| 00:00 | <hober> | updating... |
| 00:01 | <mbrubeck> | I get "Read more:..." when I copy more than X words. |
| 00:02 | <mbrubeck> | around 8 words |
| 00:02 | <MikeSmith> | me finds "{if(b.isPointInRange(c,0)&&!isElementWithoutText.test(a.tagName))return c}else if(e== |
| 00:02 | <MikeSmith> | 1)i" etc. |
| 00:02 | <MikeSmith> | http://tcr.tynt.com/javascripts/Tracer.js |
| 00:03 | MikeSmith | notices his paste from that JS script got truncated, but that's the culprit, anyway |
| 00:03 | <MikeSmith> | anyway, that's just obnoxious |
| 00:04 | <mbrubeck> | "aa(b) > 7" --> at least 8 words |
| 00:04 | <MikeSmith> | yeah, seems so |
| 00:04 | <mbrubeck> | thanks, http://jsbeautifier.org/ |
| 00:05 | <mbrubeck> | I'd love to have a source code beautifier integrated with Firebug |
| 00:05 | <MikeSmith> | http://www.tynt.com/ |
| 00:05 | <MikeSmith> | "The copy/paste company" |
| 00:20 | <MikeSmith> | anybody know if is there are Java collection class that has a method for just popping the last item off the collection? |
| 00:22 | <MikeSmith> | hmm, Stack? |
| 00:22 | <Dashiva> | Stack is one of those do-not-use classes |
| 00:23 | <MikeSmith> | shit.. I was bout to say, me wonders if Stack is deprecated.. most of the obviously useful and simple Java classes are deprecated and replaced by something more complex and opaque |
| 00:23 | <Dashiva> | Do you have to support Java <6? |
| 00:23 | <MikeSmith> | Dashiva: so what do the Java gods mandate that I use instead? |
| 00:24 | <MikeSmith> | Dashiva: no, I think Java 6-only is OK |
| 00:24 | <Dashiva> | ArrayDeque is great if not |
| 00:24 | MikeSmith | checks ArrayDeque doc |
| 00:25 | <MikeSmith> | Dashiva: but, really, is the whole Java-let's-deprecate-everything phenomenon some kind of absurdist joke or something? |
| 00:25 | <Dashiva> | It's not deprecation for no reason |
| 00:25 | <mbrubeck> | It's because the old container classes were pre-generics |
| 00:25 | <MikeSmith> | oh |
| 00:25 | <Dashiva> | Also, lots of them were sychronized |
| 00:26 | <MikeSmith> | I see |
| 00:26 | <Dashiva> | Spelled correctly, anyhow |
| 00:26 | <MikeSmith> | I guess it's just that every time I find something that intuitively maps to my use cases and programming experiences, I find that's in deprecated |
| 00:27 | <Dashiva> | The good names were taken first, of course :) |
| 00:27 | <MikeSmith> | yeah |
| 00:28 | <Dashiva> | That's not to say you can't use the old classes, they still work. But they're deprecated because the alternatives are generally better. |
| 00:37 | <sid__> | btw i dont suppose any one has tried working with workers API in chrome extensions....? |
| 00:37 | <sid__> | they dont seem to work in the JS environment... |
| 00:37 | <sid__> | set up for extensions |
| 01:42 | <Hixie> | if there are any forum admins around (zcorpan?), looks like DaveC426913 is sending spam PMs to people on the forum |
| 01:44 | <TabAtkins> | Anyone with more than 2 digits at the end of their name (4 if it looks like a year in the last century) is a spam bot automatically. |
| 01:50 | <KaOSoFt> | I already reported them. I guess that was the reason for the net split. |
| 01:50 | <KaOSoFt> | Perhaps restarted a server? |
| 01:50 | <KaOSoFt> | Anyways, time to go home. |
| 04:17 | <boblet> | MikeSmith: are you in charge of the sites listed in planet html5? |
| 04:17 | <boblet> | this might be a good feed to add http://doctype.com/tags/html5 |
| 04:18 | <MikeSmith> | boblet: yeah, I am |
| 04:18 | MikeSmith | looks at site now |
| 04:18 | <boblet> | not quite as active as SO, but similar in a lower level kinda way |
| 04:18 | <boblet> | well, less technical |
| 04:19 | <MikeSmith> | can you vouch for these guys? |
| 04:19 | <MikeSmith> | Doctype was built by the people who run Litmus - David Smalley, Paul Farnell and Matthew Brindley. |
| 04:19 | <MikeSmith> | (quote from about page) |
| 04:19 | <boblet> | btw thanks for adding me yo |
| 04:19 | <boblet> | litmus is a compare-in-different-browsers app for web devs |
| 04:20 | <boblet> | don’t know them personally but they haven’t done anything dodgy so far |
| 04:20 | <boblet> | they made me a site admin fwiw (not much :) |
