/irc-logs / w3c / #html-wg / 2007-12-19 / end

Options:

# Session Start: Wed Dec 19 00:00:01 2007
# Session Ident: #html-wg
# [00:05] * Quits: aaronlev (chatzilla@209.6.168.245) (Ping timeout)
# [00:05] * Quits: Sander (svl@86.87.68.167) (Quit: And back he spurred like a madman, shrieking a curse to the sky.)
# [00:12] * Disconnected
# [00:12] * Attempting to rejoin channel #html-wg
# [00:12] * Rejoined channel #html-wg
# [00:12] * Topic is 'HTML WG telcon Thu 20 Dec 4p Pacific (2007-12-21T00:00UTC) http://www.w3.org/html/wg/tracker/agenda (logs: http://krijnhoetmer.nl/irc-logs/ ) '
# [00:12] * Set by DanC on Tue Dec 18 22:48:23
# [00:12] * Quits: krijnh (krijnhoetm@213.84.148.98) (Connection reset by peer)
# [00:15] * Quits: kingryan (kingryan@66.92.219.50) (Quit: kingryan)
# [00:16] * Joins: sbuluf (glyv@200.49.132.85)
# [00:18] * Quits: tH (Rob@87.102.85.140) (Quit: ChatZilla 0.9.79-rdmsoft [XULRunner 1.8.0.9/2006120508])
# [00:36] * Joins: gsnedders (gsnedders@86.135.224.200)
# [00:41] * Disconnected
# [00:42] * Attempting to rejoin channel #html-wg
# [00:42] * Rejoined channel #html-wg
# [00:42] * Topic is 'HTML WG telcon Thu 20 Dec 4p Pacific (2007-12-21T00:00UTC) http://www.w3.org/html/wg/tracker/agenda (logs: http://krijnhoetmer.nl/irc-logs/ ) '
# [00:42] * Set by DanC on Tue Dec 18 22:48:23
# [00:42] * Quits: krijn (krijnhoetm@213.84.148.98) (Ping timeout)
# [00:42] * Philip keeps getting slightly depressed by the XML encoding of X3D
# [00:44] * Joins: Lachy_ (Lachlan@84.215.9.100)
# [00:44] <Philip> but I'm too lazy/incapable for X3D5 so I'll just continue to ignore interoperability and do what makes sense
# [00:44] <Philip> (since I assume other implementors have similarly chosen to do what makes sense, so accidental interoperability will occur)
# [00:44] * Quits: Lachy_ (Lachlan@84.215.9.100) (Quit: Leaving)
# [00:44] * Joins: Lachy_ (Lachlan@84.215.9.100)
# [00:46] * Quits: Lachy (Lachlan@84.215.9.100) (Ping timeout)
# [00:50] * Joins: kingryan (kingryan@66.92.219.50)
# [00:53] * Quits: jmb (jmb@152.78.68.189) (Ping timeout)
# [00:57] * Joins: olivier (ot@128.30.52.30)
# [00:57] * Joins: jmb (jmb@152.78.68.189)
# [01:01] <anne> we seem to be doing well in so far not doing what the TAG suggests: http://www.w3.org/QA/2007/12/version_identifiers_reconsider.html
# [01:01] <anne> (same goes for CSS, etc.)
# [01:01] * anne wonders how that's explained
# [01:02] <anne> ah
# [01:02] <anne> further on it suggests something different
# [01:02] <anne> "If a language or data format will change in incompatible ways, then indicate the language version used for each instance."
# [01:02] * DanC heads out to family time...
# [01:03] <jgraham_> Nice a firm date for publication. Thanks DanC
# [01:03] <Philip> Easy solution: don't change in incompatible ways?
# [01:03] <anne> yeah
# [01:06] <anne> so it seems we're actually aligned with the TAG now
# [01:06] <anne> on this, hmm
# [01:06] <jgraham_> I like "As long as such documents are likely to have the same or sufficiently compatible meanings per the different versions, then it may be better to omit any indication of version in the instance, and leave it to the receiving software to decide whether the document can be processed"
# [01:07] <jgraham_> Since it conveys a positive advantage to not having versioning when you do have compat.
# [01:08] <Lachy_> oh, the date is nice, but DanC didn't outline any list of requirements that need to be met for the spec to be published, as he was supposed to
# [01:08] <Hixie> i guess that means the spec already meets all requirements
# [01:08] <Lachy_> so then what's stopping us publishing now?
# [01:10] <anne> the action item summary seems to suggest patent review
# [01:10] <jgraham_> The date is the important thing, I think. Because we have to publish by that date whatever state the spec is in :)
# [01:10] <anne> it's not really clear to me what that means though
# [01:21] * Quits: hober (ted@68.101.220.172) (Client exited)
# [01:21] * Quits: billmason (billmason@69.30.57.156) (Quit: .)
# [01:24] * Joins: hober (ted@68.101.220.172)
# [01:24] * Quits: hober (ted@68.101.220.172) (Quit: ERC Version 5.3 (devel) (IRC client for Emacs))
# [01:37] * Joins: mjs (mjs@64.81.48.145)
# [02:10] * Quits: jgraham_ (james@81.86.217.3) (Quit: This computer has gone to sleep)
# [02:12] <mjs> looks like we have a publication deadline
# [02:12] <mjs> sort of
# [02:13] * Hixie replies to said mail
# [02:17] * Quits: Lachy_ (Lachlan@84.215.9.100) (Connection reset by peer)
# [02:17] * Joins: Lachy_ (Lachlan@84.215.9.100)
# [02:26] * Joins: anne-mac (annevk@82.156.27.18)
# [02:33] <anne> ooh, roc tries to merge flexbox and the advanced layout proposal
# [02:33] <anne> maybe we finally get somewhere with that...
# [02:39] * gsnedders points out to anne he's mixing English and Dutch in his latest tweet
# [02:40] <anne-mac> within the same word even
# [02:40] * gsnedders can't remember what the exact Dutch spelling is anyway
# [02:41] <anne-mac> in nl-NL it would be a lowercase "j"
# [02:41] <gsnedders> ah, I thought that.
# [02:41] <gsnedders> But that's not spelling, really.
# [02:41] <gsnedders> but there again, I didn't specify spelling, just language
# [02:41] <anne-mac> why is it not spelling?
# [02:42] <gsnedders> surely capitalisation is a matter of grammar?
# [02:42] <anne-mac> punctation is part of spelling as far as my teachers were concerned
# [02:42] * gsnedders shrugs
# [02:42] <gsnedders> we're probably both too tired :)
# [02:44] <anne-mac> actually, per the Dutch Wikipedia I'm right
# [02:44] <anne-mac> http://nl.wikipedia.org/wiki/Spelfout
# [02:44] <anne-mac> "verkeerd gebruik van hoofdletters: TV in plaats van tv;"
# [02:45] <gsnedders> That means nothing to me :)
# [02:45] <Philip> Hoofdletters is clearly a postal system based on horses
# [02:45] * gsnedders has always wanted to learn Dutch
# [02:46] <anne-mac> Philip, :)
# [02:46] <gsnedders> Philip: are you sure it isn't the letters having hoofs?
# [02:47] <anne-mac> (it's having "hoofden" which means "heads")
# [02:47] * anne-mac -> bed
# [02:47] * anne too
# [02:47] * gsnedders ought to do that too
# [02:48] * gsnedders almost wrote, "/me ought to follow anne", but then he remembered what he was told last time, "I hope you're not following him to the same bed."
