElement("a#edit-button.button[href='http://test.com']";), but i think in practice it's not so great because what you'd usually be doing is createElement("a[href='" + url + '"]') and then you end up with all the usual XSS problems of unescaped content being injected |
| 04:00 | <Hixie> | not to mention it looks way uglier than the current proposals especially after you add three or four such attributes |
| 04:07 | <jarek> | Hixie: yup, it would become messy when you use it for complex elements |
| 04:08 | <jarek> | Hixie: but from my experience, most of the time I create elements that have one class or id |
| 04:09 | <jarek> | the JSON approach discussed on the mailing list assumes the opposite |
| 04:10 | <jarek> | it would be best if we could have both JSON and CSS syntax |
| 04:10 | <jarek> | but considering the amount of bikeshedding this is probably never going to happen :P |
| 04:12 | <Hixie> | jarek: i'd be ok with simple syntax to add classes and ids (though when you're doing it via the DOM, you shouldn't really need an ID, just keep a reference to the element) |
| 04:13 | <Hixie> | jarek: but i don't think it should be one based on a special syntax that gets parsed, because that's asking for injection bugs |
| 04:14 | <jarek> | Hixie: but all browsers already ship with solid CSS parsers |
| 04:14 | <Hixie> | the injection bugs would be in the js, not in the browsers |
| 04:14 | <jarek> | I'm not sure how this could be abused |
| 04:15 | <Hixie> | all it takes is for a page to accept input from the user to determine the class name, and for the page not to remember to escape "." or "#" characters |
| 04:33 | <kennyluck> | Hixie, just curious, what kind of page would accept a class name from a user? |
| 04:34 | <Hixie> | kennyluck: no idea |
| 04:34 | <Hixie> | kennyluck: it's certainly more of a problem with arbitrary attributes |
| 05:44 | <Hixie> | wow, 'transform' is still only supported with prefixes? |
| 05:47 | <othermaciej> | yep |
| 05:48 | <othermaciej> | I can't even convince some of my fellow Apple employees that there is no net benefit to prefixing |
| 05:48 | <othermaciej> | a lot of people seem to have faith in the literal truth of <http://www.w3.org/TR/css-2010/#experimental>; |
| 05:49 | <Hixie> | putting aside the issue of whether or not they should be prefixed in the first place, shouldn't they definitely be unprefixed by now? |
| 05:49 | <othermaciej> | in my opinion, certainly yes |
| 05:49 | <othermaciej> | but officially, they are not in CR |
| 05:49 | <Hixie> | what's taking so long? |
| 05:50 | <othermaciej> | I don't follow CSS WG enough to have a comment more informed than what anyone could guess |
| 05:50 | <Hixie> | k |
| 05:51 | <othermaciej> | I did read some of the recent www-style discussion on whether to unprefix transform or not |
| 05:51 | <Hixie> | yeah i saw it's been a topic of discussion |
| 05:52 | <Hixie> | prefixing seems fine to me for actual experiments, but experiments need to be short-lived by definition |
| 05:52 | <Hixie> | if we're just gonna ship them like this, there's no benefit and a whole bunch of harm, seems to me |
| 05:52 | <Hixie> | i guess it basically depends on the spec editor |
| 06:00 | <othermaciej> | I think there is often a challenge in standards work in general, or perhaps life in general, where a rule is made for a very sensible original reason, but people start believing in the rule instead of the reason |
| 06:02 | <[tm]> | othermaciej: indeed |
