/irc-logs / freenode / #whatwg / 2009-08-05 / end

Options:

# Session Start: Wed Aug 05 00:00:00 2009
# Session Ident: #whatwg
# [00:00] <othermaciej> frankly I would be ok with publishing a WD with either Hixie's text or John Foliot's text, my only annoyance is that it's been tied to WD publication when a WD by definition doesn't have to reflect consensus or resolve any specific issue
# [00:00] <Lachy> my doc isn't ready for publication yet. It hasn't been updated in a while and is very out of sync with the spec
# [00:00] * Quits: hobertoAtWork (n=hobertoa@gw1.mcgraw-hill.com) ("Nettalk6 - www.ntalk.de")
# [00:00] <Hixie> Lachy: *shrug*
# [00:01] <Lachy> also, I was awaiting the resolution for the spec licencing issue before I published it
# [00:01] <Hixie> Lachy: i don't really have any opinion or concern about the heartbeat requirement
# [00:01] <othermaciej> furthermore I think periodic WD publication has little meaning except pro forma satisfaction of W3C Process requirements
# [00:01] <Hixie> Lachy: my interest is in addressing the actual issue
# [00:01] <othermaciej> which is why I wish we wouldn't mix it up with other things
# [00:01] <Lachy> sure, so is mine. But having the actual issue tied to the publication of a draft is proving harmful. So finding a way to untie it and get on with addressing the actual issue seems productive
# [00:03] <Lachy> that's also why I proposed to to rubys in #html-wg the other day that we simply adopt a process in the group whereby any editor can ask for publication as a WD at any time and it be done without debate.
# [00:03] * Quits: mpilgrim (n=mark@216.239.45.19) (Read error: 110 (Connection timed out))
# [00:04] <Hixie> i would be fine with replacing the entire spec with a picture of bunnies and kittens for WD, if that's what you're worried about
# [00:04] <Lachy> LOL
# [00:05] <othermaciej> bunneh!!
# [00:05] <Hixie> i really couldn't care less about the WD itself, i'm concerned about fixing the actual problems in the language
# [00:05] <Lachy> othermaciej, re the study, I'll forward you a copy of the proposal I've sent to a few other collegues
# [00:06] <othermaciej> cool, I'll watch for it
# [00:06] * Quits: Samer (n=sziadeh@CPE0024369ef3ab-CM001ac35cd4b4.cpe.net.cable.rogers.com)
# [00:06] <Lachy> I'm keeping it off list for now to avoid it being destroyed before it even begins by some of the trolls in the group
# [00:08] <othermaciej> fair enough
# [00:09] <Hixie> feel free to post it to the whatwg list
# [00:09] <Hixie> don't let trolls dictate your behaviour
# [00:10] <Lachy> sent
# [00:15] <Hixie> don't forget to make the analysis phase blind also
# [00:15] <Hixie> it's important that the people analysing the results not know which group each table is from
# [00:16] * Quits: heycam (n=cam@124-168-112-102.dyn.iinet.net.au) ("bye")
# [00:16] * Quits: Maurice` (i=copyman@5ED548D4.cable.ziggo.nl) ("Disconnected...")
# [00:20] <Lachy> for the usability testing phase at the end, that will be possible, assuming the assitive technology users don't know any other details about the study
# [00:22] <webben_> othermaciej: http://lists.w3.org/Archives/Public/public-html/2009Aug/0221.html
# [00:22] <othermaciej> webben_: yes?
# [00:22] <webben_> position: absolute; left: -999999px; works well with screen readers.
# [00:23] <othermaciej> webben_: I do know that one works - what I don't know is if visibility: hidden or using an aural media stylesheet consistently works
# [00:23] * dbaron was thinking about implementing the canvas text changes (pretty easy), but wants to know what to do about the ctx.font getter first
# [00:23] <webben_> visibility: hidden; is known to have problems with (at least) older AT, and is generally undesirable as it still leaves a layout box.
# [00:24] <webben_> aural media stylesheet doesn't work with popular screen readers.
# [00:25] <Hixie> dbaron: what's the question? (looking at e-mail now)
# [00:25] <webben_> othermaciej: compare also existing WCAG advice: http://www.w3.org/TR/WCAG-TECHS/C7.html
# [00:25] <dbaron> Hixie, pretty much everything about the ctx.font getter is undefined :-)
# [00:25] <othermaciej> webben_: what about sizing something to 0 width and 0 height?
# [00:26] <dbaron> Hixie, like if you set ctx.font="2em serif" do you get back 2em or 32px? And if the font size on the canvas then changes?
# [00:26] <Hixie> dbaron: oh, i see. i'm relying on http://dev.w3.org/csswg/cssom/#css-value
# [00:26] <dbaron> oh boy
# [00:26] <dbaron> I suppose i should look at that spec then
# [00:26] <Hixie> dbaron: i'm not sure about the units, though
# [00:26] <dbaron> "Issue: This document is in its very early stages. The intention is that this specification will in due course supersede DOM Level 2 Style. "
# [00:26] <webben_> s/WCAG/WAI/
# [00:27] <Hixie> dbaron: i guess converting to pixels makes sense
# [00:27] <webben_> I've experienced problems with WAI's advice of moving content upwards.
# [00:27] <Hixie> dbaron: especially given the example in your e-mail
# [00:27] * Joins: mpt (n=mpt@canonical/launchpad/mpt)
# [00:27] <Hixie> dbaron: let's say that all units are converted to px on setting, and that the serialisation must use the rules in cssom
# [00:27] <dbaron> ok
# [00:27] <webben_> iirc had a doubling effect on links in Window-Eyes link list and did funny things with keyboard navigation.
# [00:27] <Hixie> dbaron: (those rules are basically the canonicalisation rules i made up a few years back)
# [00:28] <Hixie> dbaron: also if you find an editor for cssom, let me know :-)
# [00:28] <Hixie> anne's unable to spend time on it at the moment
# [00:28] <webben_> othermaciej: re font size/height 0: I'm not sure. I believe authors often avoid those as potential spam flags.
# [00:28] <Hixie> dbaron: i'll update the spec as soon as i'm done adding redirects and authentication to web sockets
# [00:28] * Joins: gsnedders (n=gsnedder@c83-252-206-210.bredband.comhem.se)
# [00:29] <dbaron> k
# [00:29] <dbaron> thanks
# [00:29] <Hixie> np
# [00:30] <webben_> othermaciej: old research: http://css-discuss.incutio.com/?page=ScreenreaderVisibility and http://www.access-matters.com/screen-reader-test-results/
# [00:30] * Joins: jennb (n=jennb@72.14.227.1)
# [00:33] * Joins: ojan (n=ojan@72.14.228.129)
# [00:34] * Quits: BlurstOfTimes (n=blurstof@168.203.117.59) ("Leaving...")
# [00:34] <webben_> othermaciej: (also old, but interesting re media-specific styles): http://web.archive.org/web/20071229112251/http://www.freedomscientific.com/fs_support/BulletinView.cfm?QC=1165
# [00:34] * Quits: ojan (n=ojan@72.14.228.129) (Client Quit)
# [00:34] * Joins: ojan (n=ojan@72.14.228.129)
# [00:35] <othermaciej> webben_: thanks for the info!
# [00:36] <GPHemsley> Has the Twitter account been fixed?
# [00:37] <takkaria> yes
# [00:37] <gsnedders> Aww… how boring.
# [00:38] <gsnedders> Hixie: You want help deleting all the spam?
# [00:38] <Hixie> nah, anne helped already
# [00:38] <Hixie> thanks though
# [00:38] <gsnedders> There's still more.
# [00:39] <Hixie> there's no spam left
# [00:39] <Hixie> just messages that were posted
# [00:39] <Hixie> which is fine
# [00:39] <Hixie> we don't delete criticism
# [00:39] <gsnedders> http://twitter.com/WHATWG doesn't look like it
# [00:39] <gsnedders> http://twitter.com/WHATWG/status/3125833340 looks far from sane
# [00:39] <Hixie> hm, there's some new ones since i last looked
# [00:39] <Hixie> i guess twitter is on some sort of delay
# [00:41] * Parts: ojan (n=ojan@72.14.228.129)
# [00:44] * Joins: heycam (n=cam@clm-laptop.infotech.monash.edu.au)
# [00:51] <webben_> othermaciej: I suspect people will know what you mean, but a large amount to the left -
# [00:51] <webben_> { position: absolute; left: 9999px; }
# [00:51] <webben_> is missing the "-" sign
# [00:51] <othermaciej> oh, right
# [00:51] <othermaciej> feel free to correct me on the list if you want
# [00:51] <othermaciej> I have some meetings to attend to
# [00:51] * Philip` doesn't suppose it'll make a difference until people have monitors 10,000 pixels wide
# [00:52] <gsnedders> And use browser windows that wide.
# [00:53] * webben_ used to use em units, but found this triggered a very odd bug in Safari, adding this absurdly wide horizontal scrollbar
# [00:53] <webben_> dunno if that's still the case in 4.
# [00:55] <Dashiva> That's a whole lot of email...
# [00:56] * Quits: StationsPatient (n=kristall@c-base/crew/kristall) (Read error: 110 (Connection timed out))
# [00:57] <Hixie> hmm
# [00:57] <Hixie> do Set-Cookie headers get honoured in redirects?
# [00:57] * webben_ wonders whether <ins><details></details></ins> is a better way to meet Roy's use-case than @summary. If textual fidelity is important then textual fidelity is important.
# [00:58] <Hixie> i.e. if i get a 301 Redirect and it includes a Set-Cookie, does it get set?
# [01:02] * Joins: StationsPatient (n=kristall@c-base/crew/kristall)
# [01:08] * Quits: svl (n=me@ip565744a7.direct-adsl.nl) ("And back he spurred like a madman, shrieking a curse to the sky.")
# [01:10] <Hixie> should we allow a connection to a wss:// secure socket to be redirected to a ws:// non-encrypted socket?