# [02:50] * Quits: anne-mac (annevk@82.156.27.18) (Ping timeout)
# [02:50] <Philip> Hmm, two people going to bed makes it sound like a Recommendation, so I should go too
# [02:51] <gsnedders> I haven't, though.
# [02:51] <gsnedders> I merely said I ought to.
# [02:56] * Joins: Lachy__ (Lachlan@84.215.9.100)
# [02:56] * Quits: Lachy_ (Lachlan@84.215.9.100) (Connection reset by peer)
# [03:02] * Quits: Lachy__ (Lachlan@84.215.9.100) (Ping timeout)
# [03:03] * Quits: gsnedders (gsnedders@86.135.224.200) (Quit: gsnedders)
# [03:05] * Joins: Lachy__ (Lachlan@84.215.9.100)
# [03:10] * Quits: dbaron (dbaron@63.245.220.241) (Quit: 8403864 bytes have been tenured, next gc will be global.)
# [03:10] * Quits: Lachy__ (Lachlan@84.215.9.100) (Quit: Leaving)
# [03:10] * Joins: Lachy (Lachlan@84.215.9.100)
# [03:10] * Quits: kingryan (kingryan@66.92.219.50) (Ping timeout)
# [03:10] * Joins: timbl (timbl@209.6.134.246)
# [03:23] * Quits: Lachy (Lachlan@84.215.9.100) (Ping timeout)
# [03:34] * Disconnected
# [03:39] * Attempting to rejoin channel #html-wg
# [03:39] * Rejoined channel #html-wg
# [03:39] * Topic is 'HTML WG telcon Thu 20 Dec 4p Pacific (2007-12-21T00:00UTC) http://www.w3.org/html/wg/tracker/agenda (logs: http://krijnhoetmer.nl/irc-logs/ ) '
# [03:39] * Set by DanC on Tue Dec 18 22:48:23
# [03:50] * Joins: Lachy (Lachlan@84.215.9.100)
# [03:51] * Quits: timbl (timbl@209.6.134.246) (Quit: timbl)
# [03:53] * Joins: kingryan (kingryan@66.92.2.56)
# [03:54] * Quits: adele (adele@17.255.110.63) (Quit: adele)
# [04:17] * Joins: aaronlev (chatzilla@209.6.168.245)
# [04:31] * marcos wishes the HTML5 draft would be published for Xmas! it would be a nice gift for the web community! ..."on the first day of xmas the W3c gave to me...." :D