| 04:23 | <boblet> | oh btw I’m not posting much HTML5-relevant stuff to Tumblr now, so you may want to change my link to http://oli.jp/articles.atom |
| 04:23 | <boblet> | most of my content is going to html5doctor.com or books now tho :/ |
| 04:25 | <MikeSmith> | losted my connection |
| 04:25 | <MikeSmith> | [12:25] <MikeSmith> np |
| 04:25 | <MikeSmith> | [12:26] <MikeSmith> I'm kidding about vouching for them |
| 04:25 | <MikeSmith> | [12:26] <MikeSmith> I'll add it now |
| 04:25 | <MikeSmith> | [12:26] <MikeSmith> it looks like a useful resource |
| 04:25 | <boblet> | heh |
| 04:25 | <MikeSmith> | boblet: will change the URL for you feed too |
| 04:25 | <MikeSmith> | dammit |
| 04:25 | <MikeSmith> | this train driver loves to hit his brakes |
| 04:25 | <MikeSmith> | and make everybody stumble |
| 04:25 | <boblet> | woah, that’s a dodgy feeling huh |
| 04:26 | <boblet> | well, it is on JR West |
| 04:26 | <boblet> | :| |
| 04:26 | <MikeSmith> | train drivers should be wired to get an electric shock every time they do that |
| 04:26 | <MikeSmith> | this it Odakyu sen |
| 04:26 | <MikeSmith> | the only thing worse than JR |
| 04:26 | <boblet> | heh |
| 04:26 | <boblet> | as long as they stay on the rails I’m happy |
| 04:27 | <MikeSmith> | I'd rather they also tried harder to arrive on time |
| 04:28 | <MikeSmith> | Odakyu is the online train line I've seen in Japan that can't manage to get trains to run on time |
| 04:28 | <MikeSmith> | and plus every time it rains, all hell seems to break lose |
| 04:29 | <boblet> | maybe they’re using the British Rail model |
| 04:29 | <boblet> | leaves on the track! oh noes! |
| 04:30 | <MikeSmith> | the other day when it was raining here, they announced "We've had an accident on the ENoshima line, so we will not go all the way to Shonandai (my station), instead we will stop at Sagami Ono".. so, fuck you |
| 04:30 | <MikeSmith> | basically |
| 04:30 | <MikeSmith> | there are no other train lines that connect to Sagami Ono |
| 04:30 | <MikeSmith> | and they gave no info about, if you need to get further down to Fujisawa or whatever, this is what you should to |
| 04:30 | <MikeSmith> | *do |
| 04:31 | <boblet> | that’s … helpful |
| 04:31 | <MikeSmith> | instead, it was just "get off the train and don't let the doors hit you in the ass on the way out" |
| 04:31 | <boblet> | (for taxi companies) |
| 04:31 | <MikeSmith> | boblet: OK, doctype.com added |
| 04:31 | <MikeSmith> | willbchange your ULR now |
| 04:32 | <boblet> | MikeSmith: no rush |
| 04:32 | <boblet> | I was actually a bit surprised to see myself there. “I’d just like to thank my family and fans—without them…” ;-) |
| 04:33 | <MikeSmith> | heh |
| 04:34 | <MikeSmith> | boblet: you have made yourself a member of the cabal, for better or worse |
| 04:34 | <boblet> | feel obligated to write more :/ |
| 04:34 | <boblet> | wow, sweet! looking forward to those holographic member ids you told me about |
| 04:34 | <boblet> | the one with hypno-powers |
| 04:35 | <MikeSmith> | now, when people bitch about the "HTML clusterfuck" and the people responsible, you can count yourself among the privileged few they have in mind |
| 04:35 | <boblet> | rock! … umm I think |
| 04:47 | <MikeSmith> | so boblet |
| 04:47 | <MikeSmith> | if you can indulge me |
| 04:47 | <MikeSmith> | I have something to share |
| 04:48 | <MikeSmith> | I went to the supermarket this morning |
| 04:48 | <boblet> | I’m always down for your indulgences |
| 04:48 | <MikeSmith> | to buy some english muffins |
| 04:48 | <boblet> | mm muffins |
| 04:48 | <MikeSmith> | I got to the store and found that they had two brands of english muffins |
| 04:48 | <MikeSmith> | and they looked a lot the same |
| 04:49 | <MikeSmith> | and were the same price |
| 04:49 | <MikeSmith> | so I'm confronted with a dilemma |
| 04:49 | <MikeSmith> | how do I choose? |
| 04:49 | <MikeSmith> | what process do I use to decide |
| 04:49 | <boblet> | elementary. buy one of each and have a taste-off |
| 04:49 | <MikeSmith> | I considered the options available to me |
| 04:49 | <boblet> | don’t thank me, thank science |
| 04:50 | <MikeSmith> | I could, say, try to get some expert advice from the management |
| 04:50 | <boblet> | heh |
| 04:50 | <boblet> | good luck with that |
