| 00:36 | <Hixie> | zcorpan: yt? |
| 01:19 | <Hixie> | select.add() |
| 01:19 | <Hixie> | second argument |
| 01:19 | <Hixie> | IE: has to be a child of the <select> |
| 01:19 | <Hixie> | Safari: has to be a child of the <select> |
| 01:19 | <Hixie> | Mozilla: has to be a descendant of the <select> |
| 01:21 | <Hixie> | Opera: has to be an <option> descendant of the <select> |
| 02:37 | <jruderman> | mozilla's behavior is probably nicest for selects with optgroups |
| 03:24 | <olliej> | Hixie: you still trapped in <select> hell? |
| 07:29 | <hsivonen> | sicking: the elements that make a difference as fragment parents are at http://www.whatwg.org/specs/web-apps/current-work/multipage/parsing.html#reset-the-insertion-mode-appropriately |
| 07:30 | <hsivonen> | Philip`: I think I moved the finally block from the portable tokenizer to the Java IO driver. It is there because the SAX API contract is specified to require endDocument to be called with finally semantics |
| 07:30 | <hsivonen> | it indeed is a bit annoying sometimes |
| 07:34 | <hsivonen> | Hixie: if you want to spin graphics definitions out of the spec (or make them informative-only in the spec), PostScript Level 2 is probably the right normative reference |
| 07:34 | <Hixie> | hsivonen: ? |
| 07:37 | <hsivonen> | Hixie: Re: IRC log about the winding rule and canvas and you saying you'd love to spin stuff out |
| 07:41 | <hsivonen> | Philip`: what compression implementation is breaking HTTP streamingness? |
| 07:42 | <olliej> | yoyo roc |
| 07:43 | <roc> | gdday |
| 07:47 | <Hixie> | hsivonen: the winding rule thing is i just want a good original reference |
| 07:47 | <Hixie> | hsivonen: spinning things out is that i want a separate spec |
| 07:49 | <hsivonen> | ok. PostScript Level 2 is probably "original" from implementation point of view. I don't know if the concept has been stated earlier elsewhere than PostScript Level 1 |
| 08:05 | <zcorpan> | hmm wonder why the forums always get spammed while i'm on vacation |
| 08:06 | <zcorpan> | Hixie: here now |
| 08:06 | <Hixie> | zcorpan: spam issues on the forums |
| 08:07 | <Hixie> | zcorpan: i recommend making a few of the more effective moderators (e.g. the three who contacted me) into admins or some such |
| 08:07 | <Hixie> | hsivonen: the winding rule concept was invented in some paper or other originally, i'm sure |
| 08:08 | <hsivonen> | Hixie: I expect it to be originally from a research paper, yes. But if the details changed at all between that paper and PS Level 2, PS Level 2 compat is what drives libraries |
| 08:09 | <Hixie> | hsivonen: it's a pretty simple concept, i dunno how it could really change. but what do i know. :-) |
| 08:11 | <zcorpan> | Hixie: who contacted you? |
| 08:20 | <zcorpan> | i've made mskinner admin |
| 08:51 | <Philip`> | hsivonen: Does any contract require XmlSerializer.endDocument() to call OutputStream.close()? (I don't see why the serialiser would think it should take over control of the whole stream like that...) |
| 08:54 | <Hixie> | zcorpan: ryan jones |
| 08:55 | <Hixie> | oh and the other was mskinner again |
| 09:04 | <hsivonen> | Philip`: no, but that's the resonable behavior when you have one-to-one mapping with ContentHandler and OutputStream |
| 09:08 | hsivonen | finally managed to visualize the object relationships in Gecko's HTML and XML parsers |
