| 00:11 | <MikeSmith> | Hixie: I can give you some ISO-2022-JP content, if that does you any good |
| 00:11 | <Hixie> | it might |
| 00:12 | <Hixie> | reading http://en.wikipedia.org/wiki/ISO/IEC_2022 has not helped me in the slightest except to make me rather more alarmed |
| 00:14 | <MikeSmith> | Hixie: http://people.w3.org/mike/tests/iso-2022-jp-mail.txt |
| 00:14 | <MikeSmith> | fwiw |
| 00:15 | <MikeSmith> | oh, hang on |
| 00:15 | <MikeSmith> | I'll set the content-type on the web server for that properly |
| 00:26 | <MikeSmith> | Hixie: content-type is set correctly on that file now |
| 00:26 | <MikeSmith> | again, fwiw |
| 00:46 | <Hixie> | MikeSmith: thanks, will look |
| 00:46 | Hixie | pipes it through hexdump -C |
| 00:48 | <Hixie> | hmmm |
| 00:49 | <Hixie> | ok now to create a form that submits that |
| 00:51 | <Hixie> | in what encoding is é 0xc3 0xa9 ? |
| 00:51 | <Hixie> | wait, i have a tool to figure that out i think |
| 00:52 | <Hixie> | none that i know of, bummer |
| 00:53 | <Hixie> | well this isn't encoding as 2022, that's for sure |
| 00:53 | Hixie | pokes |
| 00:54 | <Hixie> | ooh, here we go |
| 00:54 | <Hixie> | 1b 24 42 |
| 00:55 | <Hixie> | aha, ESC $ B to switch to JIS X 0208-1983 (2 bytes per character) |
| 00:55 | <Hixie> | which matches what that e-mail has |
| 00:55 | <Hixie> | well well |
| 01:37 | <Hixie> | ok i need an ASCII-compatible encoding in which you can switch to a mode where 0x20 is not a space but is still an assigned codepoint (or part of one). |
| 01:37 | <Hixie> | anyone? :-) |
| 01:37 | <Hixie> | i feel like iso-2022 should provide the answer but it's proving quite impenetrable |
| 01:54 | <mkanat> | Hixie: That would make it not ASCII-compatible, wouldn't it? |
| 01:56 | ayo | circles "switch to a mode" |
| 01:59 | <ayo> | wouldn't utf-8 and U+0120 and the like do the trick? |
| 02:01 | <ayo> | 0120 Ġ 0220 ɐ 0320 ʹ 0420 Р 0520 Ẁ |
| 02:01 | <mkanat> | Ahh, switch to a mode. |
| 02:58 | <Hixie> | ayo: hmm, utf-8 might be an option, at least for "part of one" |
| 02:58 | <ayo> | :) |
| 04:38 | <MikeSmith> | Hixie: thanks for the quick reply to John Gregg on public-web-notification |
| 04:39 | <MikeSmith> | that spec has been languishing for a while |
| 04:39 | <MikeSmith> | would be nice to get it refined and actually implemented |
| 04:39 | <MikeSmith> | that is, implemented in more than one browser |
| 07:25 | <Hixie> | MikeSmith: yeah. |
| 07:25 | <Hixie> | MikeSmith: we really need to give spec writing lessons or something. |
| 07:25 | <Hixie> | MikeSmith: so many elementary things that few people understand |
| 07:25 | <Hixie> | MikeSmith: like the event olop |
| 07:26 | <MikeSmith> | well, some people like me learn best the hard way |
| 07:28 | <MikeSmith> | so doesn't always help me to get the lessons ahead of time, because I either am not just listening carefully then, or I'd forget by the time I get to doing whatever it is I need to do |
| 07:29 | <MikeSmith> | I think the best thing is for editors to get on IRC and ask |
| 07:29 | <MikeSmith> | while they are in the process of writing |
| 07:29 | <MikeSmith> | ask for feedback, ask for help when not sure about something |
| 07:30 | <Hixie> | getting the editors onto irc would be a good step forward |
| 07:32 | <MikeSmith> | I guess editors should actually read the HTML5 spec as well |
| 07:32 | <MikeSmith> | the "Web application APIs" at the very least |
| 07:32 | <Hixie> | not to mention the intro and infrastructure sections |
| 08:59 | <annevk> | so yeah we are not going to have more than one ID per element |
| 08:59 | <annevk> | if we align all specifications with the DOM ID concept that is |
| 09:12 | <annevk> | Hixie, so if you just have a single header row you should still use <thead> you think? |
| 09:28 | <hsivonen> | looks like Joe Hewitt is now defining "the Web" the way Tim Bray has been defining it for a while: http://joehewitt.com/post/what-the-web-is-and-is-not/ |