# [01:11] * Quits: gsnedders (n=gsnedder@c83-252-206-210.bredband.comhem.se)
# [01:12] <Dashiva> Automatically, or by script request, or by user request?
# [01:13] * Quits: othermaciej (n=mjs@17.203.15.225) (Read error: 104 (Connection reset by peer))
# [01:15] * Joins: Will_Robinson (n=chatzill@82-42-116-39.cable.ubr06.live.blueyonder.co.uk)
# [01:15] * Quits: Will_Robinson (n=chatzill@82-42-116-39.cable.ubr06.live.blueyonder.co.uk) (Client Quit)
# [01:16] * Joins: bill_robinson (n=chatzill@82-42-116-39.cable.ubr06.live.blueyonder.co.uk)
# [01:16] * Joins: othermaciej (n=mjs@17.203.15.225)
# [01:16] <bill_robinson> Hey guys
# [01:17] <othermaciej> hello
# [01:17] <bill_robinson> What's the password for the Twitter app?
# [01:17] * Joins: ap_ (n=ap@17.246.19.164)
# [01:18] <Hixie> bill_robinson: what do you want to tweet? :-)
# [01:18] <bill_robinson> HTML 5 tut
# [01:19] <Dashiva> Can I tweet the password? :P
# [01:19] * roc ignores Tom Lord harder
# [01:19] <Hixie> bill_robinson: uri?
# [01:19] <bill_robinson> Here:
# [01:20] <bill_robinson> http://snipurl.com/opruh
# [01:22] * Quits: ap_ (n=ap@17.246.19.164) (Remote closed the connection)
# [01:22] * Quits: ap (n=ap@nat/apple/x-llacaswgmtlrfmgc) (Read error: 104 (Connection reset by peer))
# [01:22] * Joins: ap (n=ap@nat/apple/x-hrvhdnqwdiqdhudm)
# [01:23] * Quits: bill_robinson (n=chatzill@82-42-116-39.cable.ubr06.live.blueyonder.co.uk) (Read error: 104 (Connection reset by peer))
# [01:23] <Hixie> aw, he left
# [01:23] <Hixie> $ curl -sD - http://snipurl.com/opruh | grep Location:
# [01:23] <Hixie> Location: http://meatspin.com
# [01:23] <Hixie> no idea what "meatspin.com" is, but it doesn't sound like a tutorial
# [01:26] * Quits: ap (n=ap@nat/apple/x-hrvhdnqwdiqdhudm) (Read error: 54 (Connection reset by peer))
# [01:26] * Joins: ap (n=ap@nat/apple/x-lghtlcjjrbxnmzjl)
# [01:27] <othermaciej> roc: does he have more new and exciting proposals?
# [01:27] <roc> no
# [01:27] <roc> he's just being difficult
# [01:27] <Dashiva> Is Tom Lord the latest dmitry or something?
# [01:30] <hober> he's pretty famous for being impossible
# [01:32] <Hixie> oops, forgot to update my checkin script's WHATWG password
# [01:35] <othermaciej> is there anyone on www-font who could be said to be contributing constructively to a likely useful end result?
# [01:36] <webben_> http://lists.w3.org/Archives/Public/www-font/ 1187 between July and Sept ... isn't it time that list got broke up into months?
# [01:37] <othermaciej> I'm looking at the archives now - glad I'm not on the list
# [01:37] <Hixie> othermaciej: as far as i can tell, the status quo is the most useful end result we can expect to get
# [01:43] * Quits: sbublava (n=stephan@77.118.248.251.wireless.dyn.drei.com)
# [01:45] <othermaciej> Hixie: I'm certainly not seeing anything useful reading the archives
# [01:45] <cying> grrr
# [01:46] <cying> that's the last time i click shortened urls at work
# [01:49] <Hixie> i'm not even at the office and i knew not to click that link, assuming you mean the one from bill earlier :-)
# [01:49] * Quits: tndH (n=Rob@cpc2-leed18-0-0-cust427.leed.cable.ntl.com) (Remote closed the connection)
# [01:57] <Hixie> wow, gsnedders|work's infinite loop is indeed implemented in gecko :-)
# [01:57] <Hixie> good work following the spec, hsivonen
# [01:57] <Hixie> guess i'd better fix that :-)
# [02:00] <dbaron> those are always fun. I think I had more trouble with :root { line-height: 1rem } than :root { font-size: 1rem; }
# [02:00] <Hixie> why is the former hard?
# [02:01] <Hixie> isn't it the same as line-height: 1em, effectively?
# [02:01] <dbaron> yes, except you need to remember to special-case the root
# [02:01] <dbaron> in our codebase, at least
# [02:01] <Hixie> ah
# [02:01] <Hixie> this particular infinite loop was that <input> in <select> implies </select> and then reprocesses the token
# [02:02] <Hixie> but if you're setting the innerHTML on a <select>, the </select> is ignored
# [02:02] <Hixie> so nothing changed, and reprocessing the token just tries again
# [02:02] <dbaron> since otherwise you get: go off and find the root element and get its style... oh wait, it doesn't have one, let's compute it ... ad inf.
# [02:02] <Hixie> nice
# [02:12] * Quits: dglazkov (n=dglazkov@nat/google/session)
# [02:16] * Quits: mpt (n=mpt@canonical/launchpad/mpt) (Read error: 113 (No route to host))
# [02:16] * Quits: smedero (n=smedero@D-128-208-133-191.dhcp4.washington.edu)
# [02:31] * Quits: ap (n=ap@nat/apple/x-lghtlcjjrbxnmzjl)
# [02:31] <cying> Hixie: yea, i'm slow today :)
# [02:36] * Joins: gimpscape (n=gimpscap@ayq42.neoplus.adsl.tpnet.pl)
# [02:49] * Joins: webben (n=benh@91.85.212.62)
# [02:52] * Quits: Amorphous (i=jan@unaffiliated/amorphous) (Read error: 110 (Connection timed out))
# [02:53] * Quits: cying (n=cying@70.90.171.153)
# [02:55] * Joins: Amorphous (i=jan@unaffiliated/amorphous)
# [02:58] <roc> othermaciej: I think there's a possible world not very different from our own where "EOT Lite" is worth supporting
# [02:59] <roc> we are making progress towards determining whether we live in that world
# [02:59] * Quits: jacobolus (n=jacobolu@adsl-76-202-196-235.dsl.pltn13.sbcglobal.net) (Read error: 54 (Connection reset by peer))
# [02:59] <othermaciej> roc: seems like there's a few big contentious points around that discussion
# [02:59] <roc> yep
# [02:59] <othermaciej> - what to do about root strings?
# [02:59] * Quits: webben_ (n=benh@dip5-fw.corp.ukl.yahoo.com) (Read error: 110 (Connection timed out))
# [02:59] <othermaciej> - what about inevitable pressure on vendors that support only Lite in the face of Classic content?
# [03:00] <othermaciej> - is it tied to a requirement to implement vanilla OTF/TTF (or is the latter otherwise placed on an equal footing)?
# [03:00] <roc> I agree these issues exist
# [03:00] <roc> I can imagine them being solved
# [03:00] <othermaciej> maybe they are solvable
# [03:01] <roc> I can't imagine Tom Lord contributing to their solution in any way
# [03:01] <othermaciej> Håkon seemed to like the idea of a format not based on EOT at all
# [03:01] <othermaciej> in the smattering of messages I saw
# [03:01] <othermaciej> it does seem like Tom Lord mostly just periodically pours fuel on the fire
# [03:06] * Joins: webben_ (n=benh@genkt-048-005.t-mobile.co.uk)
# [03:06] * Quits: webben_ (n=benh@genkt-048-005.t-mobile.co.uk) (Read error: 104 (Connection reset by peer))
# [03:10] * Joins: hdh (n=hdh@hdh-1-pt.tunnel.tserv20.hkg1.ipv6.he.net)
# [03:10] <roc> I take it back
# [03:10] <roc> Tom Lord may contribute to the solution by uniting everyone else against him
# [03:12] * Quits: webben (n=benh@91.85.212.62) (Read error: 110 (Connection timed out))
# [03:14] <Hixie> what does HTTP stand for?
# [03:14] <roc> Hyper Text Transport Protocol?
# [03:14] <roc> or is it Transfer?
# [03:14] <Hixie> hm, seems to be transfer
# [03:14] <Hixie> ok
# [03:15] <roc> presumably that was some sort of human experiment, not a genuine request for information?
# [03:15] <Hixie> neither really
# [03:15] <Hixie> i had remembered it as transport
# [03:15] <othermaciej> Transfer
# [03:15] <Hixie> and was thinking out loud
# [03:16] <othermaciej> I think by analogy with FTP
# [03:18] * Quits: jennb (n=jennb@72.14.227.1)
# [03:31] * Joins: xydyx (n=hdh@118.71.102.245)
# [03:46] * Joins: webben (n=benh@genkt-048-005.t-mobile.co.uk)
# [03:46] * Quits: webben (n=benh@genkt-048-005.t-mobile.co.uk) (Read error: 104 (Connection reset by peer))
# [03:48] * Quits: hdh (n=hdh@hdh-1-pt.tunnel.tserv20.hkg1.ipv6.he.net) (Read error: 110 (Connection timed out))
# [03:54] * Joins: jorlow (n=jorlow@nat/google/x-kfypqbwbbkkfwlda)
# [03:56] * Quits: gimpscape (n=gimpscap@ayq42.neoplus.adsl.tpnet.pl) ("Ex-Chat")
# [04:00] * Joins: Samer (n=sziadeh@CPE0024369ef3ab-CM001ac35cd4b4.cpe.net.cable.rogers.com)
# [04:03] * Quits: jorlow (n=jorlow@nat/google/x-kfypqbwbbkkfwlda)
# [04:06] * Joins: dglazkov (n=dglazkov@c-67-188-0-62.hsd1.ca.comcast.net)
# [04:10] * Quits: dglazkov (n=dglazkov@c-67-188-0-62.hsd1.ca.comcast.net) (Client Quit)
# [04:10] * Quits: Samer (n=sziadeh@CPE0024369ef3ab-CM001ac35cd4b4.cpe.net.cable.rogers.com)
# [04:13] <othermaciej> I'm thinking of sending an email that both quotes from our charter and cites first-order predicate calculus
# [04:13] <othermaciej> this seems, on the face of it, like a poor choice
# [04:23] <Hixie> right well i think i'm gonna take a break, and then read the 84 e-mails on summary that were sent today
# [04:24] * Joins: Samer (n=sziadeh@CPE0024369ef3ab-CM001ac35cd4b4.cpe.net.cable.rogers.com)
# [04:25] * Quits: paul_irish (n=paul_iri@12.33.239.250)
# [04:27] <othermaciej> I was *this* close to deciding to just stay out of the whole thing for good
# [04:29] * Joins: jorlow (n=jorlow@nat/google/x-atccqwlgwxelmoos)
# [04:30] * Quits: Samer (n=sziadeh@CPE0024369ef3ab-CM001ac35cd4b4.cpe.net.cable.rogers.com)
# [04:49] <roc> take a break and read www-font instead
# [04:52] <othermaciej> haha
# [04:53] <othermaciej> well, at least the endless flamewars on www-font aren't getting in the way of anything I care about or costing time I could spend on more useful things
# [04:53] <othermaciej> so they could be more entertaining
# [04:57] * Quits: jorlow (n=jorlow@nat/google/x-atccqwlgwxelmoos)
# [05:01] * Joins: ttepass- (n=ttepas--@p5B017CD7.dip.t-dialin.net)
# [05:07] * Joins: Samer (n=sziadeh@CPE0024369ef3ab-CM001ac35cd4b4.cpe.net.cable.rogers.com)
# [05:14] * Quits: ttepasse (n=ttepas--@p5B0165C2.dip.t-dialin.net) (Read error: 110 (Connection timed out))
# [05:15] * Quits: ttepass- (n=ttepas--@p5B017CD7.dip.t-dialin.net) (Read error: 60 (Operation timed out))
# [05:22] * Joins: Xanthir (n=chatzill@99-35-179-251.lightspeed.hstntx.sbcglobal.net)
# [05:23] <Xanthir> Yo, opinions requested: I'm putting together some js-based feature testing. Opera and Safari implement <input type=range>, but don't properly sanitize values. Should I mark them as supporting that?