| 09:15 | <Philip`> | hsivonen: It seems like there often isn't a one-to-one relationship, e.g. you'd often be serialising the document into an HTTP response where you've already written some header stuff to the stream, so the stream is not exclusively owned by the ContentHandler |
| 09:17 | <hsivonen> | Philip`: conceptually, the OutputStream you write XML to isn't the same OutputStream you write HTTP headers to but another OutputStream that just happens to delegate to the HTTP keep-alive-level stream |
| 09:17 | <hsivonen> | not only conceptually, but in practice, too |
| 10:27 | <Philip`> | Convincing LaTeX to place figures in the correct location is great fun |
| 10:34 | hsivonen | realizes 5 days later than annevk2 that there's now a blog for quoting bits out of the IRC logs http://lastweekinhtml5.blogspot.com/ |
| 10:48 | <Philip`> | http://alistapart.com/comments/webstandards2008?page=5#48 |
| 10:49 | <Philip`> | I don't think the WHATWG is the right place to come if what you want is a respectful environment |
| 10:53 | <Philip`> | (But it seems to work in terms of getting productive work done that nobody else would be doing) |
| 10:53 | <olliej> | who is that guy? and what's his problem? |
| 10:53 | <hsivonen> | it cuts both ways. the TAG uses a derogatory short name for their HTML5 meta issue (TagSoupIntegration-54) |
| 10:53 | <olliej> | hsivonen: and who/what is TAG? |
| 10:55 | <hsivonen> | olliej: the last week blog says "TAG - W3C Technical Architecture Group - members include Tim Berners Lee" |
| 10:55 | <annevk2> | olliej, http://www.w3.org/2001/tag/ |
| 10:56 | <olliej> | okayyy... and there approach to mature debate is to quote people out of context? |
| 10:56 | olliej | chooses not to pay any attention to such people |
| 10:56 | <olliej> | *their |
| 10:56 | <Philip`> | That blog is not related to the TAG |
| 10:56 | <olliej> | ah |
| 10:56 | <annevk2> | right |
| 10:56 | <olliej> | just a crazy person? |
| 10:57 | <Philip`> | And that blog isn't really quoting out of context - it has plenty of lines of the discussions, and a link to the whole context in the IRC logs |
| 10:57 | <annevk2> | misguided W3C fanboy? dunno :) |
| 10:58 | <annevk2> | I tried figuring it out, but couldn't find anything useful |
| 10:58 | hsivonen | expects this discussion to show up on the blog |
| 10:58 | <Philip`> | You're still not giving the impression of showing respect and understanding :-p |
| 10:59 | <annevk2> | Philip`, come on, do we have to be politically correct even on IRC? |
| 10:59 | <annevk2> | where is the fun in that |
| 10:59 | <Philip`> | annevk2: I don't mean to suggest that you should be showing respect and understanding |
| 10:59 | <Philip`> | It's boring when people do that |
| 11:00 | <Philip`> | Conflict with outsiders helps the group bind together more closely :-) |
| 11:01 | annevk2 | likes how it Philip` never grows tired of pointing out flaws |
| 11:06 | <hsivonen> | since the blogger seems to be pseudonymous, we can start guessing who (s)he is. |
| 11:08 | <mcarter> | anyone besides Hixie in the bay area? |
| 11:08 | <Philip`> | hsivonen: Why does Jean-Baptiste Clamence seem to be a pseudonym? |
| 11:08 | <annevk2> | the image is titled marcos.jpg |
| 11:08 | <mcarter> | i think it'd be neat to do a social event for html5/whatwg |
| 11:08 | <annevk2> | s/titled/named/ |