# [04:36] <marcos> anyway, it's great we finally have a pub date.
# [04:44] * Joins: adele (adele@67.170.232.64)
# [04:44] * Quits: adele (adele@67.170.232.64) (Client exited)
# [04:45] * Joins: adele (adele@67.170.232.64)
# [05:02] * Quits: heycam (cam@210.84.9.222) (Quit: bye)
# [05:18] * Quits: aaronlev (chatzilla@209.6.168.245) (Ping timeout)
# [06:03] * Joins: aaronlev (chatzilla@209.6.168.245)
# [06:04] * Quits: aaronlev (chatzilla@209.6.168.245) (Quit: ChatZilla 0.9.79 [Firefox 3.0b3pre/2007121805])
# [06:06] * Quits: MikeSmith (MikeSmith@mcclure.w3.org) (Quit: Less talk, more pimp walk.)
# [06:13] * Joins: MikeSmith (MikeSmith@mcclure.w3.org)
# [06:15] * Quits: MikeSmith (MikeSmith@mcclure.w3.org) (Quit: Less talk, more pimp walk.)
# [06:28] * Quits: Lachy (Lachlan@84.215.9.100) (Ping timeout)
# [06:39] * Joins: hober (ted@68.101.220.172)
# [06:47] * Joins: dbaron (dbaron@71.204.145.103)
# [07:00] * Quits: marcos (chatzilla@131.181.148.226) (Client exited)
# [07:12] * Joins: MikeSmith (MikeSmith@mcclure.w3.org)
# [07:55] * Quits: MikeSmith (MikeSmith@mcclure.w3.org) (Quit: Less talk, more pimp walk.)
# [07:55] * Joins: karl (karlcow@128.30.52.30)
# [07:55] <karl> http://lists.w3.org/Archives/Public/www-archive/2007Dec/0072.html
# [07:55] <karl> Publication request for HTML5 specification
# [07:55] * Parts: karl (karlcow@128.30.52.30) (Leaving)
# [08:08] <mjs> wow!
# [08:17] * Quits: olivier (ot@128.30.52.30) (Quit: Leaving)
# [08:41] * Quits: adele (adele@67.170.232.64) (Quit: adele)
# [09:03] * Quits: hober (ted@68.101.220.172) (Quit: ERC Version 5.3 (devel) (IRC client for Emacs))
# [09:05] * Joins: MikeSmith (MikeSmith@mcclure.w3.org)
# [09:07] * Quits: dbaron (dbaron@71.204.145.103) (Quit: 8403864 bytes have been tenured, next gc will be global.)
# [09:10] * Joins: Lachy (Lachlan@84.215.41.149)
# [09:26] * Quits: MikeSmith (MikeSmith@mcclure.w3.org) (Quit: Less talk, more pimp walk.)
# [09:31] * Quits: Lachy (Lachlan@84.215.41.149) (Quit: Leaving)
# [09:32] * Quits: Dashiva (noone@84.48.60.15) (Ping timeout)
# [09:33] * Joins: Dashiva (noone@84.48.60.15)
# [09:33] * Joins: tH_ (Rob@87.102.85.140)
# [09:33] * tH_ is now known as tH
# [09:40] * Joins: Lachy (Lachlan@84.215.41.149)
# [09:55] * Quits: Lachy (Lachlan@84.215.41.149) (Connection reset by peer)
# [10:03] * Joins: MikeSmith (MikeSmith@mcclure.w3.org)
# [10:03] * Joins: Lachy (Lachlan@84.215.41.149)
# [10:09] * Quits: Lachy (Lachlan@84.215.41.149) (Quit: Leaving)
# [10:09] * Joins: Lachy (Lachlan@84.215.41.149)
# [10:16] * Quits: MikeSmith (MikeSmith@mcclure.w3.org) (Quit: Less talk, more pimp walk.)
# [10:16] * Quits: Lachy (Lachlan@84.215.41.149) (Quit: This computer has gone to sleep)
# [10:29] * Joins: ROBOd (robod@89.122.216.38)
# [10:30] * Joins: Lachy (Lachlan@213.236.208.22)
# [10:30] * Quits: Lachy (Lachlan@213.236.208.22) (Client exited)
# [10:30] * Joins: Lachy (Lachlan@213.236.208.22)
# [10:43] * Quits: sbuluf (glyv@200.49.132.85) (Ping timeout)
# [10:47] * Joins: jgraham_ (james@81.86.217.3)
# [10:48] * Joins: sbuluf (uaohlk@200.49.132.85)
# [10:57] <anne> whoa, awesome
# [10:57] <anne> thanks guys!
# [10:59] <mjs> anne: can I ask you for some off-topic Opera help?
# [11:00] <anne> sure
# [11:01] <mjs> anne: This fails in Opera (both 9.5b1 and 9.5a1 on the Mac at least) for some reason: http://webkit.org/perf/sunspider-0.9/sunspider.html
# [11:01] <mjs> anne: when I run the same benchmark from local HTML/JS/CSS files it passes (and Opera does pretty well)
# [11:01] <mjs> anne: but the web version stops mysteriously on one of the tests
# [11:01] <mjs> (works ok in IE, Firefox and Safari)
# [11:02] <wilhelm> Strange. Have you tested it on other platforms? Works fine on Windows and Linux here.
# [11:02] <mjs> wilhelm: you tried actually running the benchmark?
# [11:02] <mjs> I have not tried it on Windows or Linux builds of Opera
# [11:03] <anne> I tried it on Linux and it worked there
# [11:03] <anne> hmm
# [11:03] <wilhelm> Hm. I'll have someone investigate.
# [11:04] <wilhelm> We are studying those tests now. (c;
# [11:04] <mjs> no NaN results?
# [11:05] <mjs> input on how to improve them would be appreciated
# [11:05] <wilhelm> No NaN results when I tested on Linux earlier today, no. I'll test some more.
# [11:08] <mjs> I'm not surprised that Opera and Mozilla people are looking at the tests, I am curious whether anyone at Microsoft is, though
# [11:10] * Quits: Lachy (Lachlan@213.236.208.22) (Quit: Leaving)
# [11:11] <wilhelm> Hm. I get NaN on format-tofte and dna.
# [11:13] <mjs> the Opera build I tried got stuck on crypto-aes
# [11:13] <mjs> some other people get NaNs
# [11:13] <mjs> not sure what is up
# [11:13] <mjs> when I tried it, I didn't see anything in Opera's error console
# [11:13] <mjs> (other than complaining about the CSS hack for IE)
# [11:14] * Joins: Lachy (Lachlan@213.236.208.22)
# [11:15] <wilhelm> I see the same result on a Windows box. NaN on the same tests. Hm.
# [11:15] <wilhelm> Does the script time out?
# [11:16] <mjs> I would not expect it to
# [11:17] <mjs> (like I said, the same test works fine run from local files)
# [11:17] <mjs> anything in your error console?
# [11:18] <wilhelm> No, just the IE hack.
# [11:18] <mjs> if you figure it out and there's a workaround, I'll gladly take patches to make it work in Opera
# [11:18] <mjs> I assume JS parse errors or exceptions would be reported
# [11:18] <wilhelm> Cool. Thanks. Would it be possible for us to get a local copy?
# [11:19] <wilhelm> They should be.
# [11:31] * Joins: zcorpan (zcorpan@88.131.66.80)
# [11:40] <mjs> svn checkout http://svn.webkit.org/repository/webkit/trunk/SunSpider
# [11:40] <mjs> you have to run the make-hosted script there though before the html version will run
# [11:40] <mjs> there's also a command-line version for standalone JS engines
# [11:41] <wilhelm> Thanks. (c:
# [11:44] * Quits: sbuluf (uaohlk@200.49.132.85) (Quit: sbuluf)
# [11:46] * Joins: heycam (cam@203.217.74.244)
# [12:16] * Parts: anne (annevk@82.156.27.18)
# [12:16] * Joins: anne (annevk@82.156.27.18)
# [12:58] * Quits: deltab (deltab@82.36.30.34) (Ping timeout)
# [12:58] * Joins: deltab (deltab@82.36.30.34)
# [13:03] * Quits: Lachy (Lachlan@213.236.208.22) (Quit: Leaving)
# [13:04] * Quits: deltab (deltab@82.36.30.34) (Ping timeout)
# [13:05] * Joins: deltab (deltab@82.36.30.34)
# [13:06] * Joins: Lachy (Lachlan@213.236.208.22)
# [13:29] <beowulf> http://www.w3.org/2007/03/mth/harness # mobile browser testing
# [13:30] <anne> hmm "were aware" -> "ware"
# [13:55] <krijnh> anne: http://fronteers.nl/ ;]
# [13:57] <krijnh> Kritiek welkom
# [14:01] <dedridge> krijnh: interesting contact form, looks cool :)
# [14:01] <krijnh> dedridge: you mean markup or style? :)
# [14:04] * Joins: myakura (myakura@123.224.119.194)
# [14:06] <dedridge> krijnh: no sorry, I meant this page -> http://fronteers.nl/inschrijven markup and style
# [14:06] <krijnh> Ah
# [14:06] <krijnh> That one needs some extra attention :)
# [14:07] <krijnh> Titles on the radio buttons for example
# [14:15] <anne> veel <div> :)
# [14:15] <anne> ziet er wel goed uit
# [14:16] <krijnh> Mjah
# [14:16] <krijnh> Voorzover ik weet is dit het minimum
# [14:16] <anne> submenu mag wel automatisch
# [14:16] <krijnh> Maar ik zal het wel mis hebben :)
# [14:16] <krijnh> Automatisch?
# [14:16] <anne> minimum is wel prima dan
# [14:16] <anne> in plaats van een extra page load
# [14:16] <krijnh> Ah, submenu links
# [14:17] <anne> euh, <table summary="Tabel bevat 4 rijen en 2 kolommen">
# [14:17] <krijnh> Dat moet ik nog uit m'n CMS slopen :]
# [14:18] <krijnh> Done
# [14:18] <krijnh> Meteen ook al m'n andere sites beter
# [14:23] <krijnh> <dl> voor comments is waarschijnlijk ook lelijk..
# [14:47] <anne> hsivonen, maybe you should have an advanced interface for your validator instead of trying to put it all in a single screen?
# [14:48] <anne> and have some kind of tracking going on on what stuff people use so you can use that to tweak things
# [14:50] * Joins: laplink (link@193.157.66.112)
# [14:51] <hsivonen> anne: yeah. I have ideas for an advanced (X)HTML5 interface long ago, but (X)HTML5 is so far from stable that it doesn't make sense to design that interface yet
# [14:52] <hsivonen> (also, I wouldn't object to other people trying out alternative UIs for generating validator parameters and redirecting to validator.nu)