| 06:02 | <Hixie> | othermaciej: no kidding |
| 06:05 | <othermaciej> | as a person who's had to invent rules, I understand pretty well that they can end up not serving their original purpose or causing unintended consequences |
| 06:11 | <[tm]> | I think I'll file a by report against the decision policy asking that the quote "a foolish consistency is the hobgoblin of little minds" be printed at the rip on big bold letters |
| 06:11 | <[tm]> | s/rip/top |
| 06:12 | <[tm]> | *bug report |
| 08:40 | <annevk> | so should we have xhr.responseURL? |
| 09:07 | <hsivonen> | annevk: use case? redirect checking? |
| 09:09 | <hsivonen> | Hixie: seems odd to use CSS syntax for HTML elements instead of an HTML tag. (Yay, reinventing IE5 features.) |
| 09:09 | <jgraham> | The CSS syntax thing seems horrendous, tbh |
| 09:10 | <annevk> | resolving relative URLs in non-HTML/XML responses |
| 09:12 | <hsivonen> | Hixie: as I understand it, Apple dropped the editing ball somewhat after submitting the Transforms/Transitions/Animations specs |
| 09:12 | <hsivonen> | Hixie: combine that with the policy that stuff stays prefixed until CR |
| 09:12 | <hsivonen> | Hixie: and combine that with upholding the policy instead of overturning policies that demonstratably bear bad fruit |
| 09:13 | <hsivonen> | Prediction: 3 months from now, Transforms/Transitions/Animations are 3 months from CR again |
| 09:14 | <hsivonen> | CSS3--always 3 months from CR |
| 09:43 | <Ms2ger> | Hmm, I hadn't seen http://dbaron.org/css/intrinsic/ before |
| 10:04 | <annevk> | pretty old :) |
| 10:05 | <annevk> | oh it's been updated again |
| 10:14 | <Ms2ger> | MikeSmith, yt? |
| 10:15 | <MikeSmith> | yup |
| 10:15 | <Ms2ger> | Is http://w3c-test.org/framework/ supposed to get the results from testharness.js tests automatically? |
| 10:15 | <MikeSmith> | yeah |
| 10:15 | <MikeSmith> | it was broke today though |
| 10:15 | <MikeSmith> | I broke it |
| 10:16 | <MikeSmith> | I think I fixed what I broke |
| 10:16 | <MikeSmith> | I mean just now fixed it, like in the last 15 minutes |
| 10:16 | <Ms2ger> | Aha |
| 10:16 | <MikeSmith> | so if you were seeing a problem, maybe retry it now |
| 10:16 | <Ms2ger> | Is it updated automatically? |
| 10:17 | <MikeSmith> | PHP and apache just give me so much heartburn |
| 10:17 | <MikeSmith> | it's updated in that I just ssh'ed into the server and edited the code there directly |
| 10:17 | <MikeSmith> | and then committed to the dvcs.w3.org remote |
| 10:18 | <Ms2ger> | Oh, duh, the checkbox |
| 10:18 | <MikeSmith> | ah |
| 10:18 | <MikeSmith> | heh |
| 10:18 | <MikeSmith> | NOT MY FAULT AFTER ALL |
| 10:18 | <MikeSmith> | I feel slightly better |
| 10:19 | <Ms2ger> | Pff, there's enough I can blame you for ;) |
| 10:19 | <MikeSmith> | heh |
| 10:19 | <MikeSmith> | I had tried to walk away from the test-framework code weeks ago so that I don't cause any more damage |
| 10:19 | <MikeSmith> | but they keep pulling me back in |
| 10:20 | <MikeSmith> | having me write php code is an accident waiting to happen |
| 10:20 | <Ms2ger> | Bah, innerHTML |
| 10:20 | <hsivonen> | MikeSmith: s/me/anyone/ |
| 10:20 | <zcorpan> | having anyone write php code is an accident waiting to happen |
| 10:20 | <MikeSmith> | hsivonen: heh |
| 10:20 | <zcorpan> | hsivonen beat me to it |
| 10:20 | <MikeSmith> | zcorpan: make a macro for "having anyone write php code is an accident waiting to happen" |
| 10:20 | <MikeSmith> | for next time |