| 11:08 | <hsivonen> | Philip`: the name is a character from a novel by Albert Camus |
| 11:09 | <annevk2> | mcarter, I'll be in San Francisco / Mountain View in the week of October 26 |
| 11:09 | <Philip`> | annevk2: He's http://en.wikipedia.org/wiki/Subcomandante_Marcos |
| 11:09 | <annevk2> | (leaving October 31) |
| 11:10 | <Philip`> | (The person in the photo, not the person who wrote the blog) |
| 11:10 | MikeSmith | chuckles at the phrase "standards youth" |
| 11:10 | <annevk2> | Philip`, k |
| 11:10 | <mcarter> | annevk2, perfect, i'll start a wiki and we can see who all can meet up thenabouts |
| 11:17 | <roc> | woohoo, I've already featured in lastweekinhtml5 |
| 11:23 | <Philip`> | roc: You were even emboldened in it, which is a rare privilege |
| 11:23 | annevk2 | is reminded of http://html4all.org/pipermail/list_html4all.org/2007-August/000054.html for some reason |
| 11:24 | <Philip`> | though clearly they should use <em> instead of <span style="font-weight: bold;"> for that |
| 11:25 | <annevk2> | it seems only natural for an anti-fan blog to do everything wrong :) |
| 11:25 | <Philip`> | (or <em style="font-weight: bold; font-style: normal"> to get the desired presentation) |
| 11:26 | <Philip`> | They're an anti-fan of the WHATWG but presumably a fan of the W3C, and the W3C says you should use semantic elements too :-) |
| 12:22 | <krijnh> | Oh, more linklove :) |
| 12:29 | Philip` | has fun trying to fix a paper which he would not be able to understand at all, had he not spent the past year working on precisely what the paper is talking about |
| 12:56 | <Philip`> | http://lastweekinhtml5.blogspot.com/2008/10/monsieur-sivonen-makes-discovery.html didn't take long |
| 13:04 | <annevk2> | maybe if we talk about it enough his blog will be filled with us quoting him :p |
| 13:05 | <Philip`> | It'll be an infinite loop of each party saying "hey look, they talked about us! it's like we're famous now" |
| 13:06 | <Philip`> | Eventually we'll write a bot to automate the whole process |
| 13:10 | <Dashiva> | Philip`: Maybe someone on the inside is making it to draw attention to our IRC channel? |
| 13:12 | <virtuelv> | I really don't get the selective highlighting thing there |
| 13:18 | <Philip`> | Dashiva: You mean a kind of subtle PR campaign to promote the existence of #whatwg? |
| 13:27 | <hsivonen_> | Monsieur Clamence uses <span style="font-size: 85%;"> instead of <big> and <small> |
| 13:28 | <Philip`> | I guess that's what happens when you use the 'font size' dropdown menu in Blogspot's editing interface? |
| 13:28 | <hsivonen_> | that's possible. I haven't used Blogspot from the publisher side |
| 13:30 | <Philip`> | "Meet the WHAT WG Youth" - sounds like he's concerned that we are on his lawn |
| 13:30 | <krijnh> | http://1.bp.blogspot.com/_27WEu4KHCYU/SODF7QM1Y_I/AAAAAAAAAAM/6zGKJZWJD1E/S220/marcos.jpg - why is it marcos.jpg? |
| 13:30 | <hsivonen_> | krijnh: the blogger borrowed his face from dictator Marcos |
| 13:31 | <krijnh> | Ah, doh |
| 13:31 | <krijnh> | Never heard of him before :) |
| 13:31 | <Philip`> | (unless it's just meant to be a reference to the Hitler Youth) |
| 13:31 | <annevk2> | also, rtfl |
| 13:31 | annevk2 | already asked that |
| 13:31 | <Philip`> | (which might be a *little* too extreme) |
| 13:33 | krijnh | adds Meet the WHAT WG Youth to his subtitles :) |