# [05:27] * Quits: Samer (n=sziadeh@CPE0024369ef3ab-CM001ac35cd4b4.cpe.net.cable.rogers.com)
# [05:27] * Quits: tkent_ (n=tkent@220.109.219.244) ("Leaving...")
# [05:30] * Quits: dbaron (n=dbaron@c-98-234-51-190.hsd1.ca.comcast.net) ("8403864 bytes have been tenured, next gc will be global.")
# [05:36] * Joins: dglazkov (n=dglazkov@c-67-188-0-62.hsd1.ca.comcast.net)
# [05:37] * Quits: othermaciej (n=mjs@17.203.15.225)
# [05:38] * Joins: tkent (n=tkent@220.109.219.244)
# [05:47] * Joins: webben (n=benh@genkt-048-005.t-mobile.co.uk)
# [05:47] * Quits: webben (n=benh@genkt-048-005.t-mobile.co.uk) (Read error: 104 (Connection reset by peer))
# [05:56] * Joins: dglazkov_ (n=dglazkov@c-67-188-0-62.hsd1.ca.comcast.net)
# [06:02] * Joins: abarth (n=abarth@c-98-210-108-185.hsd1.ca.comcast.net)
# [06:03] * Joins: Samer (n=sziadeh@CPE0024369ef3ab-CM001ac35cd4b4.cpe.net.cable.rogers.com)
# [06:04] <abarth> Hixie: do you know what the IP restrictions on RFCs are like? Can I just copy-and-paste paragraphs from an RFC to an Internet-Draft?
# [06:04] * Quits: mookid (i=mookid@ROFL.name) (Read error: 113 (No route to host))
# [06:05] * Quits: Samer (n=sziadeh@CPE0024369ef3ab-CM001ac35cd4b4.cpe.net.cable.rogers.com) (Client Quit)
# [06:11] * Joins: jorlow (n=jorlow@c-67-180-199-19.hsd1.ca.comcast.net)
# [06:12] * Quits: dglazkov (n=dglazkov@c-67-188-0-62.hsd1.ca.comcast.net) (Read error: 113 (No route to host))
# [06:19] * Joins: gavin______ (n=gavin@CPE001346f5db49-CM0018c0db9a8a.cpe.net.cable.rogers.com)
# [06:20] * gavin______ is now known as gavin
# [06:26] * Quits: weinig (n=weinig@nat/apple/x-vyrlxuadcflahvyi)
# [06:27] * Quits: Xanthir (n=chatzill@99-35-179-251.lightspeed.hstntx.sbcglobal.net) (Read error: 110 (Connection timed out))
# [06:37] * Joins: othermaciej (n=mjs@c-69-181-42-237.hsd1.ca.comcast.net)
# [06:37] * Quits: dglazkov_ (n=dglazkov@c-67-188-0-62.hsd1.ca.comcast.net)
# [06:44] * Quits: dolske (n=dolske@nat/mozilla/x-xicdceisbraqywef)
# [06:46] * Joins: dave_levin_ (n=dave_lev@72.14.224.1)
# [06:49] * Quits: dave_levin (n=dave_lev@72.14.227.1)
# [06:56] * dave_levin_ is now known as dave_levin
# [07:04] <ezyang> Oh noes, infinite loop
# [07:04] * Joins: Xanthir (n=chatzill@99-35-179-251.lightspeed.hstntx.sbcglobal.net)
# [07:16] * Joins: mpilgrim (n=mark@ip-64-139-15-210.dsl.sca.megapath.net)
# [07:17] <mpilgrim> othermaciej (or anyone who might know), how much of videoElement.canPlayType() does safari support?
# [07:17] <othermaciej> mpilgrim: we support it, I think it may be buggy in some cases (depending on what release)
# [07:18] * Joins: dolske (n=dolske@c-76-103-40-203.hsd1.ca.comcast.net)
# [07:19] <mpilgrim> including the codecs parameter?
# [07:19] * Quits: Xanthir (n=chatzill@99-35-179-251.lightspeed.hstntx.sbcglobal.net) ("ChatZilla 0.9.85 [Firefox 3.5.2/20090729225027]")
# [07:21] <othermaciej> yes, we look at the codecs parameter
# [07:22] <othermaciej> I think the last released Safari may have a bug with properly detecting that XiphQT is installed. It should be fixed in 4.0.3, once that is out.
# [07:22] * Joins: jorlow_ (n=jorlow@72.14.224.1)
# [07:22] <mpilgrim> use case: http://wearehugh.com/public/2006/12/video.html
# [07:22] <mpilgrim> via markup, it works fine (tested in latest release of safari 4.0.x, on windows)
# [07:23] <mpilgrim> i want to write a script that asks the browser whether it can play that video
# [07:24] <mpilgrim> "mp4creator -list 20061225.mp4" says it contains an MPEG-4 AAC LC audio track
# [07:24] <othermaciej> I believe asking about video/mp4 with the proper codecs parameter will work from script
# [07:24] <mpilgrim> and an H264 Baseline (level 1.3) video track
# [07:25] <mpilgrim> (in an MP4 container, obviously)
# [07:25] <mpilgrim> how would i format the codecs parameter to test that particular combination?
# [07:25] <othermaciej> I am not an expert on the details of the mpeg-4 codecs parameter values, but I think the example of one in the spec would work
# [07:25] <othermaciej> though it may not encapsulate exactly that profile set
# [07:25] <mpilgrim> the example in the spec always returns "maybe"
# [07:25] <mpilgrim> is that the most confident answer i can expect?
# [07:26] <mpilgrim> or is there another input that would return "probably"?
# [07:27] <othermaciej> I think we might be wrong about when exactly we say "maybe" vs "probably"
# [07:27] <othermaciej> but we won't say "no" (or, in the next version, "") for a codec we can actually play
# [07:31] <mpilgrim> i have quicktime installed
# [07:31] <mpilgrim> but v.canPlayType("video/mp4;codecs=avc1.42E01E,mp4a.40.2") returns "maybe" in win/safari 4.0.2
# [07:32] * Quits: xydyx (n=hdh@118.71.102.245) (Remote closed the connection)
# [07:32] <mpilgrim> but returns "probably" in chrome 3.0.196.2
# [07:32] <mpilgrim> (both browsers can play a video matching that codec parameter)
# [07:38] * Quits: jorlow (n=jorlow@c-67-180-199-19.hsd1.ca.comcast.net) (Read error: 110 (Connection timed out))
# [07:41] <roc> chrome lies, so that's not necessarily a good data point
# [07:42] <roc> although I think they may have fixed that bug
# [07:45] <mpilgrim> i think that's fixed now
# [07:46] <mpilgrim> hmm
# [07:46] <mpilgrim> v.canPlayType("video/mp4;codecs=foobar") returns "maybe" (chrome 3.0.196.2)
# [07:56] * Joins: shepazu (n=schepers@75-149-110-74-Chattanooga.hfc.comcastbusiness.net)
# [08:02] * Joins: clarkeyboy (n=darryl_c@78.149.170.205)
# [08:04] <othermaciej> I think Safari and Chrome both have bugs here
# [08:05] <othermaciej> some are in the cross-platform media element code perhaps, and some are likely in the media engine back ends
# [08:06] * Joins: maikmerten (n=merten@ls5dhcp196.cs.uni-dortmund.de)
# [08:07] * Joins: webben (n=benh@91.85.212.62)
# [08:12] * Joins: webben_ (n=benh@dip5-fw.corp.ukl.yahoo.com)
# [08:13] * Parts: clarkeyboy (n=darryl_c@78.149.170.205)
# [08:16] <mpilgrim> sigh
# [08:16] <mpilgrim> thanks othermaciej, roc
# [08:16] * Joins: jorlow (n=jorlow@c-67-180-199-19.hsd1.ca.comcast.net)
# [08:17] * Quits: jorlow (n=jorlow@c-67-180-199-19.hsd1.ca.comcast.net) (Remote closed the connection)
# [08:20] * Quits: heycam (n=cam@clm-laptop.infotech.monash.edu.au) ("bye")
# [08:24] * Quits: webben (n=benh@91.85.212.62) (Read error: 110 (Connection timed out))
# [08:29] * Joins: heycam (n=cam@dyn-130-194-69-76.infotech.monash.edu.au)
# [08:32] * Quits: jorlow_ (n=jorlow@72.14.224.1) (Read error: 110 (Connection timed out))
# [08:35] * Quits: mpilgrim (n=mark@ip-64-139-15-210.dsl.sca.megapath.net) (Read error: 110 (Connection timed out))
# [08:41] * Quits: virtuelv (n=virtuelv@201.169.34.95.customer.cdi.no) ("Ex-Chat")
# [08:41] <hsivonen> I wish that a week or so ago, someone had taken the time to do the WCAG 2 spec lawyering that I just did.
# [08:50] * Quits: sebmarkbage (n=miranda@h-6-72.A146.priv.bahnhof.se) ("http://calyptus.eu/")
# [08:52] * Joins: Samer (n=sziadeh@CPE0024369ef3ab-CM001ac35cd4b4.cpe.net.cable.rogers.com)
# [08:58] * Joins: jacobolus (n=jacobolu@c-98-248-43-68.hsd1.ca.comcast.net)
# [09:03] <othermaciej> you raised a darn good point hsivonen
# [09:03] <othermaciej> I did notice the part that cited summary="" was non-normative
# [09:04] <othermaciej> I did not notice it claimed HTML 4.01 and XHTML 1.x as scope
# [09:04] <othermaciej> also it looks like WAI people are working on submitting the HTML5 techniques (plus probably something ARIA-based) to the relevant section of WCAG2 techniques
# [09:05] <othermaciej> I still think my proposal is a reasonable deal if people take are willing to take it
# [09:06] <othermaciej> so far most of the objection is from summary fans, and mainly the ones staking out a more extreme position than the people closely involved in WAI
# [09:14] <abarth> othermaciej: do you know anything about the licensing of RFC text?
# [09:14] <othermaciej> abarth: no idea about it
# [09:15] <abarth> do you know who would be good to ask?
# [09:15] <othermaciej> abarth: I know there's an RFC describing the IETF's IPR policy, which may or may not cover this
# [09:15] <othermaciej> I read over it once, can't remember the details.
# [09:16] <abarth> thanks
# [09:18] <hsivonen> abarth: IIRC, RFC text used to be very non-Free. However, the policy was recently adjusted, but, IIRC, it's still non-Free. I don't know the details.