# [14:52] <hsivonen> in fact, there's already an independently developed Validator.nu front end for CML
# [14:55] <anne> wow, out of the blue or they knew you somehow?
# [14:55] <hsivonen> anne: out of the blue but they did ask if it was OK to deploy it
# [14:55] <hsivonen> anne: http://cmlcomp.org/validator/
# [14:57] <anne> heh, cool
# [15:01] * Quits: Yudai (Yudai@59.146.9.126) (Ping timeout)
# [15:13] * Quits: Lachy (Lachlan@213.236.208.22) (Ping timeout)
# [15:18] * Joins: Yudai (Yudai@59.146.9.126)
# [15:22] * Joins: timbl (timbl@128.30.5.98)
# [15:30] * Joins: smedero (smedero@158.130.16.191)
# [15:30] * Quits: jgraham_ (james@81.86.217.3) (Quit: This computer has gone to sleep)
# [15:33] * Joins: matt (matt@128.30.52.30)
# [15:47] * Quits: myakura (myakura@123.224.119.194) (Ping timeout)
# [16:08] * Quits: dedridge (opera@121.72.18.70) (Ping timeout)
# [16:24] * Joins: Lachy (Lachlan@84.215.41.149)
# [16:29] * Quits: Lachy (Lachlan@84.215.41.149) (Quit: Leaving)
# [16:29] * Joins: Lachy (Lachlan@84.215.41.149)
# [16:33] * Joins: billmason (billmason@69.30.57.156)
# [16:40] <hsivonen> Validator.nu now supports XHTML5 + SVG1.1 + MathML 2.0
# [16:40] * Joins: aaronlev (chatzilla@146.115.126.114)
# [16:40] <hsivonen> feedback appreciated
# [16:48] <anne> dare I say, smedero +1
# [16:48] <smedero> thanks anne.
# [16:49] <smedero> i hate to be a list-mom
# [16:49] <smedero> but you know, we'd actually gotten to a fairly decent signal-noise rational on the list
# [16:49] <smedero> and now i feel like it is going south again
# [16:49] <anne> it's certainly going better
# [16:49] <smedero> yeah, overall much better.
# [16:50] * Parts: timbl (timbl@128.30.5.98)
# [16:51] <zcorpan> hsivonen: http://validator.nu/?doc=http%3A%2F%2Fgolem.ph.utexas.edu%2F%7Edistler%2Fblog%2F&schema=http%3A%2F%2Fs.validator.nu%2Fxhtml5-rdf-svg-mathml.rnc+http%3A%2F%2Fs.validator.nu%2Fhtml5%2Fassertions.sch+http%3A%2F%2Fc.validator.nu%2Fall%2F&parser=xml
# [16:51] <zcorpan> hsivonen: source extracts stop working after message 59, it seems
# [16:52] <zcorpan> hsivonen: um, and then work again at message 83
# [16:52] <zcorpan> (but are not correct)
# [16:58] <hsivonen> zcorpan: thanks.
# [16:58] * Joins: gsnedders (gsnedders@86.135.224.200)
# [17:03] <hsivonen> zcorpan: actually, it breaks between messages 5 and 6
# [17:04] <zcorpan> yeah
# [17:09] * Quits: Lachy (Lachlan@84.215.41.149) (Quit: This computer has gone to sleep)
# [17:20] * Joins: Lachy (Lachlan@213.236.208.22)
# [17:22] <Philip> Argh, Opera 9.2's <input type=url> has worse usability than <input type=text>, since it's impossible to type in a prefix of something in the autocomplete list and then press enter without it autocompleting
# [17:27] <krijnh> And it's uber slow
# [17:27] <krijnh> If if you have a lot of stuff in your history
# [17:27] <krijnh> And <input type=email> doesn't work either, cause it inserts readable names in the boxes, not the addresses
# [17:27] <Philip> hsivonen: The Warnings on http://html5.validator.nu/?doc=http://www.microsoft.com&showsource=yes link to #l1c10482 etc, but the source has <b id="cl1c10482">"</b> instead so the link doesn't work
# [17:28] <krijnh> But that's probably fixed in 9.5 :)
# [17:32] <hsivonen> Philip: thanks.
# [17:48] <hsivonen> Philip: hmm. It looks like I broke that when I misfixed a bug reported by zcorpan.
# [17:48] * hsivonen searches irc logs for the case zcorpan reported
# [17:51] <hsivonen> found zcorpan's report
# [17:59] * Joins: timbl_ (timbl@128.30.5.98)
# [18:01] * Joins: adele (adele@67.170.232.64)
# [18:01] <zcorpan> i wonder if it's useful for <script src> to have an empty content model
# [18:02] <zcorpan> i've seen people leave a comment in <script src> that says what the script is for or so
# [18:03] * Joins: Sander (svl@86.87.68.167)
# [18:03] <hsivonen> zcorpan: the microsoft.com comment is not very useful, though. :-)
# [18:03] <hsivonen> /**/
# [18:03] <zcorpan> indeed
# [18:03] <zcorpan> but  is conforming :)
# [18:04] <zcorpan> (in pcdata)
# [18:04] <Philip> <script src="">alert('You\'re using IE')</script>
# [18:09] <zcorpan> Philip: didn't know about that one
# [18:16] * Quits: anne (annevk@82.156.27.18) (Ping timeout)
# [18:18] * Quits: timbl_ (timbl@128.30.5.98) (Quit: timbl_)
# [18:19] * Quits: Lachy (Lachlan@213.236.208.22) (Ping timeout)
# [18:22] * Joins: timbl (timbl@128.30.5.98)
# [18:27] * Quits: adele (adele@67.170.232.64) (Quit: adele)
# [18:36] <hsivonen> heh. the w3.org front page uses lang='en-uk'
# [18:36] <hsivonen> hreflang actually
# [18:37] <hsivonen> Philip, zcorpan: fixed the "cl" vs. "l" issues in source location links
# [18:38] <hsivonen> with the XML parser the source location tracking is still broken
# [18:52] <Philip> hsivonen: When I try to run the validator, it says
# [18:52] <Philip> java.lang.ClassCastException: org.xml.sax.InputSource cannot be cast to nu.validator.xml.TypedInputSource at nu.validator.servlet.VerifierServletTransaction.schemaByUrl(VerifierServletTransaction.java:1037) at nu.validator.servlet.VerifierServletTransaction.<clinit>(VerifierServletTransaction.java:369)
# [18:52] <Philip> just after 'DEBUG nu.validator.servlet.VerifierServletTransaction - Will load schema: http://s.validator.nu/svg11/svg11.rnc'
# [18:52] <hsivonen> Philip: please rerun dldeps
# [18:53] <Philip> Nothing happens when I do that
# [18:53] <hsivonen> Philip: or rather, checkout, then checkout again and then dldeps
# [18:53] <Philip> Things happen when I do that
# [18:54] <hsivonen> Philip: happen as in "work"?
# [18:54] <Philip> Happen as in "probably work in a few minutes when it's finished doing stuff"
# [18:54] <hsivonen> ok. good
# [18:55] <hsivonen> I'd be interested in better ways to self-update the build script
# [18:55] <Philip> hsivonen: It looks happy now - thanks :-)
# [18:56] <Philip> (dldeps didn't do anything, probably because I'd already modified it to use the newer Xalan)
# [18:56] <hsivonen> Philip: it probably made two symlinks
# [18:57] <Philip> Ah, okay
# [19:24] <DanC> teleconference tomorrow or not? any preferences?
# [19:34] * Joins: dbaron (dbaron@63.245.220.241)
# [19:39] <Dashiva> I hope Thierry doesn't get misunderstood when he said "I am resenting this email"
# [19:41] * Quits: Yudai (Yudai@59.146.9.126) (Quit: SIGTERM received; exit)
# [19:41] * Joins: Yudai (Yudai@59.146.9.126)
# [19:45] * Quits: mjs (mjs@64.81.48.145) (Quit: mjs)
# [19:46] * Joins: adele (adele@17.255.110.63)
# [19:58] * Quits: kingryan (kingryan@66.92.2.56) (Quit: kingryan)
# [20:05] * Joins: brianlandau (brianlanda@24.106.184.150)
# [20:20] * Joins: Lachy (Lachlan@84.215.41.149)
# [20:20] * Quits: Lachy (Lachlan@84.215.41.149) (Client exited)
# [20:20] * Joins: Lachy (Lachlan@84.215.41.149)
# [20:26] * Quits: smedero (smedero@158.130.16.191) (Quit: smedero)
# [20:35] * Quits: brianlandau (brianlanda@24.106.184.150) (Quit: brianlandau)
# [20:42] * Joins: jgraham_ (james@81.86.217.3)
# [21:14] * Quits: jgraham_ (james@81.86.217.3) (Quit: Leaving)
# [21:16] * Joins: kingryan (kingryan@66.92.219.50)
# [21:23] * Quits: aaronlev (chatzilla@146.115.126.114) (Ping timeout)
# [21:25] * Joins: hasather (hasather@90.231.107.133)
# [21:28] * Quits: timbl (timbl@128.30.5.98) (Quit: timbl)
# [21:32] * Joins: jgraham_ (james@81.86.217.3)
# [21:36] * Joins: aaronlev (chatzilla@209.6.168.245)
# [21:42] * Joins: timbl_ (timbl@128.30.5.98)
# [21:49] <DanC> interesting... IE8 does Acid2 http://blogs.msdn.com/ie/archive/2007/12/19/internet-explorer-8-and-acid2-a-milestone.aspx