| 10:21 | <zcorpan> | the s/// trick is fast enough |
| 10:23 | MikeSmith | finally steps away now for real for some R&R |
| 10:23 | jgraham | is going to blame all the bugs he is currently hunting on the legacy PHP system his code intereacts with (and not on his new code, obviously :) |
| 10:29 | <annevk> | did not forget eggs |
| 10:29 | <annevk> | yay |
| 10:30 | <jgraham> | What about yoghurt? |
| 10:30 | <annevk> | healthy supply of yoghurt too, cherry tomatoes, enough until I head to Norway for bacon cheeseburger / taco diet |
| 10:37 | <zcorpan> | what can you make from eggs, yoghurt and cherry tomatoes? |
| 10:37 | <Ms2ger> | A meal |
| 10:38 | <jgraham> | Art |
| 10:39 | <tomasf> | a tomato yoghurt omelette |
| 10:39 | <charlvn> | just throw it into a wok and be done with it |
| 10:40 | <zcorpan> | i think i'd boil the eggs, eat the yoghurt as it is, and put the tomatoes in the oven or so. good breakfast |
| 10:44 | <kennyluck> | Are there mailing lists for spec changes on dvcs.w3.org? Or you need to use RSS/ATOM? |
| 10:44 | smaug____ | kicks some students who just copy-paste code and try to claim it is theirs code. |
| 10:45 | <Ms2ger> | kennyluck, the latter, I think |
| 10:45 | <kennyluck> | oh fine. |
| 10:47 | <hsivonen> | smaug____: in Bugzilla? or have you become a course assistant? |
| 10:48 | <smaug____> | hsivonen: I've been "mentoring" some students |
| 10:48 | <smaug____> | not in bugzilla |
| 10:48 | <hsivonen> | smaug____: I see |
| 10:50 | <jgraham> | zcorpan: Would be better if you poached the eggs, I think |
| 10:50 | <jgraham> | But otherwise I agree |
| 10:52 | <Ms2ger> | Scrambled, of course |
| 10:56 | <Ms2ger> | MikeSmith, seems like the automatic reporting went into an infinite loop: http://w3c-test.org/framework/details/dom4-Ms2ger-submissions/Document-getElementById.html/engine/gecko/ |
| 10:59 | <zcorpan> | hmm i have never made poached eggs. gotta try that |
| 11:02 | annevk | had fried eggs |
| 11:45 | <[tm]> | Ms2ger: つうは |
| 11:46 | <[tm]> | Ms2ger: thanks for the heads-up |
| 11:46 | <[tm]> | will take a look at it when i get back to my laptop |
| 11:59 | <Ms2ger> | Thanks |
| 12:00 | <Ms2ger> | "Communication is"? |
| 12:14 | <annevk> | hmm http://dvcs.w3.org/hg/domcore/raw-file/tip/Overview.html#concept-event-dispatch does not initialize to AT_TARGET for normal objects |
| 12:14 | <annevk> | that seems like a bug |
| 12:15 | <annevk> | maybe we should just flatten that structure and let the ancestor list become the empty list if there's no tree? |
| 12:26 | <annevk> | done |
| 12:50 | <Velmont> | Reading the IDB-spec, -- this seems like it should work: add: 1:"one", 2:"two", 3:"two", add an index on the values. Make a cursor on the values. when at key 2, ("two"), it should be possible to run cursor.continue("two") and get to key 3. |
| 12:52 | <Velmont> | Currently mozilla throws for it, -- webkit doesn't seem to have updated their idb implementation, and ie is impossible to run without loads of stress. :P Ohwell, and they're also implementing an old spec. |
| 13:12 | <annevk> | A : B |
| 13:12 | <annevk> | A is derived from B |
| 13:12 | <annevk> | is that okay? |
| 13:23 | <Velmont> | annevk: That's how the specs looks now, -- what should be not okay with it? |
| 13:27 | <annevk> | Velmont: just wondering about using the word "derived" |
| 13:27 | <annevk> | I should maybe be a bit more verbose :) |
| 13:28 | <Velmont> | Oh. Dunno. :] |
| 14:13 | annevk | is writing an introduction to events |