# [09:18] <abarth> hsivonen: i'm trying to figure out if i can use rfc2109 text in a new cookie internet-draft
# [09:18] <abarth> this document
# [09:18] <abarth> http://www.rfc-editor.org/copyright.html
# [09:18] <abarth> seems to say yes
# [09:19] <abarth> as long as i cite
# [09:19] <abarth> because my use is within the IETF standards process
# [09:19] <othermaciej> you're writing a new draft for cookies?
# [09:20] <othermaciej> I need to join these http-related mailing lists I think
# [09:20] <abarth> othermaciej: i;m considering it
# [09:20] <othermaciej> abarth: are you going to include spec language for the actual real cookies people really use, and not just the imaginary "v2" cookies?
# [09:20] <abarth> yeah, i want to do real cookies
# [09:20] <abarth> with very little deviation (if any) from existing reality
# [09:21] <othermaciej> because if so, you will have even more of my gratitude than you already do for all the various awesome things you do
# [09:21] <abarth> its kind of an embarrassment that no cookie spec exists
# [09:21] <abarth> thanks :)
# [09:21] <abarth> if you're interested, you should subscribe to http-state@ietf.org
# [09:21] <abarth> here's the git repo i'm working in
# [09:21] <abarth> http://github.com/abarth/http-state/tree/master
# [09:22] <abarth> i'm triaging rfc2109 text at the moment
# [09:22] <abarth> and i'm trying to figure out if i can pull it to the public repo
# [09:24] <othermaciej> yeah I think I need to subscribe to http-state and ietf-http-wg
# [09:24] <othermaciej> maybe I can convince ap and/or weinig to sign up too, since they know this area and have been known to be useful standards participants
# [09:25] <abarth> i haven't quite gotten the swing of IETF yet
# [09:25] <abarth> there seem to be a lot of naysayers
# [09:25] <abarth> i think they're used to rejecting crappy ideas
# [09:27] <hsivonen> good ones, too :-/
# [09:27] * Joins: jorlow (n=jorlow@72.14.224.1)
# [09:30] * Joins: gunderwonder (n=gunderwo@garage.upstruct.com)
# [09:31] <abarth> alright, i pushed my branch to github. hopefully everything is kosher w.r.t. IPR
# [09:35] * Joins: Maurice (i=copyman@5ED548D4.cable.ziggo.nl)
# [09:37] <Hixie> http://softwareas.com/whatwg-twitter-day
# [09:37] <abarth> haha
# [09:39] <abarth> did you have an open twitter on purpose?
# [09:39] <Hixie> yes
# [09:39] <Hixie> that article is entirely accurate
# [09:50] * Joins: mpt (n=mpt@canonical/launchpad/mpt)
# [09:50] <hsivonen> http://stackoverflow.com/questions/472274/why-are-xhtml-5-and-xhtml-2-separate-standards/1061773#1061773
# [09:57] <hsivonen> Hixie: are you tracking StackOverflow as one of your feedback sources?
# [09:57] <Hixie> no
# [09:58] <hsivonen> seems useful for finding out what developers want to know or didn't figure out on their own
# [10:01] <Hixie> i see it occasionally
# [10:01] * Joins: mat_t (n=mattomas@89.127.177.74)
# [10:01] <Hixie> i just haven't been tracking it
# [10:02] * Quits: mat_t (n=mattomas@89.127.177.74) (Client Quit)
# [10:04] * Quits: mpt (n=mpt@canonical/launchpad/mpt) (Read error: 60 (Operation timed out))
# [10:10] * Quits: shepazu (n=schepers@75-149-110-74-Chattanooga.hfc.comcastbusiness.net)
# [10:11] * Quits: heycam (n=cam@dyn-130-194-69-76.infotech.monash.edu.au) ("bye")
# [10:12] * Quits: webben_ (n=benh@dip5-fw.corp.ukl.yahoo.com) (Read error: 104 (Connection reset by peer))
# [10:19] * Quits: othermaciej (n=mjs@c-69-181-42-237.hsd1.ca.comcast.net)
# [10:21] * abarth is now known as abarth|zzz
# [10:23] * Quits: roc (n=roc@203-97-204-82.dsl.clear.net.nz)
# [10:25] <gsnedders|work> Wee! Exam results time!
# [10:25] * Joins: archtech (n=sv@83.228.56.37)
# [10:27] * Joins: virtuelv (n=virtuelv@pat-tdc.opera.com)
# [10:28] * Philip` sees http://www.khronos.org/news/press/releases/khronos-webgl-initiative-hardware-accelerated-3d-graphics-internet/
# [10:35] * Quits: jwalden (n=waldo@nat/mozilla/x-ipitrffidnihcnzh) ("ChatZilla 0.9.85 [Firefox 3.5.3pre/20090804032127]")
# [10:39] * gsnedders|work marks another part of his life plan as failed, and wonders how he did so badly in English especially
# [10:41] * Quits: jorlow (n=jorlow@72.14.224.1)
# [10:45] * Joins: mat_t (n=mattomas@89.127.177.75)
# [10:48] * Joins: svl (n=me@ip565744a7.direct-adsl.nl)
# [10:50] <jgraham> gsnedders|work: Oh
# [10:51] <jgraham> Maybe they didn't like Nabokov as much as you?
# [10:51] * Joins: mpt (n=mpt@canonical/launchpad/mpt)
# [10:51] <gsnedders|work> jgraham: And the other 60% of the marks? :P
# [10:51] <Philip`> Did you accidentally write all your answers in French?
# [10:52] * Joins: zcorpan (n=zcorpan@pat.se.opera.com)
# [10:52] <gsnedders|work> Non, je ne pense pas…
# [10:52] <Philip`> Oh, okay, just a thought
# [10:52] * Joins: heycam (n=cam@124-168-112-102.dyn.iinet.net.au)
# [10:52] <zcorpan> Hixie: i note that  is equivalent to 
 in html, although i tried \r with script with the same result
# [10:53] <zcorpan> i wonder if there are more characters that are stripped before parsing
# [10:54] <zcorpan> U+000B?
# [10:54] <jgraham> German?
# [10:54] <gsnedders|work> jgraham: Nah, I lack the eloquence to write German.
# [10:54] <jgraham> How about Finnish?
# [10:55] <gsnedders|work> If I tried to write Finnish, I wouldn't have passed.
# [10:56] * Joins: webben (n=benh@nat/yahoo/x-bevwajvlwgicihjd)
# [11:01] * Quits: Samer (n=sziadeh@CPE0024369ef3ab-CM001ac35cd4b4.cpe.net.cable.rogers.com)
# [11:05] <annevk42> sad that the first person asking for the twitter account is a spammer
# [11:05] <annevk42> I guess that is to be expected
# [11:07] <gsnedders|work> Almost inevitable after word got out that it was open yesterday though
# [11:07] <annevk42> more coverage: http://blog.gingertech.net/2009/08/05/amusement-at-whatwg/
# [11:14] * Joins: Rik` (n=Rik`@pha75-2-81-57-187-57.fbx.proxad.net)
# [11:19] * Joins: roc (n=roc@121-72-186-57.dsl.telstraclear.net)
# [11:22] <jgraham> Hixie: The review comment status box on the whatwg spec is pretty annoying nad impossible to hide if your browser window is too narrow
# [11:23] <jgraham> It would be better if it was closed by default but said something like "Make comment" so that people would know to open it
# [11:24] <zcorpan> Hixie: maybe you could also make it 100% wide (with the input box flexing) when open
# [11:24] <jgraham> Then only once you have opened it would clicking do anything. It would also be nice to have more visual feedback about which section you were commenting on (e.g. by changing the backgroung colour of the relevant section(
# [11:25] <jgraham> and what zcorpan said so it is always possible to hide
# [11:29] * Joins: poe (n=poe@unaffiliated/xerox)
# [11:36] <Hixie> jgraham: how narrow is your screen?!
# [11:37] <Hixie> jgraham: send me an updated .js file and i'll replace the one on the site
# [11:37] <jgraham> Hixie: I typically have two browser windows side by side
# [11:37] <jgraham> I'll take a look at the js file
# [11:37] <jgraham> (sometime)
# [11:44] * Joins: adactio (n=adactio@host86-132-125-223.range86-132.btcentralplus.com)
# [11:48] * Joins: erikvold (n=erikvvol@96.49.192.204)
# [11:50] <Philip`> You can hide the review comment status box?
# [11:50] <Philip`> Oh, seems to be the unlabelled unintuitive arrow button
# [12:05] * Quits: erikvvold (n=erikvvol@96.49.192.204) (Read error: 110 (Connection timed out))
# [12:09] <gsnedders|work> Huh. perl -p and perl -ni do different things.
# [12:18] <Philip`> That's because they're different commands
# [12:23] * Quits: mpt (n=mpt@canonical/launchpad/mpt) ("Ex-Chat")
# [12:24] <jgraham> What happened to TMTOWTTI?
# [12:25] <annevk42> http://lists.w3.org/Archives/Public/public-html/2009Aug/0239.html nice
# [12:25] <jgraham> s/TT/TD/
# [12:25] <gsnedders|work> Philip`: I mean in terms of what they do to the file
# [12:30] * Joins: ROBOd (n=robod@89.122.216.38)
# [12:34] * Joins: yusukes (n=yusukes@220.109.219.244)
# [12:35] <GPHemsley> So... id attributes can contain any characters?
# [12:35] <zcorpan> except whitespace and can't be empty string
# [12:35] <GPHemsley> is that what HTML4 allows, too?
# [12:35] <zcorpan> i think html4 was more restrictive
# [12:36] <GPHemsley> that's what I thought
# [12:37] * Quits: Rik` (n=Rik`@pha75-2-81-57-187-57.fbx.proxad.net) (Read error: 113 (No route to host))
# [12:38] * Joins: mpt (n=mpt@canonical/launchpad/mpt)
# [12:41] * Joins: annevk2 (n=annevk@5355732C.cable.casema.nl)
# [12:42] <annevk2> hmm, the messages on @whatwg that start with @{something} are not shown in the listing where I could remove them...