| 04:50 | <MikeSmith> | well, I could even do a lifeline call to a english-muffin expert somewhere |
| 04:50 | <MikeSmith> | or I could attempt to do some detailed analysis there myself, at the stor |
| 04:50 | <MikeSmith> | close examination |
| 04:51 | <MikeSmith> | I mean, I could have spent all day there at the store deciding what english muffins to buy |
| 04:51 | <MikeSmith> | and agonizing and hand-wringing over whether I was making the right decision |
| 04:51 | <MikeSmith> | because, you know, I had to make *some* decision |
| 04:52 | <MikeSmith> | I needed english muffins |
| 04:52 | <boblet> | obviously |
| 04:52 | <MikeSmith> | I had an english-muffing consumer market waiting at home to consume some english muffins |
| 04:53 | <MikeSmith> | and I thought, I have a risk here of making the wrong decision |
| 04:53 | <MikeSmith> | and alienating my english-muffin consumer market at home |
| 04:53 | <boblet> | I stand by my scientific experiment suggestion (which would also ensure seconds for your aforementioned consumer market), but other variables worthy of evaluation are |
| 04:53 | <MikeSmith> | having them tell me, boy, Mikey, you sure really fucked up that english-muffin-buying decision |
| 04:54 | <MikeSmith> | "what a mess you made of the english-muffin-buying task that you were given responsibility for" |
| 04:54 | <MikeSmith> | etc. |
| 04:54 | <MikeSmith> | anyway, you know the punchline |
| 04:55 | <boblet> | weight, relative springiness, length of ingredients list, relative average length of ingredients in ingredients list (in both cases shorter=better), and due-by date |
| 04:55 | <boblet> | the other option is to just lie and say that’s all they had |
| 04:55 | <MikeSmith> | well, after 10 seconds of agonized though about it, I just grabbed one and took it the counter and paid |
| 04:56 | <MikeSmith> | and I got back in time to ship the code |
| 04:56 | <MikeSmith> | er, I mean I got back in time to have breakfast |
| 04:56 | <MikeSmith> | anyway, it would be a mistake for anybody to read too much into this little story |
| 04:57 | <MikeSmith> | I just felt like sharing |
| 04:57 | <boblet> | damn you and your finely crafted metaphors |
| 04:57 | <MikeSmith> | "get a decision made in time to actually eat breakfast" |
| 04:58 | <boblet> | don’t forget that it’s entirely possible you would still have got into trouble for choosing the wrong muffin, for no apparent reasons (consumer markets can be unforgiving in the morning) |
| 04:59 | <MikeSmith> | yeah, the damned-if-you-do-or-damned-if-you-don'tness of life |
| 04:59 | <MikeSmith> | Huck Finn wrote about that |
| 04:59 | <MikeSmith> | well, Mark Twain did |
| 04:59 | <MikeSmith> | I think Huck's conclusion was, I guess I'll just be damned. |
| 05:00 | <MikeSmith> | but he got Jim out to safety |
| 05:00 | <MikeSmith> | and he had a pretty good time along the way doing it |
| 05:00 | <MikeSmith> | and he learned a lot |
| 05:02 | <boblet> | “just be damned” that’d make a good quote for the IRC logs header |
| 05:02 | <MikeSmith> | Huck shipped the code |
| 05:10 | <MikeSmith> | boblet: oli.jp feed added |
| 05:11 | <boblet> | heh, and cheers |
| 05:11 | <boblet> | will post wisely |
| 05:17 | <MikeSmith> | no need to do it wisely |
| 05:18 | <MikeSmith> | abuse the Force |
| 05:33 | <MikeSmith> | boblet: let's try to be more like this guy: http://twitpic.com/1rfowk |
| 05:33 | <MikeSmith> | or perhaps better yet: http://twitpic.com/1rfp6m |
| 05:47 | <boblet> | lol |
| 05:48 | <boblet> | MikeSmith: http://www.collegehumor.com/article:1794889 |
| 06:05 | <MikeSmith> | so there's a patch copied into a code.google.com bug report that I would like to e-mail to hsivonen |
| 06:05 | <MikeSmith> | but I notice that the code.google.com UI has wrapped and borked it |
| 06:05 | <MikeSmith> | http://code.google.com/p/jing-trang/issues/detail?id=35 |
| 06:05 | <MikeSmith> | anybody know if there's some way I can get a raw/unwrapped version of the comment? |
| 06:09 | MikeSmith | finds that the patchutils distro has a Unwrapdiff command |
| 07:00 | <hsivonen> | OK. I guess I should email fantasai about URLs on syntax.whattf.org |
| 07:19 | <Peter`> | MikeSmith: The blue background-colour is gone now, so yes, text is readable |
| 07:19 | <Peter`> | Thank you! |