| 14:31 | <zcorpan> | autofocus on links |
| 14:31 | <zcorpan> | $ grep -aPc "\.focus\(\)" web200904 |
| 14:31 | <zcorpan> | 124167 |
| 14:31 | <zcorpan> | $ grep -aPc "document\.(forms|getElementsByTagName\s*\(\s*[\"']([iI][nN][pP][uU][tT]|[bB][uU][tT][tT][oO][nN]|[tT][eE][xX][tT][aA][rR][eE][aA]|[kK][eE][yY][gG][eE][nN]|[sS][eE][lL][eE][cC][tT])[\"']\s*\))\S*\.focus\(\)" web200904 |
| 14:31 | <zcorpan> | 2713 |
| 14:31 | <zcorpan> | $ grep -aPc "document\.(links|getElementsByTagName\s*\(\s*[\"'][aA][\"']\s*\))\S*\.focus\(\)" web200904 |
| 14:31 | <zcorpan> | 7 |
| 14:35 | <Philip`> | zcorpan: Easier to write ((?i)input|button|...) for case-insensitivity |
| 14:36 | <Philip`> | (assuming you want it scoped to part of the pattern, not global) |
| 14:38 | <zcorpan> | Philip`: ah, thanks |
| 14:38 | <zcorpan> | i tried the -i flag but for some reason that gave 0 results |
| 14:39 | <hsivonen> | Philip`: is ?i equivalent, though? that is, is it ASCII-case-insensitive or Unicode-case-insensitive? |
| 14:40 | <hsivonen> | someone could use İNPUT |
| 14:46 | <zcorpan> | $ grep -aPc "(İ|ı)((?i)nput)" web200904 |
| 14:46 | <zcorpan> | |
| 14:46 | <zcorpan> | searching for just "(İ|ı)" gives plenty of results though |
| 14:47 | <Philip`> | hsivonen: I presume grep is being run with locale settings like LANG=C so it'll treat the input and command-line arguments as 8-bit ASCIIish strings and won't do any encoding/decoding itself, so it can't be Unicode-case-insensitive |
| 14:48 | Philip` | doesn't like how grep seems to go about a hundred times slower when using the default UTF-8 locales on modern Linuxes |
| 15:06 | <AryehGregor> | I just found some transition tests that only IE and Opera pass. |
| 15:06 | <AryehGregor> | (transitions on elements with display: none -- both Gecko and WebKit don't transition) |
| 15:06 | <AryehGregor> | It happens sometimes. |
| 15:06 | <zcorpan> | does transition between 0 and auto work yet? |
| 15:07 | <zcorpan> | for height |
| 15:12 | <annevk> | http://dvcs.w3.org/hg/domcore/raw-file/tip/Overview.html#events |
| 15:13 | <zcorpan> | heh, // great success |
| 15:15 | <zcorpan> | "but do not implement the EventTarget interface and can therefore not be dispatched on themselves." is this something that is necessary to say in an introduction? :-) |
| 15:16 | <zcorpan> | annevk: seems more useful to mention that the listener gets called with the event as its argument |
| 15:18 | <zcorpan> | annevk: "Apart from signaling events " add a comma before events |
| 15:19 | <annevk> | ah yeah, I should rename event listener's listener to event listener's callback |
| 15:20 | <annevk> | all your other comments sound good too |
| 15:21 | <AryehGregor> | zcorpan, there's a spec bug on that, I think. |
| 15:22 | <zcorpan> | AryehGregor: on what? |
| 15:22 | <AryehGregor> | zcorpan, transition to/from auto. |
| 15:22 | <zcorpan> | ah. ok |
| 15:23 | <zcorpan> | it's the first thing i wanted to do with transitions when i tried to use it a few years ago when webkit implemented it and opera had an internal impl |
| 15:23 | <zcorpan> | disappointing that it still doesn't work |
| 15:24 | <annevk> | it's the one thing my friends keep bugging me about |
| 15:24 | <AryehGregor> | I don't know if it works in implementations. |
| 15:24 | <annevk> | I raised it too at one point, didn't really go anywhere |
| 15:27 | <zcorpan> | AryehGregor: bratell wonders how you keep the selection right when manipulating the tree in the editing spec |