# [12:43] * Joins: Lachy_ (n=Lachy@pat-tdc.opera.com)
# [12:44] <zcorpan> hmm, it seems firefox, safari and chrome don't support the event handler dom attributes for <video>, other than .onload :-(
# [12:45] * Joins: MadAtWork2 (n=not_@83.139.194.2)
# [12:48] <zcorpan> but they do support the event handler content attributes
# [12:53] * Quits: mat_t (n=mattomas@89.127.177.75) ("This computer has gone to sleep")
# [12:54] * zcorpan switches to the more verbose addEventListener
# [12:56] * Joins: mat_t (n=mattomas@89.127.177.75)
# [12:56] <annevk2> I think for <ruby> we should just make support mandatory and remove <rp>
# [12:57] <annevk2> especially since it has been stated that the fallback is inadequate
# [12:59] <MadAtWork2> hm, http://www.w3.org/TR/2009/ED-html5-diff-20090801/ is a big fat 404.
# [12:59] * Parts: heycam (n=cam@124-168-112-102.dyn.iinet.net.au) ("bye")
# [12:59] * Joins: heycam (n=cam@124-168-112-102.dyn.iinet.net.au)
# [12:59] * Joins: danbri (n=danbri@78-2-188-237.adsl.net.t-com.hr)
# [13:00] <MadAtWork2> Linked from the latest dev draft of the html5 diff file annevk2.
# [13:00] <annevk2> it's not actually supposed to point to anything
# [13:01] <annevk2> it's just auto-generated every time I do a commit and when we publish ED changes to WD, the date changes to the publication date and the webmaster will make sure the file is actually there
# [13:01] <MadAtWork2> Okay.
# [13:01] * Quits: Yudai (n=Yudai@p929c7b.kngwnt01.ap.so-net.ne.jp) (Read error: 110 (Connection timed out))
# [13:01] <annevk2> feel free to point out any other errors though :)
# [13:01] <MadAtWork2> Sure!
# [13:02] * Philip` discovers that the #whatwg approach of having no ops on a channel is a pretty bad idea on QuakeNet, because if there's no ops for some time period then anybody can automatically request ops and take over the channel
# [13:02] <MadAtWork2> It could be worse.
# [13:03] <MadAtWork2> It could be ircnet.
# [13:12] * Quits: dave_levin (n=dave_lev@72.14.224.1)
# [13:14] <Dashiva> How did this message end up in www-archive? http://lists.w3.org/Archives/Public/www-archive/2009Aug/0008.html
# [13:14] <gsnedders|work> Dashiva: The "To" field.
# [13:17] * Quits: Lachy_ (n=Lachy@pat-tdc.opera.com) ("Leaving")
# [13:18] <hsivonen> +1 to Lachy's reply to Hixie on public-html, even though we don't do the +1 thing
# [13:19] * Quits: mpt (n=mpt@canonical/launchpad/mpt) (Remote closed the connection)
# [13:20] * Joins: Lachy_ (n=Lachy@pat-tdc.opera.com)
# [13:22] * Quits: StationsPatient (n=kristall@c-base/crew/kristall) ("Wünsche weiterhin guten Flug")
# [13:24] * Joins: mpt (n=mpt@canonical/launchpad/mpt)
# [13:26] * Quits: danbri (n=danbri@unaffiliated/danbri)
# [13:27] * Quits: mpt (n=mpt@canonical/launchpad/mpt) (Remote closed the connection)
# [13:32] <MikeSmith> hsivonen: +1 to your +1
# [13:33] * MikeSmith figures he should probably actually read Lachy's reply before indirectly +1'ing it
# [13:34] <takkaria> MikeSmith: +1 (score quotes implied)
# [13:34] <MikeSmith> takkaria: brilliant
# [13:34] <jgraham> -1 to MikeSmith's +1 since he did it without due dilligence
# [13:37] <MadAtWork2> If you are referring to http://lists.w3.org/Archives/Public/public-html/2009Aug/0261.html I add my +0.1 to it.
# [13:37] * gsnedders|work adds 2 + 3i to it
# [13:37] <MadAtWork2> I've seen plenty of companies stuck with IE6 and the subsequent security problems because of poopy intranet apps.
# [13:37] <gsnedders|work> See, I'm just unreal.
# [13:38] <jgraham> You can only add numbers within the unit circle
# [13:39] <MadAtWork2> Right now I'm working on a little intranet app, using css3 and html5.
# [13:39] <MadAtWork2> People will have to use webkit/presto or gecko :P
# [13:39] <jgraham> Like sqrt(0.5) - sqrt(0.5)i
# [13:39] <Dashiva> jgraham: +Δe
# [13:40] <gsnedders|work> jgraham: Only within the unit circle? :\
# [13:40] * gsnedders|work doesn't remember that
# [13:42] <jgraham> gsnedders|work: The set of z such that |z|<=1
# [13:42] <gsnedders|work> Yes, I know what you mean.
# [13:42] <gsnedders|work> But I just don't remember that restriction.
# [13:43] <jgraham> Well we can't have people going around adding +2 or +3 or whatever
# [13:43] <gsnedders|work> Oh, duh.
# [13:43] * gsnedders|work adds 2 + i^2 then
# [13:45] * Joins: mpt (n=mpt@canonical/launchpad/mpt)
# [13:46] <Dashiva> jgraham: You can just truncate the value
# [13:47] <takkaria> whoever put "device-width" in CSS media queries should be shot
# [13:48] <gsnedders|work> takkaria: Well, there's a 50% chance it was an Opera employee
# [13:51] <jgraham> takkaria: What do you particularly dislike about it?
# [13:52] <takkaria> jgraham: well, it's pretty redundant with "width" there
# [13:52] * Joins: Rik` (n=Rik`@pha75-2-81-57-187-57.fbx.proxad.net)
# [13:52] <hsivonen> IMO people have no business querying my device. They should stick to querying the view port I choose to grant them.
# [13:52] * Quits: svl (n=me@ip565744a7.direct-adsl.nl) ("And back he spurred like a madman, shrieking a curse to the sky.")
# [13:52] <takkaria> jgraham: and it means that people write media queries using "device-width", test in destop browsers and find it works, only for someone else to use a screen whose width is acutally quite small and find that it doesn't work as intended
# [13:53] <takkaria> pretty much what hsivonen said
# [13:53] <hsivonen> being able to query the device means the site can nag "hey, I know your screen is wider, please, please maximize me!!!!"
# [13:53] <takkaria> if I have a narrow viewport, then you should be switching on that, not the theoretical minimum or maximum viewport I could have
# [13:53] <jgraham> I agree width seems like the right thing to use in almost all situations
# [13:53] <Dashiva> Next up: Making UAs report viewport-width as the value of device-width
# [13:54] <jgraham> So maybe the world would be better without device-width
# [13:54] <Dashiva> No, because then people would be bugging others to implement it
# [13:56] <annevk2> i tried to kill device- features, fwiw
# [13:56] * hsivonen deletes 1025 emails to non-public-html w3.org mailing lists that have also been sent to public-html
# [13:57] <annevk2> you could raise the issue yet again
# [13:57] * Lachy_ tries to calculate the total that has been added to his mail, and then gets stuck with gsnedders' use of imaginary numbers
# [13:57] <annevk2> i'd again argue on the side of removal
# [13:57] <takkaria> annevk2: is it worth doing, do you think?
# [13:57] * Joins: myakura (n=myakura@p3137-ipbf7310marunouchi.tokyo.ocn.ne.jp)
# [13:58] <annevk2> i don't think hsivonen's concern has been brought up before actually
# [13:58] <annevk2> my main argument has been is that you do not need them and that therefore we should not have them
# [13:59] <takkaria> I guess I should go poke chaals and get member access so I can post, then
# [13:59] <annevk2> though Björn Höhrmann did have a somewhat reasonable case for device-aspect-ratio
# [13:59] <annevk2> takkaria, everyone can subscribe to www-style
# [14:00] <takkaria> ah, excellent
# [14:00] * hsivonen is glad a lot of the ratholing moved to the font list
# [14:00] <hsivonen> glad in the sense that there's now less ratholing on www-style
# [14:00] <Lachy_> what's the status of the ratholing on www-font?
# [14:01] <jgraham> Lachy_: just treat the real and imaginary parts seperately and remember that i^2 === -1
# [14:01] * annevk2 has not been following www-font at all
# [14:01] * hsivonen neither
# [14:01] <Lachy_> I read on ars technica that there seems to be some support for the .webfont proposal, but it only mentioned support from the font foundries, not the browsers
# [14:02] <annevk2> i'm hoping we can just stick with what we have now
# [14:02] <Dashiva> I wish widgets had its own list too
# [14:02] <annevk2> i told howcome as much
# [14:02] <MadAtWork2> one thing about html/head/body.
# [14:02] <annevk2> but he's willing to compromise I think
# [14:02] <Lachy_> jgraham, ok, then that gives a total of 5.1 + 3i
# [14:02] <MadAtWork2> If I don't use them, does they get created in the DOM?
# [14:03] <MadAtWork2> Like, if I style body with css but do not define it in the html, does it still make sound?
# [14:04] <MadAtWork2> Hm, looks like it does, according to anne's blog.
# [14:04] <jgraham> MadAtWork2: Yes
# [14:04] <MadAtWork2> I candidate as resident idiot for idiot-proofing the specs.
# [14:05] <Dashiva> Someone will develop a better idiot before you're finished
# [14:06] <MadAtWork2> I'm sure.
# [14:08] <MadAtWork2> By the way, that site is using most of the bleeding edge techniques available right now, with a noticeable exception, no rgba/hsla for the hover effect.
# [14:08] <hsivonen> what's WARP in the widgets context?
# [14:10] * Parts: zcorpan (n=zcorpan@pat.se.opera.com)
# [14:21] <Philip`> It seems the current idea on www-font is to create a format that's a subset of EOT, without the encryption or compression or root strings, so it's compatible with old IE and doesn't have objectionable DRM-ish features and isn't TTF
# [14:22] * Quits: Lachy_ (n=Lachy@pat-tdc.opera.com) ("Leaving")
# [14:22] <Philip`> and then there's a hundred posts a day about it, adding no new information
# [14:23] <gsnedders|work> Sounds like public-html, really.
# [14:27] * Joins: Lachy_ (n=Lachy@pat-tdc.opera.com)
# [14:27] * Quits: Rik` (n=Rik`@pha75-2-81-57-187-57.fbx.proxad.net)
# [14:29] * Quits: Lachy_ (n=Lachy@pat-tdc.opera.com) (Client Quit)
# [14:31] * Joins: Lachy_ (n=Lachy@pat-tdc.opera.com)
# [14:32] * Quits: Maurice (i=copyman@5ED548D4.cable.ziggo.nl) (Read error: 60 (Operation timed out))
# [14:35] * Quits: mpt (n=mpt@canonical/launchpad/mpt) (Read error: 113 (No route to host))
# [14:35] <hsivonen> are there Process discussions on www-font?
# [14:35] * MadAtWork2 is now known as MadAtWork
# [14:36] <Philip`> hsivonen: I haven't seen any (in the past ~week)
# [14:40] <Philip`> but there are discussions about e.g. whether someone could hypothetically make an EOT file which happens to be processed correctly by EOT-Lite UAs except for ignoring root strings that limit the font to particular domains, and then the font would be rendered and someone could be sued for violating the DMCA
# [14:40] <Philip`> or something like that
# [14:40] * Quits: Lachy_ (n=Lachy@pat-tdc.opera.com) ("Leaving")
# [14:43] * Joins: BlurstOfTimes (n=blurstof@168.203.117.59)
# [14:43] * Joins: Maurice (i=copyman@5ED548D4.cable.ziggo.nl)
# [14:44] * Joins: pmuellr (n=pmuellr@nat/ibm/x-hmavzrasyfvlxpxe)
# [14:45] <annevk2> that's seems technical
# [14:46] <virtuelv> fuzzy booleans in canPlayType!?
# [14:47] <Philip`> DMCA discussions don't seem to be technical, because nobody understands the legal issues so they just make up claims
# [14:47] <annevk2> http://googleblog.blogspot.com/2009/08/innovation-in-video-on-web.html
# [14:47] <annevk2> virtuelv, not a boolean
# [14:47] <virtuelv> annevk2: I know that, technically, this is a DOMString
# [14:48] <virtuelv> but in reality, this is no different from a fuzzy boolean
# [14:48] <virtuelv> either way, I don't want to wrestle strings
# [14:48] <virtuelv> I'd much prefer an integer
# [14:49] <annevk2> also: http://www.on2.com/index.php?564
# [14:49] <annevk2> but it'll take another 4 months or so before we get the details
# [14:50] * Quits: webben (n=benh@nat/yahoo/x-bevwajvlwgicihjd) ("Lost terminal")
# [14:50] <annevk2> also: http://www.on2.com/index.php?594
# [14:51] <annevk2> if the "patent pool free" part is correct and Google "does the right thing" I guess it would mean that everyone can implement their H.264 impl RF?