| 09:29 | <hsivonen> | short version: Android and iOS advocates want to call their stuff the Web |
| 09:31 | <annevk> | I am now even more confused |
| 09:32 | <annevk> | First he wants one browser, now he wants new browsers... |
| 09:33 | <annevk> | In a couple of days HTML might not be that bad after all |
| 09:41 | <jgraham> | Using google.com in chrome isn't using the web? |
| 09:43 | <hsivonen> | jgraham: I thought "Web" was one of Google's services alongside +, Gmail, Calendar, Documents, Photos and Sites |
| 09:44 | <hsivonen> | jgraham: "Web" is to GApps what Windows Desktop is to Metro :-) |
| 09:50 | <hsivonen> | looks like ze frank is another person who is famous enough for Google to bend their Real Names Policy |
| 16:39 | <dglazkov> | good morning, Whatwg! |
| 16:43 | <charlvn> | hi dglazkov |
| 16:50 | <timeless> | > Wonderful. We've no tracker, so try not to lose the slip of paper :-) |
| 16:50 | timeless | chuckles |
| 17:37 | <jwalden> | !summon ayo |
| 17:46 | <GlitchMr42> | http://gavinsharp.com/irc/whatwg.html |
| 17:46 | <GlitchMr42> | 404 |
| 17:48 | <gavin> | hrm |
| 17:48 | <gavin> | I should fix that |
| 17:51 | <jwalden> | gavin: why the log-rotation into old/? |
| 17:51 | <gavin> | welcome gavin_. he will generate stats for you. |
| 17:52 | <gavin> | jwalden: the machine previously logging/generating stats has moved to my parents basement |
| 17:52 | <jwalden> | heh |
| 17:52 | <gavin> | the new machine doing it has a smaller set of logs |
| 17:52 | <gavin> | rather than lose the historical stats I just moved them over to /old/ |
| 18:15 | <gavin> | http://gavinsharp.com/irc/whatwg.html |
| 18:30 | <Hixie> | ok i need an ASCII-compatible encoding in which you can switch to a mode where 0x20 is not a space but is still an assigned codepoint (or part of one). |
| 18:31 | <Hixie> | can't use utf-16, as it's not ascii-compatible |
| 18:31 | <Hixie> | iso-2022-jp seems like the way to go but i can't work it out :-/ |
| 18:35 | <AryehGregor> | Isn't such an encoding non-ASCII-compatible, by construction? |
| 19:01 | <hober> | AryehGregor: ascii-compatible when in another mode |
| 19:02 | <hober> | AryehGregor: but not while in the weird-0x20-mode |
| 19:22 | <AryehGregor> | Seriously? Google Docs doesn't allow font sizes below 8pt or line spacing below 1.00? How are you supposed to do a half-line break? |
| 19:22 | <AryehGregor> | Real users don't need precise control over font size or line spacing, I guess, right? |
| 19:22 | <AryehGregor> | . . . |
| 19:39 | <Hixie> | AryehGregor: what hober said |
| 19:40 | <Hixie> | looks like 0x20 is always a space in sio-2022-jp |
| 19:40 | <Hixie> | iso |
| 19:40 | <Hixie> | maybe there simply isn't an encoding with the property i describe |
| 19:41 | <Hixie> | that would certainly make my life easier |
| 19:42 | <hober> | Hixie: such an encoding would be perverted |
| 19:42 | <Hixie> | there are plenty of perverted encodings |
| 19:42 | <Hixie> | iso-2022-jp |
| 19:42 | <Hixie> | utf-7 |
| 19:42 | <hober> | indeed |
| 19:42 | <Hixie> | cesu |
| 19:43 | <Hixie> | gsm03.38 |
| 19:43 | <Hixie> | ebcdic |
| 19:43 | <Hixie> | and variants of the above |
| 19:43 | <Philip`> | Someone could always make up an encoding which has the property you describe |
| 19:43 | <Hixie> | just to name a few! |
| 19:44 | <Hixie> | Philip`: sure, but my concern only exists if it's an encoding that has legacy content to worry about (so i can't ban it) |
| 20:00 | <jmb> | Hixie: last I looked, iso 2022 didn't permit the redefinition of 0x20 (or 0x7f, for that matter) even if you redefined the mapping of G0. |
| 20:00 | <Hixie> | yeah that seems to indeed be the case |
| 20:00 | <Hixie> | i've decided to just assume no encodings allow it for now |
| 20:02 | <jmb> | that sounds sane :) |
| 20:02 | <Ms2ger> | In that case, it's probably wrong |
| 20:08 | <AryehGregor> | + <p class=note>This form data set encoding is in many ways an |
| 20:08 | <AryehGregor> | + aberrant monstrosity, the result of many years of implementation |