| 07:20 | <MikeSmith> | cheers |
| 08:38 | MikeSmith | sends some patches to hsivonen for review |
| 08:40 | <annevk> | http://docs.google.com/Doc?docid=0AaYxrITemjbxZGNmZzc5cHpfM2Ryajc5Zmhx&hl=en scroll to the end -- DOCTYPE rejoicing |
| 08:40 | <annevk> | via Peter- |
| 08:41 | <annevk> | or maybe Peter` if they're different :) |
| 08:42 | <Peter`> | Nah, two computers |
| 08:44 | <slartsa> | whoa, AI? |
| 08:47 | <hsivonen> | annevk: tag-format="semantics/1.0" is awesome |
| 08:52 | <Rich_Clark> | can anyone tell me why there isn't a type="year" in the spec? |
| 08:53 | <annevk> | use case? |
| 08:54 | <MikeSmith> | hey, it's Rich_Clark!! |
| 08:54 | <MikeSmith> | HTML5 Doctors are taking over |
| 08:55 | <Rich_Clark> | hey Mike |
| 08:55 | <MikeSmith> | hey man |
| 08:55 | <MikeSmith> | for the record, I don't have clue about the rational for not having type=year |
| 08:55 | <MikeSmith> | it seems to me like it'd be useful |
| 08:55 | <Rich_Clark> | anne, I don't have one, someone just asked me the other day and I didn't know so thought I'd see if anyone did |
| 08:55 | <MikeSmith> | but as always, whatta I know.. |
| 08:56 | <annevk> | well, the answer is lack of a use case afaik |
| 08:56 | <annevk> | MikeSmith, more than most people, I'd say |
| 08:56 | <MikeSmith> | annevk: great to see http://www.w3.org/TR/speech-grammar/grammar.dtd |
| 08:56 | <Rich_Clark> | right ok, cheers guys |
| 08:56 | <MikeSmith> | keeping hope alive for DTDs |
| 08:56 | <MikeSmith> | talk about lack of clue |
| 08:57 | <hsivonen> | Rich_Clark: as far as I am aware, the use cases are addressed by type=number min=something step=1 |
| 08:57 | <annevk> | even has <meta>, including the broken http-equiv |
| 08:57 | <annevk> | awesome |
| 08:57 | <MikeSmith> | we really need to have an executive ban on DTDs across working groups |
| 08:57 | <MikeSmith> | seriously |
| 08:57 | <MikeSmith> | the madness must end |
| 08:57 | <MikeSmith> | otherwise these jackasses will still be churning out DTDs 50 years from now |
| 08:58 | <Rich_Clark> | hsivonen: yes that's what I advised the guy to use in the end. Cheers |
| 08:58 | <hsivonen> | is special (NULL | VOID | GARBAGE) #IMPLIED |
| 08:58 | <hsivonen> | legal in DTDs? |
| 08:58 | <annevk> | another ban: if you intend your spec to end up the web browser vendors must be involved; if they're not and you continue ahead anyway you lose the right to complain later on |
| 08:59 | <MikeSmith> | I'm trying hard to un-learn anything I ever knew about DTDs |
| 08:59 | <MikeSmith> | annevk: amen |
| 08:59 | <MikeSmith> | again, seriously |
| 09:01 | <MikeSmith> | groups of mad scientists off quietly spending years in their remote labs developing Frankenstein monsters |
| 09:01 | <MikeSmith> | then unleashing them on the world and asking that they be welcomed with open arms |
| 09:58 | <zcorpan_> | hsivonen: i think so, although the html4 dtd usually sets a default value instead of #IMPLIED for enumerated attributes that aren't booleans |
| 09:59 | <zcorpan_> | method (GET|POST) GET |
| 09:59 | <zcorpan_> | checked (checked) #IMPLIED |
| 10:35 | <jgraham> | https://wiki.mozilla.org/Tantek-Mozilla-projects - the bit about form controls is interesting |
| 10:43 | <Lachy> | "Take a look at what Opera has done for example (in terms of dangers to avoid)." - haha |
| 10:54 | <zcorpan_> | http://twitter.com/bobdvb/statuses/14818754527 |
| 10:54 | <Mitsurugi> | hi all |
| 10:59 | <jgraham> | hi |
| 11:04 | <jgraham> | Does anyone know if it is defined somewhere that new InterfaceObject() and InterfaceObject() are equivalent? Or does it have to be specified on a case-by-case basis? |
| 11:09 | <jgraham> | (I would expect WebIDL 4.4.1 to define this, but it doesn't) |
| 11:09 | <jgraham> | (unless I am missing something) |
| 11:12 | zcorpan_ | notes 15.3.1.1 in es5 |
| 11:13 | <gsnedders> | zcorpan_: That's not relevant |
| 11:13 | <gsnedders> | zcorpan_: That's only relevant for the Function object |
| 11:15 | <gsnedders> | As in what is bound to "Function" in the global scope be default |
| 11:16 | jgraham | wonders if any work on WebIDL is actually happening |
| 11:26 | <zcorpan_> | so for xhr it seems firefox and opera return an object while safari and chrome throw TypeError |