| 15:27 | <AryehGregor> | zcorpan, I just reset it sometimes. The range mutation algorithm isn't good enough by itself. |
| 15:28 | <AryehGregor> | Like: http://dvcs.w3.org/hg/editing/raw-file/tip/editing.html#deleting-the-selection "If direction is "forward", call collapseToStart() on the context object's selection." |
| 15:28 | <AryehGregor> | Stuff like that. |
| 15:29 | <AryehGregor> | "Call collapse(start node, start offset) on the context object's selection. Call extend(end node, end offset) on the context object's selection." |
| 15:30 | <zcorpan> | thanks |
| 15:38 | <matjas> | zcorpan: thanks for the link, i saw this behavior mentioned in a few old bug reports as well (ES3 era). i couldn’t find this in the es5 spec though… any pointers? |
| 15:39 | <annevk> | hsivonen: maybe warning in console too? |
| 15:39 | <annevk> | hsivonen: for sync xhr |
| 15:40 | <zcorpan> | matjas: it's not in the es5 spec afaik |
| 15:40 | <hsivonen> | annevk: I'd be OK with that. |
| 15:40 | <zcorpan> | matjas: that's why we have the wiki page |
| 15:40 | <matjas> | zcorpan: ok, i’ll just add a compatibility note then. thanks! |
| 15:41 | matjas | scrolls up on http://wiki.whatwg.org/wiki/Web_ECMAScript#Identifiers and reads the very first paragraph… |
| 15:41 | <matjas> | #facepalm |
| 15:42 | <zcorpan> | i hear ya |
| 15:48 | zcorpan | just learned about error.stack |
| 15:49 | <zcorpan> | since window.onerror got a column argument, should we add column info to exceptions too? |
| 15:50 | <zcorpan> | and add stack to window.onerror? |
| 15:54 | <zcorpan> | or use the 4th argument to onerror for the actual exception object |
| 15:57 | <AryehGregor> | function any_unit_to_num(str) { return Number(String(str).match(/^([\d.]+)/)[1]); } |
| 15:58 | <AryehGregor> | Is that supposed to be intentionally different from parseFloat(), I wonder? |
| 16:02 | <zcorpan> | where's that from? |
| 16:08 | zcorpan | finds "Octal literals" in http://kangax.github.com/es5-compat-table/non-standard/ |
| 16:12 | zcorpan | wonders how to grep for octal literals in scripts |
| 16:12 | <tomasf> | ugh. octal literals in JS? the syntax is so stupid |
| 16:12 | <AryehGregor> | zcorpan, a Gecko mochitest. |
| 16:13 | <zcorpan> | AryehGregor: ? |
| 16:13 | <AryehGregor> | An internal JS-based test. |
| 16:13 | <AryehGregor> | test_transitions_per_property.html. |
| 16:13 | <AryehGregor> | Is where it's from. |
| 16:13 | <zcorpan> | ah |
| 16:13 | AryehGregor | is rewriting to be more readable and less Gecko-specific |
| 16:14 | <zcorpan> | my mind has always went on to something else when you answer me something. :-) |
| 16:14 | <AryehGregor> | That's what backscroll is for. :) |
| 16:55 | <matjas> | zcorpan: http://mathiasbynens.be/notes/javascript-escapes#octal has this regex: \\(?:[1-7][0-7]{0,2}|[0-7]{2,3}) |
| 16:56 | <matjas> | (note that \0 is not an octal escape) |
| 16:56 | <matjas> | ooh octal literals, not escapes; ignore me |
| 17:27 | <Philip`> | zcorpan: Might be worth using some code to parse the pages and extract the script content rather than trying to force everything through grep |
| 17:28 | <Philip`> | Also might be worth resolving references to external .js files and downloading some of those |
| 19:03 | <jhawkins> | Hixie: ping |
| 19:07 | <twisted`> | is it correct that a margin-right on the most right position of a website is covered by a scrollbar? |
| 20:05 | <Hixie> | jhawkins: pong |