| 13:33 | <hsivonen_> | oops. I had the wrong Marcos in mind. |
| 13:34 | hsivonen_ | managed to confuse various political Marcoses |
| 13:36 | <Philip`> | I have the advantage that I've never heard of any political Marcoses, so I just did an image search for "marcos" and the pipe and balaclava and hat were sufficiently distinctive to recognise him |
| 13:37 | <virtuelv> | Philip`: Ferdinand Marcos? |
| 13:37 | <hsivonen_> | virtuelv: I had Ferdinand in mind, but it's a Mexican Marcos |
| 13:37 | <Philip`> | Strangely Wikipedia gives the impression that he was actually serious and not just clowning around looking like that |
| 13:38 | <virtuelv> | http://en.wikipedia.org/wiki/Subcomandante_Marcos then? |
| 13:38 | <Philip`> | virtuelv: Who? :-) |
| 13:38 | <hsivonen_> | virtuelv: yes |
| 13:38 | <Philip`> | virtuelv: That one looks like the picture |
| 13:39 | <virtuelv> | Baader Meinhof syndrome |
| 13:39 | <virtuelv> | I encountered that WP article this morning in an entirely different context |
| 13:39 | <krijnh> | Perhaps it's just Marcos Caceres making a joke :) |
| 16:07 | <annevk2> | fyi: http://twitter.com/diveintomark/statuses/949726163 |
| 17:20 | <BenMillard> | hsivonen, whoever runs that blog seems pretty small-time |
| 17:21 | <BenMillard> | and nasty |
| 17:21 | gsnedders | bites |
| 17:26 | <BenMillard> | none of these images have alt text: http://lastweekinhtml5.blogspot.com/2008/10/meet-what-wg-youth.html |
| 17:26 | <BenMillard> | although that could be a red herring... |
| 17:39 | <BenMillard> | doing it anonymously is cowardly |
| 17:40 | <BenMillard> | deliberately choosing unflattering photos says more about the blogger's (or bloggers'?) personality than those photographed, imho |
| 17:40 | <BenMillard> | so the blogger is nasty-minded but too cowardly to own up to it |
| 17:40 | <BenMillard> | that narrows it down |
| 17:47 | <annevk_win32> | tell us when you get him sherlock :) |
| 17:47 | <BenMillard> | heh...it's like Cleudo :) |
| 17:47 | <BenMillard> | annevk_win32, whoever it is, what they're doing can only be good for WHATWG as it shows how petty the criticism has become |
| 18:32 | <annevk_win32> | there was a time when http://flickr.com/photos/novemberborn/4046399/ was the first hit for my name on image search... |
| 18:33 | <annevk_win32> | anyway, time to go |
| 18:56 | <Philip`> | I presume the blogger's identity is narrowed down by either being French, or being an actor capable of sounding very French |
| 19:34 | <Hixie> | so nobody noticed that i added lastweekinhtml5 to the list of ways you can track our community on the wiki? :-) |
| 19:35 | <Hixie> | if the blog was more cleanly set up and more readable, i'd have blogged about it from blog.whatwg.org already |
| 19:35 | <Hixie> | unfortunately while he does link to the irc logs, which is good, he doesn't realy do a good job of highlighting the controversies |
| 19:36 | <Hixie> | it just comes off as a conspiracy nut's blog |
| 19:36 | <Hixie> | it's quite sad really |
| 19:36 | <Hixie> | if it was better layed out it would be much more fun |
| 19:48 | <gsnedders> | Hixie: Can we not just invite him to #whatwg-cabel? |
| 20:15 | <erlehmann> | Hixie: would it make sense to ask for |
| 20:16 | <erlehmann> | a way to mark taglouds up on the mailing list ? |
| 20:16 | <erlehmann> | like @frequency for links or something ? |