# [21:52] * gsnedders wonders if the IE team have started to get around the issues of being unable to say anything
# [21:52] <zcorpan> "IE8 standards mode"
# [21:52] <zcorpan> hmm
# [21:53] <zcorpan> does acid2 opt in to ie8 standards mode?
# [21:54] * Quits: hasather (hasather@90.231.107.133) (Quit: leaving)
# [21:55] <Hixie> they use perforce? interesting
# [21:57] * Quits: zcorpan (zcorpan@88.131.66.80) (Ping timeout)
# [21:59] <gsnedders> LONGHORN_IE8…
# [21:59] <gsnedders> does that mean it'll be included with Server 2008?
# [22:00] <gsnedders> I guess it also means data URI support
# [22:01] <hsivonen> gsnedders: well, it says //depot/longhorn_ie8/Inetcore/lib/network/dataprot/dataprot.cxx
# [22:01] <gsnedders> ah. I hadn't started looking through files yet :)
# [22:02] * gsnedders wonders how many bugs we can work out they've fixed from that alone
# [22:03] <gsnedders> display: table, actually, is in acid2
# [22:03] * Joins: hasather (hasather@90.231.107.133)
# [22:03] <Hixie> i know they've worked on display:table
# [22:04] <Hixie> for several months they e-mailed me every other week with a display:table question
# [22:04] <gsnedders> :P
# [22:04] <gsnedders> position: fixed is in there too
# [22:06] <ROBOd> hmm... it's "too good to be true", there must be a catch somewhere
# [22:06] <ROBOd> we'll see it when we get ie8
# [22:06] <ROBOd> or ... even worse... after several months of developing for ie8
# [22:06] <gsnedders> ROBOd: why? look at Op6 -> Op7 when they redid the rendering engine from scratch
# [22:06] <gsnedders> ROBOd: it's totally possible that by redoing the layout engine they've got rid most major issues
# [22:07] <ROBOd> gsnedders: not saying it's not possible. but i don't believe (yet) they just decided to give us a free lunch :P
# [22:07] <ROBOd> as in - will we be able to *really* develop *one* site for all? opera, firefox, safari and ie8?
# [22:08] * hsivonen wonders how "scratch" the opera 6 to 7 change was
# [22:08] <gsnedders> hsivonen: huh?
# [22:08] * hsivonen wonders if IE8 has any Tasman bits in it
# [22:08] <Hixie> ROBOd: it's quite possible to pass acid2 and be extremely buggy
# [22:08] <gsnedders> hsivonen: yeah. IE8 is just a need layout engine.
# [22:08] <gsnedders> *just is
# [22:08] * gsnedders headdesks
# [22:08] <ROBOd> i believe ie8 *will* have some "techniques" to constrain devs to work on ie8-only sites
# [22:08] * gsnedders misread Tasman for Trident
# [22:09] <hsivonen> gsnedders: I mean I wonder if Opera 7 reused some bits from Opera 6
# [22:09] <gsnedders> ROBOd: just hope MS management isn't that evil :P
# [22:10] <ROBOd> gsnedders: they proved, for me, they are
# [22:10] * Joins: mjs (mjs@17.255.106.224)
# [22:11] * gsnedders thinks the only major thing they've done regarding IE/Win is stopping dev for a few years
# [22:11] <ROBOd> gsnedders: yep, true. that's the worst thing they did
# [22:11] <gsnedders> I doubt IE8 will have any new proprietary stuff in it
# [22:12] <ROBOd> the second worst thing is they released ie7 as-is: we have now second "layer" for patching layouts for IE
# [22:12] <ROBOd> we now use if IE 7, and if IE 6 conditional comments (i do)
# [22:12] <gsnedders> ROBOd: so what would you've done? just waited out for the new layout engine for IE7?
# [22:13] <ROBOd> gsnedders: ... maybe that would've been better. they made a mistake with releasing ie7 as-is. ie7 breaks old pages, and it's still not up-to-par with modern UAs
# [22:13] <ROBOd> so we didn't get anything. everybody lost
# [22:13] <ROBOd> they lost compatibility with older pages, and we have another headache
# [22:14] * Joins: anne (annevk@82.156.27.18)
# [22:14] * gsnedders didn't have anything break in IE7 :P
# [22:14] <ROBOd> gsnedders: i did. not only once
# [22:15] <ROBOd> gsnedders: and it breaks differently in ie7 versus ie6. you need different "patching"
# [22:15] <hsivonen> ROBOd: by working around IE6 brokenness you are prolonging your need to work around IE6 brokenness
# [22:15] <ROBOd> hopefully ie6 will be made obsolete soon by users
# [22:15] <anne> Hixie, time to finish Acid3 methinks
# [22:16] <gsnedders> display: run-in anyone? :P
# [22:16] <hsivonen> Likewise, I'd like to see X-philes who were devout enough to go application/xhtml+xml only
# [22:16] <Hixie> anne: send me things to test :-)
# [22:16] <ROBOd> hsivonen: true. however, i've made my decision. as long as a sizeable number of users come with ie6, i have to make some of my sites work with ie6
# [22:17] <ROBOd> hsivonen: it's not up to me. i could turn away all my ie6 visitors ... but that's not helping me much
# [22:17] <ROBOd> if it were up to me: everyone should stop coding for ie6 and ie7 now. go on strike. :)
# [22:18] * Quits: aaronlev (chatzilla@209.6.168.245) (Quit: ChatZilla 0.9.79 [Firefox 3.0b3pre/2007121805])
# [22:19] <ROBOd> anyway, i'm not saying this is not good news from microsoft regarding ie8
# [22:19] <hsivonen> the X-philes should go application/xhtml+xml-only and publish killer XHTML+SVG content
# [22:19] * Joins: ChrisWilson (cwilso@131.107.0.72)
# [22:19] <ROBOd> congrats to microsoft - the blog post made me "happy"
# [22:19] <ROBOd> but i remain skeptical
# [22:20] <shepazu> hi, ChrisWilson, congrats on Acid2
# [22:21] <ChrisWilson> howdy.
# [22:21] <ChrisWilson> thanks.
# [22:21] <ChrisWilson> skeptical on what ROBOd?
# [22:21] <ROBOd> ChrisWilson: at the milestone. i just presented my opinion previously, just before you came in
# [22:21] * Joins: aaronlev (chatzilla@209.6.168.245)
# [22:22] <ChrisWilson> Ah. Sorry I missed it.
# [22:22] <gsnedders> ChrisWilson: congrats, now you've walked in just into this :)
# [22:22] <ROBOd> ChrisWilson: beware ... conspiracy theory ahead :) - i know you do not like this stuff
# [22:22] <Hixie> ChrisWilson: my offer to add whatever you think should be added to acid3 is still open, btw, if you have specific things you'd like me to test. non-rendering, dom-related things in particular.