| 11:27 | <jgraham> | TypeError is what I would expect given the spec |
| 11:27 | <jgraham> | But I don't really trust the spec :) |
| 11:28 | <zcorpan_> | although typeof XMLHttpRequest == 'function' |
| 11:28 | <gsnedders> | That's expected |
| 11:28 | <gsnedders> | It's just the difference between [[call]] and [[construct]], so even if it only implements [[construct]] I'd expect that to be true |
| 11:29 | <zcorpan_> | ok |
| 11:29 | <annevk> | zcorpan_, I don't get that twitter message |
| 11:30 | <zcorpan_> | annevk: then we're two |
| 11:30 | <annevk> | though I've heard the sentiment about not bundling codecs before and "letting the market decide" but I don't really buy that either |
| 11:30 | <jgraham> | It is kindof supposed to be an invariant of the language that typeof x === "function" <=> x implements [[call]] |
| 11:31 | <jgraham> | Although regexp objects break that (in some implementations?) |
| 11:31 | <annevk> | I think everything with [Constructor] should not have [[call]] |
| 11:31 | <jgraham> | annevk: That would require a change to ECMAScript |
| 11:31 | <annevk> | e.g. x = Image() rather than x = new Image() is wrong imo |
| 11:32 | <jgraham> | Oh I see |
| 11:32 | <jgraham> | Well all the builtins already work like that |
| 11:32 | <annevk> | with the same semantics? |
| 11:32 | <jgraham> | Yeah |
| 11:32 | <annevk> | I thought there was some difference |
| 11:33 | <annevk> | maybe I should decide not to care and let someone else figure it out |
| 11:33 | <jgraham> | Actually taht's not quite true |
| 11:33 | <jgraham> | it is true for Array |
| 11:33 | <jgraham> | But not for String |
| 11:34 | <jgraham> | String(a) cocerces a to a string primitive, new String(a) coerces a to a string primitive and wraps that primitive in a string object |
| 11:35 | <jgraham> | (the language bug here is that String objects exist) |
| 11:35 | <hsivonen> | zcorpan_: ah. shows how little I remember about DTDs. I didn't remember that enumerated tokens don't need to be quoted |
| 11:36 | <jgraham> | Date is also weird |
| 11:36 | <jgraham> | Date() returns a string new Date() returns a Date object |
| 11:38 | <MikeSmith> | has anybody here reviewed and/or commented on http://www.w3.org/2006/WSC/drafts/rec/rewrite.html |
| 11:38 | <MikeSmith> | Web Security Context: User Interface Guidelines |
| 11:39 | <jgraham> | So basically ES follows [[Call]] === [[Construct]] except for String, Number, Boolean (none of which should exist), Date (which is a mess) and RegExp in the case where the first argument is a preexisting regexp object |
| 11:40 | <jgraham> | MikeSmith: Did they have any browser vendors involved in writing the document? |
| 11:40 | <MikeSmith> | yeah, some |
| 11:40 | <MikeSmith> | Ian Fette was somewhat involved in that WG at times |
| 11:40 | <MikeSmith> | don't know about recently |
| 11:40 | <MikeSmith> | I was in the group long ago too |
| 11:40 | <MikeSmith> | when I was working at Opera |
| 11:41 | <MikeSmith> | some Mozilla security folks too |
| 11:43 | <MikeSmith> | hsivonen: you got other document.write tests? you saw abarth discussion over on #webkit? |
| 11:43 | <MikeSmith> | or other "crazy parsing" tests |
| 11:43 | <MikeSmith> | other than html5lib ones |
| 11:46 | <MikeSmith> | zcorpan_: btw, I guess I need to try finishing the v.nu checker code I was writing for script element text-content checking |
| 11:46 | <MikeSmith> | so I might be bugging you for test cases |
| 11:47 | <zcorpan_> | MikeSmith: ok |
| 11:47 | <MikeSmith> | I really hate that part of the spec |
| 11:48 | <hsivonen> | MikeSmith: just Hixie's tests and sicking's tests in mochitest on m-c |
| 11:48 | <MikeSmith> | hsivonen: ah, OK |
| 11:49 | <MikeSmith> | I figured you guys might have written some new tests after the document.write-related evangelism bugs that came up |
| 11:49 | <MikeSmith> | iirc at least there were some document.write-related ones |
| 11:49 | <hsivonen> | oh, and then http://hsivonen.iki.fi/test/moz/detect-html5-parser.html has an evil document.write test that WebKit doesn't pass :-) |
| 11:51 | <MikeSmith> | excellent |
| 11:51 | jgraham | wonders what is being discussed in #webkit |
| 11:57 | <Peter`> | jgraham: they're working on an HTML5 tokenizer |