| 20:05 | <annevk> | whoa, surprise return from Mark Baker |
| 20:06 | <Hixie> | twisted`: scrollbar should be between padding and border, iirc |
| 20:06 | <Hixie> | twisted`: unless it's one of the popup scrollbars, in which case it should just overlap the padding, inside the border |
| 20:21 | <jhawkins> | Hixie: no need to chat now, thanks |
| 20:22 | <Hixie> | k |
| 20:22 | <Hixie> | man, i go skiing for like one day and i just get deluged in e-mail |
| 20:23 | <TabAtkins_> | I know what you mean. My inbox blew up yesterday. |
| 20:28 | <annevk> | I ended up with +60 emails over the weekend |
| 20:28 | <annevk> | and I worked some during the weekend |
| 20:28 | <annevk> | it's weird |
| 20:31 | <Hixie> | haha, a mime type vs media type discussion |
| 20:31 | <Hixie> | turns out the ietf calls them "ietf types" btw, at least based on the mailing list name ;-) |
| 20:32 | <TabAtkins_> | ietf ptyes |
| 20:33 | <TabAtkins_> | Or I suppose I could infer a "vowels then consontans, both reverse-sorted" rule, so "ietf eytsp". |
| 20:37 | <Hixie> | Ms2ger: on your union types patch, why do you give the optional argument for dom-select-add a default value? |
| 20:37 | Ms2ger | checks |
| 20:39 | <Ms2ger> | Oh, because I found it somewhat unclear if the algorithm at #dom-htmloptionscollection-add handled the omitted argument case |
| 20:40 | <Hixie> | ah k |
| 20:44 | <Hixie> | i don't really see why it's ambiguous, but defaulting to null seems harmless enough so i'll leave it |
| 20:44 | <Hixie> | was the change in capitalisation in the title="" earlier intentional also? |
| 20:46 | <Ms2ger> | I don't think so |
| 20:46 | <Hixie> | k |
| 20:46 | <Ms2ger> | No, wasn't |
| 20:46 | <Hixie> | thanks for the patch |
| 20:47 | <Hixie> | ooh, we went over r7000 and i didn't notice |
| 20:47 | <Hixie> | you also added a newline at the end of my file :-P |
| 20:49 | <Hixie> | 1 Line 16 of IDL #74: at '=': expected ')' |
| 20:49 | <Hixie> | 1 Line 4 of IDL #110: at 'callback': expected identifier |
| 20:49 | <Hixie> | 1 Line 6 of IDL #4: at '=': expected ')' |
| 20:49 | <Hixie> | 1 Line 6 of IDL #56: at 'callback': expected identifier |
| 20:49 | <Hixie> | i wish the webidl checker would tell me which one was IDL #110... |
| 20:49 | <annevk> | oh, 7000 commits |
| 20:49 | <annevk> | I was wondering what to put in the WHATWG Weekly besides the JavaScript Unicode stuff |
| 20:50 | <Hixie> | "hixie was skiing and then had a lazy weekend, so nothing happened on html this week" :-P |
| 20:51 | <Hixie> | aah, i see, the complaint is that "callback" is a reserved word now |
| 20:51 | <Hixie> | and i use it as my argument name |
| 20:51 | <Hixie> | fair enough |
| 20:51 | <annevk> | oh shit |
| 20:51 | <annevk> | that's not allowed? |
| 20:51 | <annevk> | Hixie: hehe |
| 20:52 | <Hixie> | you have to use _callback or something |
| 20:52 | <miketaylr> | use hollaback |
| 20:52 | <Hixie> | to escape the identifier |
| 20:52 | <annevk> | argument names should not be identifiers |
| 20:52 | <annevk> | I think heycam|away should fix that |
| 20:52 | <annevk> | argument names are nothing |
| 20:52 | <annevk> | should be easy enough to disambiguate, no? |
| 20:53 | <Hixie> | well without lookahead you need some way to disambiguate the argument name from more information about the argument |
| 20:53 | <annevk> | name always follows type afaik |
| 20:53 | <annevk> | and then there's a comma or ) |