# [14:51] <virtuelv> http://wiki.flyinglogic.com/wiki/Fuzzy_boolean
# [14:52] <annevk2> virtuelv, I don't really have a stake in the matter and philipj already argued it should just be a boolean, but apparently it's "too late now"
# [14:53] <nessy> annevk2: on2 has no part in h.264 afaik
# [14:53] <nessy> but it would be awesome if Google open sourced the On2 codecs!
# [14:54] * Joins: leeky (n=leeky@62.121.18.221)
# [14:54] <annevk2> nessy, ah, some line put me of, but it seems they have their own codecs indeed
# [14:54] * Quits: leeky (n=leeky@62.121.18.221) (Remote closed the connection)
# [14:54] <nessy> annevk2: Theora is a derivative of On2's older VP3 codec
# [14:55] <nessy> also, flv is On2's VP6
# [14:55] <nessy> (when it's not H.264)
# [14:55] <annevk2> ah, and now they're at VP8
# [14:55] <nessy> yup
# [14:56] <annevk2> it would make sense for Google to give it away
# [14:56] <annevk2> would make YouTube cheaper for instance
# [14:56] <annevk2> thanks nessy
# [14:57] <hsivonen> oh. interesting times regarding the codec issue!
# [14:57] <Lachy> that would be cool if Google makes VP5 through 8 royalty free :-)
# [14:57] <MadAtWork> That would imply that youtube will eventually ever get to use <video>
# [14:57] <nessy> yeah, totally!
# [14:58] <Lachy> I wonder if that's part of their motivation for acquiring On2
# [14:58] <nessy> that's my guess :)
# [14:58] <annevk2> I'm not sure how much clearer they can put it without giving away details: "we think that video compression technology should be a part of the web platform"
# [15:02] <Lachy> annevk2, that's still a long way from being a definitive statement about it
# [15:03] <annevk2> sure
# [15:07] <nessy> maybe with google wave and some other stuff that google is working on they realised they needed a free codec :)
# [15:08] * Joins: fearphage (n=fearphag@xbmc/user/fearphage)
# [15:08] <nessy> I just wonder if they will only open source some or all of them
# [15:12] <hsivonen> it seems that for the ecosystem, it would be best to push VP8 instead of a portfolio of old codecs
# [15:13] <annevk2> or VP9 if they're working on that already
# [15:13] <annevk2> but yeah, one format with multiple implementations is complexity enough :)
# [15:14] <hsivonen> I wonder if they lost hope with MPEG-LA and are now firmly on a non-MPEG-LA track or if this is like AOL acquiring Netscape
# [15:21] <hsivonen> TechCrunch says Nokia is a customer of On2. That's interesting if true.
# [15:22] <fearphage> so where do i direct comments about html5 again? mailing list only?
# [15:23] <annevk2> bug database
# [15:23] <annevk2> fyi, canPlayType was discussed recently and is highly unlikely to change
# [15:23] <annevk2> (also mailing list of course)
# [15:24] <virtuelv> isn't on2 the historical origin of theora?
# [15:24] <annevk2> yes, see above
# [15:26] <fearphage> annevk2: it would make more sense if they were constants like XMLHttpRequest.OPENED and XpathResult and others
# [15:26] * Joins: mpt (n=mpt@canonical/launchpad/mpt)
# [15:26] <fearphage> strings like "probably" nad "maybe" just come accross very poorly
# [15:30] <annevk2> http://twitter.com/annevk/statuses/3144668418
# [15:31] * Quits: MikeSmith (n=MikeSmit@71-218-60-163.hlrn.qwest.net) ("Tomorrow to fresh woods, and pastures new.")
# [15:35] * Joins: kristallpirat (n=kristall@c-base/crew/kristall)
# [15:43] * Joins: paul_irish (n=paul_iri@12.33.239.250)
# [15:44] <jgraham> virtuelv: I think there is some agreement that canPlayType was a mistake. Sadly it was a mistake that was caught after major browsers shipped it
# [15:45] <jgraham> And having dual apis this ealy in the game is seen as worse
# [15:45] <jgraham> (e.g. a boolean mightPlayType)
# [15:49] <virtuelv> jgraham: I figured as much
# [15:53] * Quits: nessy (n=nessy@124-168-244-132.dyn.iinet.net.au) ("This computer has gone to sleep")
# [16:00] * Joins: dbaron (n=dbaron@c-98-234-51-190.hsd1.ca.comcast.net)
# [16:09] * Joins: dbaron_ (n=dbaron@c-98-234-51-190.hsd1.ca.comcast.net)
# [16:11] * Quits: dbaron (n=dbaron@c-98-234-51-190.hsd1.ca.comcast.net) (Read error: 60 (Operation timed out))
# [16:11] * Joins: Xanthir (n=chatzill@99-35-179-251.lightspeed.hstntx.sbcglobal.net)
# [16:12] <Xanthir> So, I'm building a thin feature-testing library for myself for html5. Safari and Opera both support <input type=range>, but neither of them properly validate input. Should I mark them as supporting or not?
# [16:13] <Xanthir> I'm mainly going to use this to decide whether or not to feed the user a jQuery library to simulate the control.
# [16:13] <virtuelv> Xanthir: define "properly validate"
# [16:13] * Quits: dbaron_ (n=dbaron@c-98-234-51-190.hsd1.ca.comcast.net) (Read error: 113 (No route to host))
# [16:14] <Xanthir> Given an invalid @value, they'll return "" (should be halfway between min and max). Given an out-of-bounds @value, they return the value unchanged.
# [16:15] <Xanthir> The latter should be clamped to [min,max].
# [16:15] <virtuelv> filed bugs?
# [16:15] * Joins: dbaron (n=dbaron@98.234.51.190)
# [16:15] <jgraham> Xanthir: If they are both wrong in the same way maybe the spec should be changed
# [16:15] * jgraham hasn't looked at this in detail
# [16:16] <Xanthir> virtuelv: Not yet - I was just working on this last night for myself.
# [16:16] <virtuelv> neither have I, but the behavior Xanthir is suggesting seems more sensible
# [16:16] <Xanthir> jgraham: possibly. The behavior they exhibit with an invalid value is consistent with how other sanitizing types work (like the date and time types).
# [16:17] <Xanthir> But I'm definitely not happy with how they handle an overflow/underflow value, nor a non-provided value.
# [16:17] <Xanthir> Oh, and sorry if I was implying that the behavior I'm wanting is my own invention. It's from the spec, virtuelv.
# [16:18] <virtuelv> Xanthir: no, I'm not seeing that implication
# [16:18] <Xanthir> Ah, k. Good. ^_^
# [16:18] <virtuelv> I'm just saying that the spec seems to make sense in this case
# [16:19] <virtuelv> the spec is missing error handling when setting HTMLInputElement.value, though
# [16:19] * Joins: dglazkov (n=dglazkov@c-67-188-0-62.hsd1.ca.comcast.net)
# [16:19] <Xanthir> Nod. It guarantees that the return value of an <input type=range> is a number within the range at all times.
# [16:20] <Xanthir> Hm, possibly, but I was setting the value on initialization. I was using jQuery with $("<input type=range value=foo>").val()
# [16:20] <Xanthir> Which, by spec, should return 50 (default min/max is 0/100, so midway between them is 50).
# [16:23] <jgraham> Xanthir: I agree that not clamping seems odd.
# [16:23] <jgraham> The halfway between min and max seems totally arbitary
# [16:23] * Philip` points out http://philip.html5.org/data/charsets.html in relation to the charset discussion
# [16:25] <Xanthir> It's just a way of defining the 'default' value. Safari will start a range input on the halfway mark.
# [16:25] <Xanthir> But it still returns "" if you didn't set a @value, and haven't touched it yet.
# [16:26] <Xanthir> Oh wait, that's lies. It correctly returns 50 if you just put a range input onto the page and immediately ask it for the value.
# [16:26] <Xanthir> But it still sanitizes an invalid value to "".
# [16:27] <Xanthir> Ah, no, I got it. Safari will return 50 for the value of a static <input type=range> in the page. It returns "" for a freshly-created one that hasn't been inserted into the document yet. Not sure of exact details on how this works.
# [16:29] <Xanthir> I guess I'll go file some bugs today.
# [16:30] * Joins: zcorpan_ (n=zcorpan@c83-252-201-53.bredband.comhem.se)
# [16:32] * Quits: mpt (n=mpt@canonical/launchpad/mpt) (Remote closed the connection)
# [16:33] * Joins: mpt (n=mpt@canonical/launchpad/mpt)
# [16:38] <hsivonen> http://lists.xiph.org/pipermail/theora-dev/2009-August/003959.html
# [16:42] * Joins: svl (n=me@ip565744a7.direct-adsl.nl)
# [16:47] * Quits: zcorpan_ (n=zcorpan@c83-252-201-53.bredband.comhem.se) (Read error: 110 (Connection timed out))
# [16:48] * Joins: shepazu (n=schepers@75-149-110-74-Chattanooga.hfc.comcastbusiness.net)
# [16:50] * Quits: heycam (n=cam@124-168-112-102.dyn.iinet.net.au) (Connection timed out)
# [16:53] * Quits: maikmerten (n=merten@ls5dhcp196.cs.uni-dortmund.de) (Remote closed the connection)
# [16:58] * Quits: dglazkov (n=dglazkov@c-67-188-0-62.hsd1.ca.comcast.net)
# [17:04] * Joins: foolip (n=philip@pat.se.opera.com)
# [17:06] * Joins: heycam (n=cam@124-168-112-102.dyn.iinet.net.au)
# [17:22] * Joins: dglazkov (n=dglazkov@nat/google/session)
# [17:41] * Joins: mpilgrim (n=mark@ip-64-139-15-210.dsl.sca.megapath.net)
# [17:56] * Joins: Yudai (n=Yudai@p929c7b.kngwnt01.ap.so-net.ne.jp)
# [18:01] * Joins: weinig (n=weinig@nat/apple/session)
# [18:04] * Quits: gavin (n=gavin@firefox/developer/gavin) (Read error: 110 (Connection timed out))
# [18:04] * Joins: gavin (n=gavin@firefox/developer/gavin)
# [18:04] * Joins: Sidnicious (n=Sidney@unaffiliated/sidnicious)
# [18:09] * Quits: gunderwonder (n=gunderwo@garage.upstruct.com)
# [18:14] * Quits: poe (n=poe@unaffiliated/xerox)
# [18:22] <Sidnicious> It appears that storage events for changes to sessionStorage don't fire at all in Firefox and only fire in the sending window in Safari
# [18:22] <Sidnicious> What's expected behavior here?
# [18:23] <Sidnicious> Example page: http://dev.deeptechinc.com/sidney/messaging/sender.html
# [18:24] <Sidnicious> Changes to localStorage generate storage events on the child iframe and a spawned window in both browsers.