# [22:22] <shepazu> geez, guys, give it a rest
# [22:22] <ChrisWilson> gsnedders: yes, well, been stepping in things all week.
# [22:23] <ROBOd> by the way microsoft has acted in the past years ... i don't really believe the "hype" around ie8. yes, this is a great milestone, i'm looking forward to it. however, i believe microsoft will introduce some constrain to keep devs coding for ie8 only
# [22:23] <gsnedders> ChrisWilson: Could you (or someone else in the IE team) post a list of things that are fixed leading to Acid2 passing? Seeming you've now announced it I guess you could less it
# [22:23] * shepazu sighs
# [22:23] <gsnedders> s/less/post/
# [22:24] <ROBOd> shepazu: they didn't give us a rest for several years ;)
# [22:24] <ChrisWilson> gsnedders: you mean, could we post the list of features we had to add, or bugs we needed to fix, in order to get Acid2 rendering properly?
# [22:24] <gsnedders> ChrisWilson: yeah
# [22:24] <gsnedders> (both)
# [22:24] <ROBOd> ChrisWilson: i'd say post both
# [22:24] <Lachy> ChrisWilson, what's the new trigger for IE8 standards mode?
# [22:25] <ChrisWilson> (sorry, that should have been an and, not an or)
# [22:25] <ROBOd> now i have to go to sleep guys
# [22:25] <ChrisWilson> Hmm, good idea, I'll suggest it. We certainly have that, at a high level - in fact, Markus brings up his deconstruction doc during the video.
# [22:25] <ROBOd> good night to all
# [22:25] * gsnedders had issues getting something to play the WMV
# [22:25] <ChrisWilson> ROBOd, if you haven't left yet, one thing:
# [22:26] <ROBOd> yes
# [22:26] <ChrisWilson> Oh, nevermind. I didn't understand what you meant by "constrain to keep devs coding for IE8 only", but on third read I do. Oh well, time will tell.
# [22:27] <ROBOd> i'm not yelling, or some such
# [22:27] <ChrisWilson> Hixie: I think that we have a lot of edge cases for better interop on rendering now, actually. :) But then we're starting to contribute to the CSS2.1 Test Suite IIUC.
# [22:27] * Lachy wonders if ChrisWilson can/will answer his question about the standards mode trigger
# [22:28] <gsnedders> is that (standards mode) even decided yet?
# [22:28] <ROBOd> i'm not trying to be rude. i'm just making my point. it's built after years of frustration with ie. i hope this is a true milestone, not something we'll feel frustrated to code for.
# [22:28] <Hixie> ChrisWilson: yeah, i hope rendering will be mostly catered for by the CSS2.1 test suite
# [22:28] * Joins: zcorpan (zcorpan@83.227.33.203)
# [22:28] <Hixie> ChrisWilson: hence my focus on the DOM for acid3, which hasn't had as much attention
# [22:30] <ChrisWilson> Lachy: sorry, splitting attention in multiple directions. The standards mode trigger is likely a META tag; I can't remember exactly what we landed on. We'd been getting advice from a number of web developers privately on what they thought would work best in their workflow.
# [22:31] <Lachy> oh no, not a meta element! :-(
# [22:31] <gsnedders> it's hard to say what would work, though, sadly
# [22:31] <ChrisWilson> yeah, I know. It wasn't my choice either.
# [22:32] <gsnedders> ChrisWilson: what happens when we have quirks mode DOCTYPE and this meta tag?
# [22:32] <Lachy> can you or someone else blog about that on the IE blog and get some feedback from web devs?
# [22:32] <wilhelm> Legacy content will be rendered like in IE7?
# [22:33] <ChrisWilson> good question, and will be answered when the opt-in is finalized and published.
# [22:33] <ChrisWilson> (that was re:gsnedders)
# [22:33] * Quits: matt (matt@128.30.52.30) (Quit: matt)
# [22:33] <ChrisWilson> Lachy:yes.
# [22:33] * Quits: ROBOd (robod@89.122.216.38) (Quit: good night)
# [22:33] <ChrisWilson> wilhem: yes.
# [22:34] <Lachy> ChrisWilson, shouldn't you get feedback about the possible opt-in *before* you make a final decision and ship IE, so we don't just suddenly get stuck with whatever decision (good or bad) you guys make?
# [22:34] <hsivonen> ChrisWilson: was using the HTML5 doctype as the opt-in unacceptable considering Web dev workflows?
# [22:34] <shepazu> Lachy: I guess you mean feedback by the HTML WG? he just said they've been asking devs...
# [22:34] <ChrisWilson> Lachy, as I said earlier, we HAVE been getting feedback about the opt-in from web developers - just from a more limited set of web developers, and privately.
# [22:35] <Lachy> I mean *public* feedback from everyone, but was mostly talking about web developers
# [22:35] <ChrisWilson> hsivonen: actually, I expect the HTML5 doctype, whatever it might be, to automatically opt in to IE8 standards mode, considering it's an "unknown doctype".
# [22:35] <ChrisWilson> I wouldn't advise webdevs to write HTML5 content today, though.
# [22:36] <ChrisWilson> but I do (obviously, because I'm not stupid) want IE8 to use IE8 standards mode for HTML5 content in the future. I can read an adoption graph. :)
# [22:36] <hsivonen> ChrisWilson: Interesting. (especially the part about unknowns getting IE8 standards)
# [22:36] * Quits: aaronlev (chatzilla@209.6.168.245) (Quit: ChatZilla 0.9.79 [Firefox 3.0b3pre/2007121905])
# [22:37] <ChrisWilson> Sorry, that looks a little harsh in print, and I didn't mean it too. I picked up a cold yesterday, and have been swamped, so I'll plead temporary grumpiness. Sorry about that.
# [22:37] <Lachy> unknown doctypes always need to trigger the latest standards mode. if they don't, then it prevents and future defined doctype from triggering a future standards mode
# [22:37] <ChrisWilson> I meant: we do certainly have the goal of interoperating with our best standards behavior for HTML5 content in the future.
# [22:38] <ChrisWilson> Lachy: yes, that's the argument I've been using.
# [22:38] * Joins: aaronlev (chatzilla@209.6.168.245)
# [22:38] * Lachy is surprised he and ChrisWilson actually agree on something :-)
# [22:39] <ChrisWilson> well, you knew it had to happen sometime.
# [22:42] <gsnedders> wow. the insides of the building look really horrible :P
# [22:43] <ChrisWilson> Hah!
# [22:43] <gsnedders> reminds me of the lower half of my school, just with doors closer together :)
# [22:43] <ChrisWilson> Not really. It is a large building, though, and lacks much personality - but on the other hand, I think our last building had a little too much personality, so I'm not complaining.