| 20:17 | <erlehmann> | (since i'm fairly new, i don't know if it has been discussed and struck down before) |
| 20:25 | <gsnedders> | erlehmann: Nested <em> has always been the conclusion |
| 20:26 | <Hixie> | nested <em> seems dubious to me, but i don't think we'll add a formal way to do tag clouds |
| 20:26 | <Hixie> | feel free to ask if you want though |
| 20:27 | <erlehmann> | gsnedders: nested em ? why would we neet <strong> then ? also, how semantic is that ? |
| 20:27 | <erlehmann> | need |
| 20:27 | <gsnedders> | erlehmann: strong is importance, em is emphasize per HTML 5 |
| 20:28 | <Hixie> | em is stress |
| 20:28 | <Hixie> | i don't think a tag cloud is strength |
| 20:28 | <Hixie> | is stress even |
| 20:28 | <Hixie> | or strength for that matter |
| 20:56 | <webben> | erlehmann: I think http://24ways.org/2006/marking-up-a-tag-cloud works reasonably well. |
| 21:02 | <Philip`> | gsnedders: We have a WHATWG cable channel? |
| 21:03 | <gsnedders> | Philip`: I didn't say cable. I don't expect you to read though :) |
| 21:04 | <Philip`> | gsnedders: You didn't say cabal either :-p |
| 21:04 | <gsnedders> | Philip`: No, as that's what everyone thinks exists. We actually called it something else, just to mislead people :) |
| 21:59 | <Hixie> | is there a unix command line program that takes stdin input, and if it is greater than zero bytes long, outputs a preset line of text followed by stdin? |
| 21:59 | <Hixie> | e.g.: expensive-command-line-that-may-output-warnings | echo-and-cat-if-non-blank "Warnings:" |
| 22:09 | <Hixie> | http://software.hixie.ch/utilities/js/live-dom-viewer/?%3C!DOCTYPE%20html%3E...%3Cform%3E%3Cselect%20name%3Da%20multiple%3E%3Coption%3Ea%3Coption%3Eb%3C%2Fselect%3E%3Cinput%20type%3Drange%20oninput%3D%22a.selectedIndex%20%3D%20value%22%20min%3D-1%20max%3D2%3E%3C%2Fform%3E |
| 22:11 | <zcorpan_> | Hixie: s/user should/user agent should/g |
| 23:19 | <erlehmann__> | Hixie: i'd apuse grep for that |
| 23:19 | <erlehmann__> | abuse |
| 23:20 | <Hixie> | how? |
| 23:23 | Philip` | would use Perl |
| 23:23 | <Philip`> | perl -lpe'print "Warnings:" if $. == 1' or something |
| 23:24 | <Philip`> | but that's just because I'm too lazy to learn Bash |
| 23:27 | <zcorpan_> | [_] I accept the _terms_* |
| 23:27 | <zcorpan_> | how do you mark that up? |
| 23:30 | <Hixie> | hm, |perl -lpe'print "Warnings:" if $. == 1'| is good. |
| 23:30 | <Hixie> | i didn't think of using $. |
| 23:32 | <Philip`> | Of course you should golf it to be as short as possible, because anything else would be un-Perlish |
| 23:33 | <Philip`> | perl -lpe'$.-1||print"Warnings:"' is the obvious thing to do, but I can't think of how to improve on that :-( |
| 23:33 | <Hixie> | your definition of "improve" needs work. :-P |
| 23:34 | <Hixie> | perl -lpe'$1-1||$_.="Warnings:\n" is even more obscure |
| 23:34 | <Hixie> | er |
| 23:34 | <Hixie> | that puts it on the wrong end |
| 23:34 | <Hixie> | perl -lpe'$1-1||$_="Warnings:\n$_" maybe |
| 23:34 | <Philip`> | The rules don't allow to prouce incorrect output :-p |
| 23:34 | <Philip`> | Uh |
| 23:34 | <Philip`> | The rules don't allow you to produce incorrect output :-p |
| 23:35 | <Philip`> | (It'd be better if I didn't forget to type so many letters) |
| 23:35 | <Hixie> | perl -lpe'$.-1||$_="Warnings:\n$_" even |