# [18:29] <annevk2> sessionStorage should work for iframes
# [18:29] <annevk2> but not separate top-level windows (or top-level browsing contexts, as they're called)
# [18:30] * Joins: othermaciej (n=mjs@c-69-181-42-237.hsd1.ca.comcast.net)
# [18:34] * Quits: mpilgrim (n=mark@ip-64-139-15-210.dsl.sca.megapath.net) (Read error: 110 (Connection timed out))
# [18:36] <ezyang> Whoo XML parsing DOS.
# [18:36] <ezyang> Which reminds me of the recent infinite loop fix in the HTML5 spec :-)
# [18:39] <Philip`> Are you referencing any particular XML parsing DOS?
# [18:41] <Sidnicious> annevk2: Thanks, I just looked through the spec again and it's making sense. I'm trying to accomplish inter-window communication in situations where the windows may not have references to each other, and postMessage can't do that.
# [18:42] <Sidnicious> It looks like I'm not the only one who wants this... I just spoke to someone in ##javascript who's using localStorage to communicate between open instances of his site, and this poster had a similar question: http://stackoverflow.com/questions/1100336/sending-a-message-to-all-open-windows-tabs-using-javascript
# [18:45] <annevk2> you want localStorage for that
# [18:45] <annevk2> although we probably should've introduced a separate API rather than putting it on top of the storage API :/
# [18:45] <Sidnicious> using localStorage works, but it feels like this problem should be solved with cross-window messaging rather than web storage.
# [18:45] <annevk2> right
# [18:45] <Sidnicious> :)
# [18:46] <annevk2> you're not the first to point out this issue, though it prolly bears repeating: whatwg@whatwg.org
# [18:46] <weinig> you want a window.broadcast("message") type thing
# [18:46] <annevk2> yup
# [18:46] <weinig> interesting
# [18:48] <Sidnicious> annevk2: I'll write up a message.
# [18:49] <annevk2> ta
# [18:53] * Joins: cying (n=cying@70.90.171.153)
# [18:53] * abarth|zzz is now known as abarth
# [18:53] * Quits: abarth (n=abarth@c-98-210-108-185.hsd1.ca.comcast.net)
# [18:56] * Parts: MadAtWork (n=not_@83.139.194.2)
# [18:58] * Joins: gsnedders (n=gsnedder@c83-252-197-103.bredband.comhem.se)
# [18:59] * Joins: hobertoAtWork (n=hobertoa@gw1.mcgraw-hill.com)
# [19:15] * Quits: Yudai (n=Yudai@p929c7b.kngwnt01.ap.so-net.ne.jp) (Read error: 110 (Connection timed out))
# [19:16] * Joins: Yudai (n=Yudai@p6ea250.kngwnt01.ap.so-net.ne.jp)
# [19:17] * Joins: maikmerten (n=maikmert@Z86dc.z.pppool.de)
# [19:17] * Quits: mat_t (n=mattomas@89.127.177.75) ("This computer has gone to sleep")
# [19:20] * Philip` tries to work out what language code to use for Chilean Spanish
# [19:20] * Quits: gavin (n=gavin@firefox/developer/gavin) (Read error: 60 (Operation timed out))
# [19:20] <Philip`> Is something like es-CL valid?
# [19:20] * Joins: gavin (n=gavin@firefox/developer/gavin)
# [19:20] <Philip`> or maybe it's close to es-419?
# [19:20] <Philip`> Languages are too hard :-(
# [19:20] <gsnedders> Am I right at thinking per ES5 foo = [10, 2]; foo.sort() should result in [10,2]?
# [19:20] * gsnedders must be reading the spec wrong, somehow…
# [19:26] * Joins: dave_levin (n=dave_lev@72.14.224.1)
# [19:27] * Joins: sbublava (n=stephan@77.119.68.106.wireless.dyn.drei.com)
# [19:35] * Parts: adactio (n=adactio@host86-132-125-223.range86-132.btcentralplus.com)
# [19:41] * Quits: kristallpirat (n=kristall@c-base/crew/kristall) ("Wünsche weiterhin guten Flug")
# [19:42] * Joins: abarth (n=abarth@adsl-71-132-207-33.dsl.pltn13.pacbell.net)
# [19:48] * Quits: myakura (n=myakura@p3137-ipbf7310marunouchi.tokyo.ocn.ne.jp) ("Leaving...")
# [19:54] <Lachy> gsnedders, although that is unintuitive, that does seem to be the correct result
# [19:54] <gsnedders> WTF?
# [19:54] * Quits: dbaron (n=dbaron@98.234.51.190) ("8403864 bytes have been tenured, next gc will be global.")
# [19:55] <Lachy> because unless you pass a compareFunction, the values are stringified and sorted, instead of being sorted numerically
# [19:55] <gsnedders> Right, I get that from reading the spec. But it seems whacky as hell.
# [19:55] <Lachy> yeah, I know
# [19:56] * Quits: foolip (n=philip@pat.se.opera.com) (Read error: 110 (Connection timed out))
# [19:57] <Philip`> What do you expect [1, 2, "three", 4].sort() to do, if it doesn't stringify everything?
# [19:57] <ezyang> I think Python sorts strings either before/after
# [19:58] * gsnedders thinks that it is undefined in Python
# [19:58] <ezyang> No, it's definitely defined
# [19:58] <gsnedders> (But CPython goes for after)
# [19:59] <ezyang> I know this has resulted in some hilarious bugs
# [19:59] <Philip`> Python says 1 < "x"
# [19:59] <gsnedders> Philip`: It's undefined.
# [19:59] <gsnedders> *ezyang
# [19:59] <gsnedders> ezyang: "Otherwise, objects of different types always compare unequal, and are ordered consistently but arbitrarily. "
# [19:59] <ezyang> Aha
# [19:59] <ezyang> "Consistency! Consistency is all I ask!"
# [20:00] <gsnedders> ezyang: So, yes, it is undefined. :P
# [20:00] <Philip`> Hmm, Python doesn't seem very consistent at sorting nans
# [20:00] <ezyang> Ok.
# [20:00] * Quits: othermaciej (n=mjs@c-69-181-42-237.hsd1.ca.comcast.net)
# [20:01] <Lachy> Philip`, people shouldn't be silly enough to mix numbers and strings in the same array, so that will never happen!
# [20:01] * Joins: webben (n=benh@nat/yahoo/x-a13c2451db14dc3f)
# [20:01] * Quits: abarth (n=abarth@adsl-71-132-207-33.dsl.pltn13.pacbell.net)
# [20:04] <hober> nobody sorts NaNs well, assuming your sorting function uses < and >
# [20:06] <Philip`> You could be consistent and stick them at the end or something
# [20:06] <Philip`> instead of saying sorted([2, nan, 1]) == [2, nan, 1]
# [20:18] * gsnedders wonders what to do having done so badly in exams
# [20:18] * Joins: abarth (n=abarth@dhcp-45-156.EECS.Berkeley.EDU)
# [20:22] <Lachy> gsnedders, how badly?
# [20:22] <Lachy> did you fail?
# [20:22] * Joins: dbaron (n=dbaron@nat/mozilla/x-876033018060112e)
# [20:22] <gsnedders> I passed everything, just (with the exception of computing which I did no work for and did very well in).
# [20:23] <Lachy> are you planning to go to uni?
# [20:23] <gsnedders> I won't get in to almost any university outwith Scotland with the grades I got, though (and only there because the traditional starting age is a year younger, so what I did last year is relevant, not this)
# [20:25] <Lachy> you may not need to go to uni, depending on what sort of work you want to do
# [20:26] <Philip`> You could go to a university and be a janitor
# [20:26] * Philip` probably isn't helping much
# [20:26] * gsnedders takes out trout
# [20:27] <Lachy> I hear being a dole bludger is quite an exciting life style
# [20:28] * Joins: mat_t (n=mattomas@89.127.177.74)
# [20:28] * Lachy picks up a salmon in prepration for a fish fight with gsnedders
# [20:32] <gsnedders> Lachy: What sort of work do I want to do? :P
# [20:34] <Dashiva> gsnedders: es3 had the same sort behavior
# [20:34] <gsnedders> Dashiva: I don't care about ES3 because nobody now is aiming to impl it :P
# [20:35] <Dashiva> And I'm saying the Array.sort train left years ago :P
# [20:35] <gsnedders> Well I was hardly expecting it to change :P
# [20:36] * Joins: sebmarkbage (n=miranda@h-6-72.A146.priv.bahnhof.se)
# [20:39] <Lachy> gsnedders, you want to be a fireman.
# [20:42] * Joins: gunderwonder (n=gunderwo@172.80-202-84.nextgentel.com)
# [20:43] * Quits: gunderwonder (n=gunderwo@172.80-202-84.nextgentel.com) (Client Quit)
# [20:43] * Joins: gunderwonder (n=gunderwo@172.80-202-84.nextgentel.com)
# [20:45] * Quits: mat_t (n=mattomas@89.127.177.74) ("This computer has gone to sleep")
# [20:46] <gsnedders> Lachy: I disagree.
# [20:49] <Lachy> gsnedders, that's a poor attempt at refuting my statement.
# [20:50] <gsnedders> Lachy: I do not have the courage or physical ability to be confronted by fire.
# [20:51] <Lachy> ok, let me have another try...
# [20:53] <Lachy> how about being a hair dresser?
# [20:53] <gsnedders> no
# [20:54] <Lachy> drug trafficker?
# [20:54] <Philip`> You could become a Polish builder
# [20:55] <Lachy> Philip`, a small technical problem with that is that it would require gsnedders to be Polish.
# [20:55] * Quits: annevk2 (n=annevk@5355732C.cable.casema.nl) (Read error: 110 (Connection timed out))
# [20:56] * Quits: dave_levin (n=dave_lev@72.14.224.1)
# [20:56] * Joins: ttepasse (n=ttepas--@p5B013602.dip.t-dialin.net)
# [20:57] * Quits: annevk42 (n=annevk@5355732C.cable.casema.nl) (Read error: 110 (Connection timed out))
# [20:58] <Dashiva> gsnedders: You could be a spec writer roadie
# [20:59] <gsnedders> LOL
# [20:59] * Joins: annevk42 (n=annevk@5355732C.cable.casema.nl)
# [21:00] * Joins: othermaciej (n=mjs@17.203.15.225)
# [21:01] * Quits: Chris_Wilson (n=cwilso@nat/microsoft/x-burdhghsxrxcnjgy) (Read error: 60 (Operation timed out))
# [21:01] * Joins: jwalden (n=waldo@nat/mozilla/x-4fb52bdf589180ff)
# [21:02] * Quits: pmuellr (n=pmuellr@nat/ibm/x-hmavzrasyfvlxpxe)
# [21:02] * Joins: ChrisWilson (n=cwilso@nat/microsoft/x-016f42e79386a66d)
# [21:05] <Dashiva> Is that a spammer submitting those new bugs?
# [21:05] <Lachy> yes
# [21:06] <Philip`> No
# [21:06] <Lachy> Philip`, clearly, repeatedly posting links to abu---wisam.ahlamontada.net without any explanation constitues spam.