| 12:00 | <slartsa> | does an iframe go over all z-indexes? |
| 12:03 | <MikeSmith> | hsivonen: is there any reason I should not (or could not) create two separate checker classes for script checking? |
| 12:03 | <MikeSmith> | that is, one for http://dev.w3.org/html5/spec/scripting-1.html#restrictions-for-contents-of-script-elements |
| 12:03 | <MikeSmith> | and one for http://dev.w3.org/html5/spec/scripting-1.html#inline-documentation-for-external-scripts |
| 12:05 | <slartsa> | oh, wrong channel |
| 12:05 | <MikeSmith> | zcorpan_: I think I may have already finished my attempt at implementing support for http://dev.w3.org/html5/spec/scripting-1.html#inline-documentation-for-external-scripts |
| 12:06 | <MikeSmith> | for the script/@src case |
| 12:06 | <MikeSmith> | that part of the checking |
| 12:06 | <zcorpan_> | MikeSmith: cool |
| 12:06 | <MikeSmith> | deployed on http://www.w3.org/html/check |
| 12:06 | <MikeSmith> | for testing |
| 12:19 | <zcorpan_> | should we make buffered.start() and end() work without an argument? |
| 12:19 | <zcorpan_> | defaulting to 0 for start and buffered.length-1 for end? |
| 13:22 | <foolip> | zcorpan_: sounds a bit odd, did anyone ask for that? |
| 13:26 | <zcorpan_> | foolip: no |
| 13:26 | <zcorpan_> | foolip: i thought it would be a convenience shortcut |
| 13:27 | <zcorpan_> | i think webkit acts as if undefined was passed as argument if you don't have any, which will be converted to 0 |
| 13:28 | <zcorpan_> | whereas opera would throw WRONG_ARGUMENTS_ERR |
| 13:29 | zcorpan_ | notes that there's <meta name=application-name> that installable web apps could use |
| 13:39 | <Lachy> | zcorpan_, yes, I mentioned application-name in my reply on whatwg earlier. |
| 13:41 | <foolip> | zcorpan_: is WebKit or Opera wrong? |
| 13:43 | <zcorpan_> | foolip: behavior is undefined (in webidl) |
| 13:44 | <zcorpan_> | foolip: webkit has the ecmascript-like behavior |
| 13:45 | <zcorpan_> | http://lists.w3.org/Archives/Public/public-webapps/2009AprJun/1171.html |
| 13:46 | <foolip> | undefined? |
| 13:46 | <foolip> | oh my |
| 13:51 | <jgraham> | foolip: No other behaviour really makes sense in ECMAScript |
| 13:51 | <jgraham> | foo() should be exactly like foo(undefined) if foo takes a single formal argument |
| 13:52 | <jgraham> | Oh, wait |
| 13:52 | <jgraham> | You are complaining about the behaviour being undefined |
| 13:52 | <jgraham> | Yeah, that sucks |
| 13:57 | <boblet> | Rich_Clark: hey hey! |
| 14:01 | <boblet> | annevk: re: form input for type=year", what about adding the year of a wine on corkd.com? |
| 14:02 | <boblet> | annevk: alternatively, is there a way to get year-only input validated/via an appropriate browser widget? |
| 14:05 | <Rich_Clark> | boblet: hey mate |
| 14:05 | <boblet> | good to see you’ve answered MikeSmith’s siren call |
| 14:06 | <boblet> | (he’s luring us all onto the rocks of the cabal y’know) |
| 14:06 | <MikeSmith> | walk towards the light |
| 14:06 | <boblet> | I’m pretty sure Gollum said the opposite |
| 14:07 | <boblet> | I must be in the wrong movie |
| 14:07 | <jgraham> | I read that as "rocks of the canal" and wondered what kind of braindead architects would design a shipping channel with obvious hazartds to navigation |
| 14:07 | <boblet> | jgraham: now there’s a loaded metaphor |
| 14:50 | <erlehmann> | can anyone tell me whats so wrong with the audio element used in here that chrome does not play it ? http://blog.dieweltistgarnichtso.net/podcast-podcasts-feat-sikk |
| 14:50 | <erlehmann> | maybe i am using it wrong … |
| 15:04 | <erlehmann> | seems to be this http://code.google.com/p/chromium/issues/detail?id=33437 |
| 15:09 | <Lachy> | erlehmann, that's not working in Opera either. I don't know why |
| 15:10 | <Lachy> | hmm, it works when I load it locally |
| 15:14 | <erlehmann> | but Content-Type is: audio/ogg … Lachy, does it have to be more specific ? |
| 15:16 | <Lachy> | no, that should be fine |
| 15:16 | <Philip`> | erlehmann: I've had troubles when audio is sent with Content-Encoding: gzip |
| 15:16 | <Philip`> | (which it looks like yours is) |
| 15:17 | <Philip`> | Should try disabling compression for that file |