| 23:35 | <Hixie> | see this is why one doesn't make things so hard to read! |
| 23:38 | <Philip`> | If you want hard to read, you'd have to do something more like perl -pe'@.[1]="Warnings:\n";$_=$.[$.].$_' |
| 23:39 | <Philip`> | Uh, that's a bit silly, I should have said perl -pe'$.[1]="Warnings:\n";$_=$.[$.].$_' though it doesn't make any difference in practice |
| 23:40 | <Philip`> | The @ looks prettier anyway |
| 23:40 | <Hixie> | wtf is @. |
| 23:40 | <Philip`> | It's just a variable |
| 23:40 | <Hixie> | oh it's not a magical one? |
| 23:41 | <Philip`> | No, or if it's magic then I'm just clobbering whatever magic value it contained |
| 23:41 | <Hixie> | dude. don't use punctuation as the name of your own variables. :-P |
| 23:41 | <Philip`> | perldoc doesn't say it's magic |
| 23:41 | <Hixie> | it's bad enough that i have to deal with perl's own magical |
| 23:41 | <Hixie> | variables |
| 23:41 | <Philip`> | I just wanted an excuse to write $.[$.].$_ |
| 23:42 | <Philip`> | It has a pleasing symmetry |
| 23:42 | <zcorpan_> | looks like a weird smiley |
| 23:42 | <zcorpan_> | $_[^_^]_$ |
| 23:43 | <Philip`> | zcorpan_: That looks like a pretty nasty ear infection |
| 23:43 | <erlehmann__> | is that valid perl code ? |
| 23:44 | <Philip`> | erlehmann__: $_[^_^]_$ isn't |
| 23:44 | <Hixie> | m$_[^_^]_$ is |
| 23:44 | <Hixie> | :-D |
| 23:44 | <erlehmann__> | D:< |
| 23:45 | <Philip`> | #$_[^_^]_$ is too - it's easy if you're allowed to modify the code arbitrarily :-p |
| 23:45 | <Hixie> | it is an expression that returns true if $_ contains the string "_x_" for any value of x that isn't "_" or "^". |
| 23:46 | <zcorpan_> | wow |
| 23:46 | <Hixie> | note that the occurance of $_ in the expression has nothing to do with the variable $_ though |
| 23:47 | <zcorpan_> | you should go through a list of smileys and try to interpret them as perl |
| 23:48 | <Philip`> | No language is as fun as Perl |
| 23:49 | <erlehmann> | #_[^_^]_# is too ? |
| 23:49 | <Hixie> | # starts a comment |
| 23:49 | <Hixie> | so anything starting with # is valid |
| 23:49 | <Hixie> | (except if the # is preceeded by an m or an s or a tr) |
| 23:49 | <Hixie> | (but that's another story) |
| 23:51 | <Philip`> | s#\##//##\\##/#s |
| 23:51 | <Philip`> | replaces Perl/sh/etc-style comments with C++-style comments |
| 23:53 | <Hixie> | s#\##//##\\##/#s is really "s#\##//#" with a comment |
| 23:53 | <Hixie> | i don't think relying on comments is really fair game |
| 23:53 | <Philip`> | But I want symmetry :-( |
| 23:53 | <Hixie> | (unless you're writing a polyglot in which case at the commented portion should be no a comment in some language) |
| 23:56 | <Philip`> | On that subject, and just out of interest: is there a nicer/conciser way than |
| 23:57 | <Philip`> | ''/* |
| 23:57 | <Philip`> | document.write("VB") |
| 23:57 | <Philip`> | '*/document.write("JS") |
| 23:57 | <Philip`> | to write a script that can tell whether it's being run as VBScript or JScript? |
| 23:57 | <Hixie> | nicer or conciser? |
| 23:57 | <Philip`> | Conciseness is nice, so both :-) |
| 23:58 | <sicking> | Hixie, got a sec re importScripts? |
| 23:58 | <Hixie> | Philip`: i don't know VB well but i imagine the document. part can be abstracted out |
| 23:58 | <Hixie> | sicking: sure |
| 23:58 | <Hixie> | (i am deep in wf2 edits right now, but i can discuss whatever) |