| 20:54 | <Hixie> | e.g. foo(optional long long bar) - is the type "long long" or is the type "long" and the argument called "long"? |
| 20:54 | <Hixie> | "long long" might be the only time this is a problem |
| 20:54 | <annevk> | and long long is predefined |
| 20:55 | <Hixie> | only because "long" is a reserved word |
| 20:55 | <annevk> | right |
| 20:55 | <annevk> | I'm gonna raise this, might be worth fixing |
| 21:01 | <annevk> | oh you just escape them in the IDL? hmm still ugly |
| 21:05 | <AryehGregor> | Who decided we should have both String.substr() and String.substring()? Really? |
| 21:05 | <TabAtkins_> | Java. And Eich. |
| 21:06 | <othermaciej> | are they distinct? |
| 21:06 | <TabAtkins_> | Yes. |
| 21:06 | <TabAtkins_> | One takes an end index, the other a length. |
| 21:06 | <TabAtkins_> | Guess which! |
| 21:07 | <othermaciej> | that's awesome |
| 21:08 | <AryehGregor> | That was actually deliberately copied from Java? |
| 21:08 | <AryehGregor> | I assumed it was some kind of fluke where two different browsers came up with them independently and we compromised by picking both. |
| 21:08 | <TabAtkins_> | Ah, I guess not. |
| 21:08 | <TabAtkins_> | Blame Eich anyway. He's used to it. |
| 21:09 | <TabAtkins_> | Java has a proper start/end function named substring. |
| 21:14 | <Ms2ger> | We've got one java-style, one python-style and one perl-style, no? |
| 21:14 | <gsnedders> | othermaciej: One is deprecated and spec'd in Appendix B (substr), one isn't (and spec'd in the main body of the spec). |
| 21:15 | <annevk> | so |
| 21:15 | <annevk> | WTWTF, OMOMG |
| 21:15 | <gsnedders> | TabAtkins_: The only things from Java are java.util.Math and java.util.Date |
| 21:15 | <annevk> | BBBBQ? |
| 21:15 | <annevk> | dino ^^ |
| 21:15 | <Ms2ger> | gsnedders, and the name? :) |
| 21:18 | <annevk> | last time I wrote JavaScript I had to replace all my ands with && and ors with || |
| 21:18 | <annevk> | crazy |
| 21:19 | <annevk> | and all those parenthesis |
| 21:20 | <gsnedders> | Ms2ger: Well, that's not really taken from, it more just is it. |
| 21:21 | <gsnedders> | Ms2ger: And it's called ECMAScript!111!!!11111!!!!11!!!eleventy!!! |
| 21:21 | <Ms2ger> | :) |
| 21:22 | <annevk> | ECMAScript is just a posh name for JavaScript |
| 21:23 | <gsnedders> | A rose by any other name would smell as sweet? |
| 21:24 | <gsnedders> | But I guess plenty of people wouldn't claim JS/ES/whatever smells sweet. |
| 22:04 | <odinho> | 23:01 -!- Topic set by annevk42 [] [Mon Oct 19 23:03:06 2009] |
| 22:04 | <odinho> | That is some old topic, sure! That's why the stats say: A topic was never set on this channel. |
| 22:06 | <smaug____> | gavin's stats aren't updating |
| 22:07 | <annevk> | annevk42 is awesome |
| 22:08 | <gavin> | smaug____: hmm? http://gavinsharp.com/irc/whatwg.html looks up to date |
| 22:09 | <TabAtkins_> | I'm pretty close to catching up to hsivonen! |
| 22:09 | <TabAtkins_> | Just gotta spam it up for a while, gain another 300 lines. |
| 22:09 | <gavin> | oh... 2012 might have broken the config |
| 22:10 | <annevk> | there was a reset of the logs right? |
| 22:11 | <annevk> | anne-mac is no longer there |
| 22:11 | <gavin> | that was a while ago |
| 22:11 | <gavin> | but yeah, I changed the machine hosting them, and so stats lost ~3years of historical logs |
| 22:11 | <gavin> | http://gavinsharp.com/irc/old/whatwg.html is the archived copy |
| 22:13 | <annevk> | oh cool |
| 22:14 | <gavin> | I still have all of the old logs, I just can't easily merge them all together |