| 15:17 | <Lachy> | The content-encoding shouldn't affect playback. If that's the cause, file a bug so we can get that fixed |
| 15:18 | <erlehmann> | Philip`, how do i check for compression ? |
| 15:19 | <Philip`> | erlehmann: "curl --compressed -I http://daten.dieweltistgarnichtso.net/audio/podcasts/podcasts-feat-sikk.oga"; would do it |
| 15:19 | <erlehmann> | the resource tab in chrome says it spends 6ms to wait for the resource, then cancels after 1ms |
| 15:22 | <erlehmann> | i'll see what wireshark says about google chromes behaviour |
| 15:41 | <asmodai> | hsivonen: looks like the latest 3.6.4 stabilized the memory issues. |
| 15:42 | <asmodai> | At least, that's the initial idea. Need to test some more, obviously |
| 15:49 | <hsivonen> | asmodai: ok. have you tried on trunk? |
| 15:54 | <asmodai> | hsivonen: not yet |
| 16:23 | <erlehmann> | Philip`, it plays now. Thank you. |
| 16:23 | <erlehmann> | !seen mpilgrim |
| 16:31 | <Philip`> | erlehmann: Because of the gzip thing? |
| 16:31 | <erlehmann> | Philip`, seems so. |
| 16:31 | <MikeSmith> | so I'm trying to consider the ABNF in http://dev.w3.org/html5/spec/scripting-1.html#inline-documentation-for-external-scripts in terms of what the error cases are that v.nu would need to report about |
| 16:32 | <erlehmann> | i'll shoot mpilgrim a mail, his server also serves gzipped ogg content |
| 16:32 | <AryehGregor> | erlehmann, mpilgrim reads logs here. |
| 16:32 | <MikeSmith> | I know one is the same as with the style element |
| 16:32 | <AryehGregor> | You should be able to say something and he'll come in to respond later. |
| 16:32 | <AryehGregor> | At least in my experience. |
| 16:32 | <Philip`> | erlehmann: I expect it'd be good to file bugs on any browsers that fail in that case |
| 16:33 | Philip` | is too lazy to file bugs himself but doesn't mind suggesting other people do the work |
| 16:33 | <erlehmann> | Dear log-reading mpligrim, I suggest you add this rule to the .htaccess file at diveintohtml5.org: SetEnvIfNoCase Request_URI \.(?:oga|ogv|ogg)$ no-gzip |
| 16:33 | <erlehmann> | Best regards, |
| 16:33 | <erlehmann> | erlehmann |
| 16:33 | <erlehmann> | :D |
| 16:34 | <jgraham> | You know he saw it if it turns up in yellow |
| 16:34 | <jgraham> | :) |
| 16:35 | <erlehmann> | Lachy, does it work in opera now ? |
| 16:35 | <MikeSmith> | error message: "Content contains the character sequence \u201c<!--\u201d without a later occurrence of the character sequence \u201c-->\u201d." |
| 16:36 | <MikeSmith> | and another is "Content contains the character sequence "<script>" without a later occurrence of the character sequence "</script>" |
| 16:36 | <erlehmann> | hmm, does opera even have a public bug tracker ? |
| 16:36 | <Philip`> | erlehmann: The public can submit bugs |
| 16:37 | <AryehGregor> | Just not actually be informed of any progress on them. |
| 16:37 | <AryehGregor> | Opera now officially has the least open bug tracker of any browser vendor. |
| 16:37 | <erlehmann> | Sounds like lots of dupes. |
| 16:37 | <AryehGregor> | I whine at the Opera employees here about it sometimes, but they haven't managed to get it fixed yet. |
| 16:37 | <jgraham> | AryehGregor: Opinions on the merits of this vary |
| 16:38 | <AryehGregor> | erlehmann, what, as opposed to the public bug trackers where half the bugs filed *aren't* dupes? :) |
| 16:38 | <erlehmann> | anyone with opera can check if http://blog.dieweltistgarnichtso.net/podcast-podcasts-feat-sikk plays now ? i am on throttled UMTS |
| 16:38 | <AryehGregor> | jgraham, yes, yes, I know. I'll continue to make fun of you as long as you're more secretive than Microsoft, though. |
| 16:38 | <Philip`> | Microsoft has a secretive bug tracker |
| 16:38 | <Philip`> | They just have a public one in addition to it |
| 16:39 | <AryehGregor> | So that's less secretive than Opera, then. |
| 16:39 | <Philip`> | but I don't think that makes it clear which is most secretive overall |
| 16:39 | <AryehGregor> | Which has only a secret bug tracker. |
| 16:39 | <Philip`> | Opera has forums and blog posts where people complain about bugs |
| 16:39 | <AryehGregor> | I imagine Apple also has a secret bug tracker for tracking non-WebKit Safari issues, unless that's also public? |
| 16:39 | <Philip`> | which are basically public bug trackers |