# [22:44] <ChrisWilson> heheh
# [22:44] <gsnedders> But I guess you can't get 30 people in a room :)
# [22:45] <gsnedders> ChrisWilson: do you foresee any further bug modes in the future?
# [22:47] * Joins: dedridge (opera@121.72.47.224)
# [22:47] <gsnedders> nice to see in the video it's the real acid2 test with data URI scheme :)
# [22:48] * Quits: aaronlev (chatzilla@209.6.168.245) (Ping timeout)
# [22:49] <gsnedders> ChrisWilson: How much work went into acid2 outwith of the normal development of IE8?
# [22:50] * Joins: aaronlev (chatzilla@209.6.168.245)
# [22:51] <zcorpan> hmm, is it possible to view the video without silverlight?
# [22:51] <gsnedders> zcorpan: click download, gives WMV though :(
# [22:52] <zcorpan> aha
# [22:53] <gsnedders> ChrisWilson: I see a list of things in the video, but too low quality really
# [22:57] <ChrisWilson> (sorry, dean walked into my office for a while)
# [22:57] <gsnedders> silly dean.
# [22:57] <gsnedders> :P
# [22:58] <ChrisWilson> you'd think he'd learn, but no.
# [22:58] * Quits: zcorpan (zcorpan@83.227.33.203) (Ping timeout)
# [22:58] <ChrisWilson> As per a comment in the blog, a few of the big ticket items: Data URL - Now implemented
# [22:58] <ChrisWilson> CSS Display Table type properties - fixed/implemented
# [22:58] <ChrisWilson> CSS Relative positioning - fixed (shrink to fit)
# [22:58] <ChrisWilson> HTML closing p tag - ??
# [22:58] <ChrisWilson> Object fallback now co?
# [22:58] <ChrisWilson> HTML abbr tag now fixed
# [22:58] <ChrisWilson> CSS generated content - fixed
# [22:58] <gsnedders> object fallback being what?
# [22:58] <gsnedders> what about <object> padding?
# [22:58] <ChrisWilson> wow, that was cute. Didn't realize paste a double CR would send lines.
# [22:59] <ChrisWilson> yeah, that's not really "object padding" - it's the fact that we weren't treating <object src="image.png"> exactly the same as an image.
# [23:00] <ChrisWilson> we were essentially just embedding ourselves (since we know how to render images), and then there was padding on the page that hosted the image. Goofy, really, but there you go.
# [23:00] <gsnedders> ah
# [23:00] * Joins: zcorpan (zcorpan@83.227.33.203)
# [23:00] <ChrisWilson> object fallback - there's a whole section of the acid2 guide about it. You need to deal with timeouts correctly, etc.
# [23:01] <zcorpan> ChrisWilson: "HTML closing p tag" sounds scary. what does it mean?
# [23:01] <ChrisWilson> (popping up a level - if an <object> can't be rendered for some reason, it's supposed to fall back to the enclosed contents. Acid2 has nested objects.
# [23:01] <gsnedders> didn't you support that?
# [23:01] * gsnedders thought you did
# [23:02] <ChrisWilson> we did - we didn't support timeouts, though.
# [23:02] <gsnedders> with hacks to avoid multiple self-embedding?
# [23:02] <zcorpan> ChrisWilson: is it "<p><table>" where <table> closes the P ?
# [23:02] <ChrisWilson> (and actually, IE7 added some support - we did fairly poorly in IE6, IIRC)
# [23:02] <ChrisWilson> zcorpan - probably. I remember it was an edge case.
# [23:02] <ChrisWilson> gsnedders: yes.
# [23:03] <zcorpan> ChrisWilson: ok
# [23:04] <ChrisWilson> brb
# [23:04] * gsnedders blames Dean
# [23:05] <hsivonen> so thanks to Acid2, we now have a structured inline text/html compat issue not only with <p><ol> but also with <p><table>...
# [23:05] <Hixie> heh
# [23:06] <Hixie> you can't blame me for writing tests that check for standards compliance, come now :-P
# [23:06] * gsnedders blames Hixie
# [23:06] <gsnedders> ChrisWilson: having now watched the video fully, how do HTML parsing changes work when you only switch into IE8 standards mode when you find a <meta> tag?
# [23:07] * mjs wonders if unknown elements are handled sanely in the new mode
# [23:07] * Quits: timbl_ (timbl@128.30.5.98) (Quit: timbl_)
# [23:11] <ChrisWilson> gsnedders: restart if necessary. Most of the HTML parsing changes wouldn't take effect by the time you hit a META tag (e.g. the </p> issue) even if you're putting it in the end of the head.
# [23:12] * ChrisWilson wonders what mjs considers sane.
# [23:12] <gsnedders> ChrisWilson: what if I put the meta tag at the end of the document?
# [23:12] <Hixie> sane = same parse model as <span>
# [23:12] <Hixie> at least, same parse model as html5 says you should use for <span>. :-)
# [23:12] <gsnedders> ChrisWilson: sorry, I do think of mad things :)
# [23:12] <mjs> ChrisWilson: something other than inserting empty elements with tag names of "foo" and "/foo"
# [23:13] <ChrisWilson> gsnedders: oh, we went through all of them.
# [23:13] <gsnedders> ChrisWilson: huh?
# [23:13] <gsnedders> ChrisWilson: my point is won't it need a total re-parse then?
# [23:13] <ChrisWilson> (all of the mad things people could do with meta, with PI, with profile, with conditional comment-like syntax)
# [23:13] <gsnedders> ah
# [23:14] <ChrisWilson> gsnedders: if there's something that could have taken effect, yes; but you won't have parsed most of the document yet.
# [23:14] <gsnedders> ChrisWilson: my point is if you put the meta tag at the end of the document, you'll have to re-parse it all, no?
# [23:15] <hsivonen> ChrisWilson: what if you have a script before the meta and the script has side effercts?
# [23:15] <hsivonen> effects
# [23:15] <gsnedders> what about other parsing changes that mean the meta tag isn't found in IE8 standards mode but is found in standards mode?
# [23:15] <gsnedders> (or do they not exist?)
# [23:15] * gsnedders should spend less time trying to break stuff :)
# [23:15] <ChrisWilson> Hixie/mjs: hmm. I doubt we've changed it so far, no; I'll see if we can. It was actually QUITE sane at the time - there were a number of pages that had tons of junk in it that ended up blowing a parsing stack.
# [23:16] <Hixie> yeah, i bet
# [23:16] <Hixie> one of my favourite pages that crashed my own html parser when i was writing the spec for html5 was a page with an infinite number of <p><font><strong> open tags
# [23:17] <gsnedders> Hixie: s/infinite/huge/ I assume?
# [23:17] <Hixie> nope, actually infinite
# [23:17] <Hixie> at least as far as i could tell
# [23:17] <gsnedders> how on earth do you do that in HTML!?
# [23:17] <Hixie> the server never stopped returning data
# [23:17] <ChrisWilson> if it crashed before ending, does it matter? ;)
# [23:18] <Hixie> ChrisWilson: yeah my solution was to just cap it like everyone else does ;-)
# [23:18] <gsnedders> I guess if the HTML file is infinite is size…
# [23:19] <ChrisWilson> My favorite was the pages with large numbers of nested unclosed <body> tags with progressive background= attributes for a fade-in effect.