| 22:14 | <annevk> | one person kicked in well over 1617 days now |
| 22:15 | <annevk> | in 1617 + 450 days |
| 22:15 | <annevk> | pretty awesome |
| 22:16 | <gavin> | ok, think I found a simple fix to the 2012 problem |
| 22:16 | <smaug____> | gavin: "today "bug 690225 is not a crasher"" is very old |
| 22:16 | <gavin> | I'll trigger a re-run with that fixed |
| 22:17 | <annevk> | smaug____: "today" and the quote are different fields |
| 22:17 | <annevk> | smaug____: "today" is just last seen |
| 22:17 | <annevk> | smaug____: the quote is completely random afaik |
| 22:17 | <gavin> | annevk: no, the quote is generally from the last day |
| 22:17 | <annevk> | oh |
| 22:17 | <gavin> | well, "last seen" |
| 22:17 | <annevk> | should not say "Random quote" then maybe |
| 22:17 | <gavin> | yeah I guess |
| 22:18 | <annevk> | I saw they updated the software btw |
| 22:18 | <annevk> | but no interesting new features |
| 22:18 | <annevk> | http://pisg.sourceforge.net/changelog |
| 22:18 | <gavin> | heh |
| 22:19 | <smaug____> | annevk: ah, right |
| 22:19 | <smaug____> | hmm |
| 22:32 | <gavin> | ok, http://gavinsharp.com/irc/whatwg.html is updated |
| 22:32 | <gavin> | smaug____: ^ |
| 22:32 | <smaug____> | did I go up in the top 25 list |
| 22:32 | <smaug____> | that is surprising |
| 22:32 | <TabAtkins_> | Dammit, hsivonen. He widened the gap. |
| 22:32 | <smaug____> | other people have been even more lazy with whatwg than I |
| 22:32 | <smaug____> | (or even busier with other stuff than I) |
| 22:33 | <gavin> | AryehGregor surpassed MikeSmith to take the #3 spot |
| 22:58 | <annevk> | am I closer to Hixie or further away? |
| 22:59 | <annevk> | better talk some more to myself about grocery shopping if I ever want to get that top spot for person talking the most in #whatwg |
| 23:03 | <annevk> | heycam: Web IDL does all the magic for exceptions to be constructable? |
| 23:03 | <annevk> | heycam: so I don't have to do anything? |
| 23:03 | <heycam> | annevk, yep |
| 23:04 | <annevk> | nice |
| 23:07 | <annevk> | whoa |
| 23:07 | <annevk> | www-style |
| 23:07 | <annevk> | >43 emails a day |
| 23:35 | <annevk> | spec for DRM http://dvcs.w3.org/hg/html-media/raw-file/tip/encrypted-media/encrypted-media.html :/ |
| 23:36 | <wilhelm> | Boo, hiss. |
| 23:37 | <TabAtkins_> | Ditto. |
| 23:37 | <smaug____> | huh |
| 23:37 | <heycam> | sigh |
| 23:38 | <zewt> | doesn't matter if nobody's clueless enough to implement it |
| 23:39 | <smaug____> | zewt: well, there are editors from two browser vendors |
| 23:39 | <zewt> | i giggled at the big nonsensical diagram |
| 23:39 | <smaug____> | and that worries me |
| 23:39 | <astearns> | the diagram is awesome - really draws you in |
| 23:42 | <zewt> | and of course the idea of anything being secure against the user browser-side is joke |
| 23:42 | <annevk> | CDN and CDM look like the real problematic parts |
| 23:42 | <rniwa> | oh dear... |
| 23:42 | <zewt> | the problematic part is ... it |
| 23:42 | <zewt> | heh |
| 23:42 | <annevk> | well sure |
| 23:45 | rniwa | secretly wishes ... |
| 23:46 | <rniwa> | and that'll be good for the Web. |
| 23:53 | <TabAtkins_> | I like how the Abstract says in its last sentence "No 'DRM' is added to the HTML5 specification" when that's precisely what the entire spec is about. These people have convinced themselves of a very narrow definition of "DRM" (or have convinced themselves that the rest of us have such a narrow definition). |