| 16:40 | <Philip`> | (just without any good bug tracking features) |
| 16:40 | <AryehGregor> | No, because you don't get told what the status is. |
| 16:40 | <jgraham> | If we did get a public bug tracker I imagine it would be disconnected from our internal one |
| 16:40 | <AryehGregor> | I mean, by that argument, me complaining to jgraham on IRC is a public bug tracker. |
| 16:40 | <jgraham> | Apple has rdar |
| 16:40 | <jgraham> | or whatever it is called |
| 16:41 | AryehGregor | doesn't touch anything manufactured by Apple with a ten-foot pole, so doesn't have reason to know |
| 16:42 | <jgraham> | Complaining on IRC is not like filing bugs. On the other hand I personally have looked through the desktop blog comments for useful bug reports |
| 16:43 | <jgraham> | The signal to noise is pretty low though (at least for the bugs that I care about rather than "this UI element is 2px off from where I would ideally like it") |
| 16:43 | <jgraham> | (which might be bugs that someone else cares about) |
| 16:43 | <AryehGregor> | Signal-to-noise drops drastically as you make the bug tracker more accessible. This is inevitable. |
| 16:44 | <AryehGregor> | But it makes an influential subset of users (the more techy people, who care most about the difference between browsers and understand them best) feel better. |
| 16:44 | <jgraham> | AryehGregor: Sure. I understand all the arguments for public bug trackers |
| 16:45 | <AryehGregor> | And I understand all the ones for private bug trackers. |
| 16:46 | <AryehGregor> | However, I operate on the theory that there is some level of persistent whining that will maximize people's chances of agreeing with you, and informal observational evidence suggests to me that this optimal level is nonzero. |
| 16:47 | <jgraham> | AryehGregor: Well in your case you are whining at the wrong people :) |
| 16:47 | <AryehGregor> | So I mention it occasionally when it comes up, to increase Opera employees' exposure to discontent about the status quo and bias them toward supporting a public bug tracker the next time it comes up internally. |
| 16:47 | <AryehGregor> | Oh well. |
| 16:47 | <AryehGregor> | Worth a try. |
| 16:47 | <AryehGregor> | If all else fails, whining is fun anyway. |
| 16:51 | AryehGregor | cannot be expected to conduct meaningful analysis on who the best person to whine at is in an organization whose internal structure is opaque, so just goes with the theory that everyone is responsible for everything. |
| 16:57 | <jgraham> | AryehGregor: I guess it is typically the case that the people working the most in public are not the ones arguning for more closedness. So anyone you can find to whine to is unlikely to be the right person to whine to. |
| 16:58 | <AryehGregor> | A good point. |
| 16:58 | <AryehGregor> | I'll be sure to only mock Opera for fun from now on, with no intent of accomplishing any useful goals. |
| 17:37 | <bellHead> | What is the current state of browser support for <datagrid>s? |
| 17:42 | <Dashiva> | Wasn't datagrid removed for now? |
| 17:43 | <bellHead> | That would be annoying, it's the capability I'm most looking forward to. |
| 17:43 | <bellHead> | Where would I find out if it had been removed? |
| 17:44 | <AryehGregor> | It has been. |
| 17:44 | <AryehGregor> | Months ago. |
| 17:45 | <AryehGregor> | Not enough implementation, I think, or not enough interest, or too complicated, or some combination of the foregoing. |
| 17:45 | <AryehGregor> | My impression was it would return in some future version, but it was thought best to concentrate on other things for this round. |
| 17:45 | <AryehGregor> | (i.e., the next few years) |
| 17:45 | <Dashiva> | Yeah, too much work remaining on more basic elements |
| 17:46 | bellHead | curses quietly. |
| 17:46 | <bellHead> | Oh well, thanks for the info. |
| 20:25 | <Dashiva> | All the important apps are being ported: http://html5zombo.com/ |
| 21:07 | <TabAtkins> | I spent an hour with that playing in the background the other day. It's strangely soothing. |
| 22:45 | <nessy> | MikeSmith: ping |
| 23:29 | <nessy> | MikeSmith: thanks - now it's the correct link :) |
| 23:30 | <MikeSmith> | nessy: which link? |
| 23:30 | <nessy> | in the email for the survey :) |