# [23:19] <Hixie> hah
# [23:19] <Hixie> nice
# [23:19] * gsnedders collapses laughing
# [23:19] <gsnedders> I really ought to finish my homework.
# [23:21] <gsnedders> ChrisWilson: nice to see you can talk some more about IE8, though, too :)
# [23:23] * gsnedders wonders what happened at MS suddenly :P
# [23:23] <ChrisWilson> I tried convince Dean to title his post "IE team drops acid" but he wouldn't go for it.
# [23:24] <gsnedders> hmm… wouldn't "takes" be more appropriate?
# [23:25] <ChrisWilson> yeah, probably.
# [23:25] <ChrisWilson> takes the red pill?
# [23:25] <gsnedders> No sort of food is naturally blue, though.
# [23:26] * gsnedders goes back to proof-reading homework (on the waterfall development model — taught as if it is the only one)
# [23:26] <gsnedders> How widely used is the waterfall model anyway (i.e., how irrelevant is what I'm being taught)?
# [23:27] <ChrisWilson> some elements of the waterfall model are very prevalent.
# [23:27] <ChrisWilson> But as pure model? Probably about as relevant as any other theoretical model.
# [23:28] * gsnedders wonders whether to scare ChrisWilson will other things in the course
# [23:28] <gsnedders> (which are so untrue it's sad)
# [23:30] <gsnedders> e.g., Unicode is 16-bit, US-ASCII is 8-bit…
# [23:30] <ChrisWilson> heh.
# [23:30] <ChrisWilson> That's okay, my first programming class in college was taught in Pascal. I have yet to ever run across anywhere that writes software in Pascal.
# [23:31] <gsnedders> ChrisWilson: come across BASIC anywhere nowadays?
# [23:31] <Hixie> i grew up on pascal. it's a nice language, though as you say, not used much in the wild.
# [23:31] <ChrisWilson> When I was a junior, the CS college migrated all classes over to be taught in Scheme. They thought this was an improvement.
# [23:31] <gsnedders> ChrisWilson: (I'm in S5 — equiv. of second last year of high school)
# [23:31] <ChrisWilson> It's not BAD; it's just harder to use than BASIC but a lot more limited than C.
# [23:32] <hsivonen> the worst thing about pascal is that some otherwise useful computer science literature uses pascal for algorithms and leaves it to the reader to convert to sane array indexing
# [23:32] <ChrisWilson> heheheh. Exactly.
# [23:32] <gsnedders> I was trying to do thing with ASCII codes, which I know in hex through unicode, but the type of BASIC we're using doesn't support hex at all :\
# [23:32] <gsnedders> what's the array indexing like in Pascal?
# [23:32] <Hixie> hsivonen: what's special about pascal array indexing? (or did i use a non-standard version of pascal?)
# [23:33] <ChrisWilson> ...AAAAAAAnd this is why I think the first round of computer classes should be taught in C. zero-based indexing, sane pointers, and plenty of rope to hang yourself with.
# [23:33] <ChrisWilson> pascal IIRC indexes arrays starting with 1.
# [23:33] <ChrisWilson> arr[1] is the first item.
# [23:33] <hsivonen> Hixie: what ChrisWilson said
# [23:33] <gsnedders> TrueBASIC does too. My computing teacher called C silly and confusing and questioned whether people would understand.
# [23:34] <gsnedders> (in terms of zero based indexing)
# [23:34] <Hixie> huh. all pascal i've ever used uses zero-based indices.
# [23:34] <mjs> pascal lets you choose the indexing
# [23:34] <Hixie> really?
# [23:34] <mjs> you can declare an array with 4-based indexing if you want
# [23:34] <mjs> or 73-based
# [23:34] <Hixie> oh
# [23:34] <Hixie> you mean you can set the range of the index
# [23:34] <Hixie> yes
# [23:34] <Hixie> sure
# [23:35] <gsnedders> ChrisWilson: my computing teacher thinks that if I tried to teach the class how I would they'd be confused quickly: one of the other people in the class thinks they'd be fine if their English is good enough :)
# [23:35] <Hixie> type foo: array[23..29] of Char;
# [23:35] <Hixie> very helpful for certain algorithms
# [23:35] * gsnedders is confused by that
# [23:35] <Hixie> but everyone does [0..n] in real code
# [23:35] <ChrisWilson> oooo, yeah, I forgot all about that.
# [23:36] * Hixie did all his win32 programming in delphi, which is pascal with classes and many other things that make it way better than regular pascal
# [23:36] <ChrisWilson> Maybe that was it - my teachers and books all suggested doing type foo: array[1..13] of Char;
# [23:36] <hsivonen> Hixie: not some books on algorithms :-(
# [23:37] <Hixie> hsivonen, ChrisWilson: sorry to hear that
# [23:37] * ChrisWilson did all my win32 programming in C and C++. With occasional assembler for device drivers. :)
# [23:37] * gsnedders has never done anything with any win32 APIs, and never anything specifically targeting Windows apart from rudimentary BASIC at school
# [23:38] <ChrisWilson> hmm. "IE8 passes ACID2 test" hit the front page of Digg.
# [23:39] <gsnedders> http://stuff.gsnedders.com/SD3.pdf and /SD4.pdf are what my computing homework is like
# [23:39] <DanC> hi Chris. I think I got all the critical business for this week done. I'm thinking about cancelling the 20 Dec telcon
# [23:40] <gsnedders> (SD3, due the Friday before last; SD4, due yesterday)
# [23:40] * DanC tunes in, scrolls back
# [23:40] <ChrisWilson> ok. As I mentioned b4, I'll be in transit during the scheduled call anyway.
# [23:41] <gsnedders> ChrisWilson: wow. the comments are amazingly good on digg :P
# [23:44] <ChrisWilson> yeah. For digg, that's a glowing review. :)
# [23:44] * DanC checks calendar for next week...
# [23:44] * gsnedders waves g'nite to *
# [23:45] * ChrisWilson waves back.
# [23:45] <DanC> ah. holidays next week... I wonder about Jan 3 too. Do you know whether you're in the office Jan 3, ChrisWilson ?
# [23:45] <ChrisWilson> nope.
# [23:46] <ChrisWilson> will be at home and not travelling, though.
# [23:46] <ChrisWilson> I could *probably* make a call. I do have visitors, though.
# [23:46] * Quits: gsnedders (gsnedders@86.135.224.200) (Quit: 404: Not Found)
# [23:46] <DanC> so if I cancel tomorrow, the next one is 10 Jan. hm.
# [23:47] * DanC likes the idea of a longish holiday break
# [23:51] <anne> "MathML isn’t written by hand in practice."
# [23:51] * anne hopes that can somehow be "fixed"
# [23:51] * DanC likes jsmath
# [23:52] <DanC> anne, picking on you randomly... ok by you if the next HTML WG telcon is 10 Jan?
# [23:53] <DanC> i.e. the next time the chairs sync up with action owners
# [23:53] <anne> seems fine, unless something goes wrong with publishing (though all seems good so far)
# [23:54] <anne> now I'm saying that, I wonder what we'll do with html4-differences...
# [23:54] <anne> and offline-webapps
# [23:57] <DanC> I'm pretty happy with putting the question on publishing html4-differences
# [23:57] * Quits: zcorpan (zcorpan@83.227.33.203) (Ping timeout)
# [23:57] <anne> I believe we already did while doing it for html5
# [23:57] <DanC> on offline-webapps, I tried the "2 reviewers" ritual, and we went 1-for-2. that's a pretty small sample.
# [23:58] <DanC> yes, and that question failed.
# [23:58] <DanC> so it's most straightforward to do it again
# [23:58] <anne> wasn't that the same question that causes HTML5 to be published?
# [23:58] * anne is slightly confused
# [23:59] <DanC> the question that causes the HTML 5 spec to be published hasn't been put yet. at least... that's the straightforward path. Hixie suggested re-using the earlier question... that seems quirky, but maybe it's more cost-effective than doing the whole thing again. Have you given that any thought, ChrisWilson ?
# Session Close: Thu Dec 20 00:00:00 2007

The end :)