| 20:08 | <AryehGregor> | + accidents and compromises leading to a set of requirements necessary |
| 20:08 | <AryehGregor> | + for interoperability, but in no way representing good design |
| 20:08 | <AryehGregor> | + practices. In particular, readers are cautioned to pay close |
| 20:08 | <AryehGregor> | + attention to the twisted details involving repeated (and in some |
| 20:08 | <AryehGregor> | + cases nested) conversions between character encodings and byte |
| 20:09 | <AryehGregor> | + sequences.</p> |
| 20:09 | <AryehGregor> | I love notes like that. |
| 20:09 | AryehGregor | thinks he only has one or so in his own spec so far |
| 20:09 | <Ms2ger> | sheppy? Still trolling |
| 20:09 | <Ms2ger> | Me? Still not spelling well |
| 20:10 | <Hixie> | AryehGregor: i couldn't bring myself to checking in the change without apologising for it |
| 20:10 | <Hixie> | AryehGregor: hence that note :-) |
| 21:38 | <nlogax> | should one be able to read dataTransfer stuff in a dragstart event? http://jsfiddle.net/9YK5s/1/ |
| 21:39 | <nlogax> | trying to figure out some nice way to show where different kinds of stuff can be dropped |
| 21:39 | <nlogax> | but first i have to read it... :D |
| 22:30 | <nlogax> | Hixie: thoughts? :) |
| 22:30 | <nlogax> | spec says it's in "Read/write mode" |
| 22:30 | <Hixie> | hm? |
| 22:31 | <nlogax> | Hixie: tried to read DataTransfer data in a bubbling dragstart event, but failed. thought it was some bug.. but can't read it even in the listener that sets it. |
| 22:31 | <nlogax> | not sure what the spec says. it says read/write mode on one hand, but only mentions setting stuff on the other.. |
| 22:31 | Hixie | looks at the spec |
| 22:32 | <nlogax> | if you weren't around when i pasted the link, http://jsfiddle.net/9YK5s/1/ |
| 22:36 | <Hixie> | nlogax: per spec, dragstart should be able to read the data |
| 22:36 | <Hixie> | looking at your code now |
| 22:37 | <Hixie> | code looks fine |
| 22:37 | <Hixie> | dunno what's up |
| 22:40 | <nlogax> | yeah me neither :) |
| 22:40 | <nlogax> | other than bug filing time |
| 23:12 | <ojan> | tpac question: if i'm a member of the html wg but want to also attend the css wg meeting, do i need to check the observer box? |
| 23:13 | <Hixie> | in practice once you're there you can do whatever you want |
| 23:13 | <Hixie> | i didn't check any of the boxes, figure i'll just sit in the corridor working until someone wants me somewhere |
| 23:13 | <ojan> | Hixie: lol |
| 23:16 | <Hixie> | anyone got a testcase with dragleave that i can poke at? |
| 23:20 | <ojan> | Hixie: http://codesearch.google.com/codesearch#search/&exact_package=chromium&q=dragleave%20file:third_party/WebKit/LayoutTests&type=cs |
| 23:21 | <ojan> | Hixie: better search: http://codesearch.google.com/codesearch#search/&exact_package=chromium&q=dragleave%20file:third_party/WebKit/LayoutTests.*html&type=cs |
| 23:21 | <Hixie> | any of those self-contained? |
| 23:21 | <Hixie> | all those i found seemed to have a bunch of helper scripts |
| 23:22 | <Hixie> | (i'm just lazy and don't want to write my own) |
| 23:22 | <Hixie> | (trying to find out whatevent.relatedTarget is in dragleave events) |
| 23:24 | <ojan> | Hixie: http://codesearch.google.com/codesearch#OAMlx_jo-ck/src/third_party/WebKit/LayoutTests/http/tests/misc/bubble-drag-events.html&exact_package=chromium&q=dragleave%20file:third_party/WebKit/LayoutTests.*html&ct=rc&cd=3&sq= seems to be self-contained except for an external image |
| 23:24 | <ojan> | Hixie: nevermind... |
| 23:24 | <ojan> | Hixie: these all depend on test hardness specific stuff |
| 23:26 | <Hixie> | eh, don't worry about it |
| 23:26 | <Hixie> | i used the debugger to show that the spec matched webkit |
| 23:26 | <Hixie> | and closed the relevant bug saying that if they wanted it to change they should attach a testcase |
| 23:54 | <Hixie> | i wonder how to spec multipart/x-mixed-replace |
| 23:54 | <Hixie> | (in the navigate algorithm) |