# [21:07] <Philip`> I have no argument to back up my claim, so you win
# [21:19] * Joins: jennb (n=jennb@72.14.227.1)
# [21:23] <beowulf> what happened to the idea of UAs auto generating table summaries?
# [21:23] <Dashiva> It's still out there
# [21:23] <gsnedders> Nobody proved it could be done
# [21:24] <Dashiva> If you're paranoid, go ahead and assume that the summary cabal smothered it to avoid competition :P
# [21:24] <beowulf> when i looked at the example summary given on the wiki i thought it might be possible with js, but perhaps i over estimated myself
# [21:24] <Lachy> beowulf, which summary?
# [21:24] <Lachy> Are you referring to the example in the esw wiki that makes no sense if you think about it?
# [21:25] <beowulf> Lachy: yes, i was searching for the link there
# [21:25] <beowulf> everything but the "this table presents" seems possible
# [21:25] <jgraham> gsnedders: How bad were our grades?
# [21:25] <beowulf> Lachy: what makes no sense?
# [21:25] <jgraham> *your
# [21:26] <gsnedders> jgraham: ACCC
# [21:26] <Lachy> summary="This table presents traveling expenses. Rows contain destinations, traveling dates, and grand total. Columns contain expense category and total. The first column contains merged table cells."
# [21:26] <jgraham> Hmm. Not great I admit. You can surely get into some univerity with that if you want
# [21:26] <gsnedders> jgraham: I doubt anywhere in England/Wales
# [21:26] <Lachy> How can rows contain different content from the columns, in the way described?
# [21:26] <beowulf> Lachy: i assumed it was from a real table, are you saying if i try and reverse that into a table it won't work?
# [21:27] <Lachy> Or, better yet, can you conceive of a table that accurately fits that description?
# [21:27] * beowulf has a go
# [21:27] <Dashiva> Lachy: Isn't the summary just listing each row/column's heading?
# [21:27] <jgraham> gsnedders: You will get in somewhere. Possibly not somewhere great unless there is somewhere tht really likes you
# [21:27] <Lachy> I look forward to your solution. I especially curious about how the merged cells in the first column fit in
# [21:28] <Lachy> Dashiva, but then the table wouldn't actually make any sense
# [21:28] * beowulf senses a fail coming
# [21:28] <Dashiva> Lachy: That wouldn't be the first time!
# [21:28] <gsnedders> jgraham: You think there's any point in applying to places in England?
# [21:28] <jgraham> gsnedders: yes
# [21:29] <jgraham> ACCC will get you in to do CS somewhere
# [21:29] <beowulf> you know I asked for one of the pro arguments to be changed on that page, i thought it was stupid to suggest vim users might be confused by the syntax highlighter
# [21:29] <Lachy> Dashiva, beowulf, in this case, since we can't see the table, we're in effectively the same position as a blind user with assistive technology would be in prior to manually scanning the table
# [21:29] <gsnedders> jgraham: Where, though?
# [21:29] <Dashiva> Well, I'm wondering if the summary uses AT idioms that we just don't know
# [21:30] <jgraham> gsnedders: Dunno exactly
# [21:30] * Quits: raek (n=raek@h-148-88.A163.priv.bahnhof.se) (Read error: 54 (Connection reset by peer))
# [21:30] * Joins: raek (n=raek@h-148-88.A163.priv.bahnhof.se)
# [21:30] <gsnedders> jgraham: I think half the problem is CS is quite badly taught in general, or is just software engineering, which means probably a lot of the places that would take me with such grades aren't overly good
# [21:31] <Dashiva> aha
# [21:31] <takkaria> gsnedders: if you write a good personal statement pointing out the public software projects you're involved in and the fact you've worked for Opera, I imagine that will work significantly in your favour
# [21:31] <jgraham> gsnedders: I agree that is more of a problem
# [21:32] <jgraham> You do have an A in CS AH though
# [21:32] <jgraham> And all the pratical experience
# [21:32] <Lachy> this is what you get when you follow the description literally, although I can't figure out how the merged cells in the first column are supposed to work
# [21:32] <Lachy> http://html5.lachy.id.au/output?data=%3C!DOCTYPE+html%3E%0D%0A%3Ctable+border%3D1%3E%0D%0A%3Ctr%3E%3Cth%3E%3Cth%3EExpense+Category%3Cth%3ETotal%0D%0A%3Ctr%3E%3Cth%3EDestinations%3Ctd%3Ex%3Ctd%3Ex%0D%0A%3Ctr%3E%3Cth%3ETravel+Dates%3Ctd%3Ex%3Ctd%3Ex%0D%0A%3Ctr%3E%3Cth%3EGrand+Total%3Ctd%3Ex%3Ctd%3Ex%0D%0A%3C%2Ftable%3E&type=text%2Fhtml%3B+charset%3DUTF-8
# [21:33] <gsnedders> jgraham: But having a C in maths is going to be a brick wall in a lot of ways
# [21:34] <jgraham> Lachy: I assume that Expense Category would be split into multiple categories
# [21:34] <jgraham> The rows make no sense to me though
# [21:34] <gsnedders> jgraham, takkaria: it was my maths that was explicitly cited in terms of rejection for Cam last year, though
# [21:34] <Dashiva> Grand total is probably a sum of totals, right?
# [21:35] <jgraham> gsnedders: Any posibility to retake maths?
# [21:35] <Dashiva> So the concept is a set of destinations, each destination has a date and a grand total, as well as a set of categories with (sub)totals
# [21:35] <jgraham> You could make a reasonable case to have over-extended yourself
# [21:35] <gsnedders> jgraham: Practically no
# [21:35] * Lachy has to leave for 30 minutes. If you come up with a reasonable table that fits, I'll look when I get back
# [21:35] * Quits: gunderwonder (n=gunderwo@172.80-202-84.nextgentel.com)
# [21:36] <gsnedders> jgraham: What seems like a far more sensible suggestion is to do something like Open University certificate in maths, which would effectively amount to retaking AH and doing a level above
# [21:36] * Joins: roc_ (n=roc@121-72-178-80.dsl.telstraclear.net)
# [21:38] * Joins: ojan (n=ojan@72.14.228.129)
# [21:38] * Quits: roc (n=roc@121-72-186-57.dsl.telstraclear.net) (Connection timed out)
# [21:38] * Quits: Sidnicious (n=Sidney@unaffiliated/sidnicious)
# [21:39] <gsnedders> jgraham: I don't think it was purely over-pushing myself that did it, really, more absence in Nov/Dec, spending my time quite unproductively trying to do well in prelims (doing things like making sure I could write about English author I had no intention of writing about in final exam), more or less giving up having done badly in them nevertheless, then having to do all the coursework in too little time, more or less not doing class
# [21:39] <gsnedders> work…
# [21:40] <gsnedders> The fact it was the first time I had actually really had to _work_ to do well at school, as opposed to just turn up to the class and get A/B without really doing anything more, rather screwed me up too.
# [21:41] <Dashiva> Lachy: http://dashiva.net/misc/summary.html
# [21:42] <jgraham> fwiw I would expect the table described by that summary to actually be laid out like http://software.hixie.ch/utilities/js/live-dom-viewer/saved/192 even though that doesn't match the summary at all
# [21:43] * Quits: shepazu (n=schepers@75-149-110-74-Chattanooga.hfc.comcastbusiness.net)
# [21:43] <jgraham> Dashiva: That doesn't seem to match the summary. At least my understanding of it
# [21:43] <Dashiva> What part doesn't match?
# [21:44] <jgraham> Although I think I have an idea
# [21:45] * gsnedders opens the spec and has Opera crash
# [21:45] <gsnedders> We really ought to fix that.
# [21:45] <Dashiva> There's a bug
# [21:45] <Dashiva> Or several
# [21:45] <gsnedders> I know.
# [21:46] <gsnedders> Several.
# [21:46] <jgraham> Dashiva: It doesn't have the seperation between rows and columns implied by the summary. Nor does it have the grand total
# [21:46] <Dashiva> It does
# [21:46] <gsnedders> Dashiva: The problem is I always fail at reproducing them when I want to, and always succeed when I don't want to :)
# [21:46] <Dashiva> Third subcolumn in the first column
# [21:47] <jgraham> Oh yeah sorry that was such an insane place to put that I don't notice
# [21:47] <jgraham> it still doesn't have the division between columns and rows implied by the summary
# [21:48] <jgraham> Although I think your table might be as close to what is meant as possible
# [21:48] <Dashiva> Actually
# [21:49] <Dashiva> I see there's a different reading possible, let's see
# [21:50] <beowulf> i got a very stupid table
# [21:50] <Dashiva> beowulf: It probably is a very stupid table
# [21:50] <Dashiva> Traveling expenses should not require a complex table in the first place :P
# [21:51] <beowulf> http://software.hixie.ch/utilities/js/live-dom-viewer/?%3C!DOCTYPE%20html%3E%0A%3Ctitle%3EAuto%20Summary%3C%2Ftitle%3E%0A%0A%3Ctable%20summary%3D%22%22%3E%0A%20%20%3Cthead%3E%0A%20%20%20%20%3Ctr%3E%0A%20%20%20%20%20%20%3Ctd%20colspan%3D2%3E%3C%2Ftd%3E%0A%20%20%20%20%20%20%3Cth%3EDestination%3C%2Fth%3E%0A%20%20%20%20%20%20%3Cth%3EDates%3C%2Fth%3E%0A%20%20%20%20%20%20%3Cth%20colspan%3D2%3EGrand%20Total%3C%2Fth%3E%0A%20%20%20%20%3C%2Ftr%3E%0A%20%20%3C%2Fthea
# [21:51] <beowulf> ouch, sorry!
# [21:52] <Dashiva> Short url?
# [21:52] <jgraham> beowulf: Use the save feature
# [21:53] <beowulf> http://software.hixie.ch/utilities/js/live-dom-viewer/saved/193 # dumb table
# [21:53] * Joins: annevk2 (n=annevk@5355732C.cable.casema.nl)
# [21:53] <beowulf> my heart goes out to any blind person who has to work out a table from that
# [21:53] <Dashiva> It would help if you add more data, so people can tell the structure :)
# [21:55] <beowulf> actually, mine is just plain wrong, i'll try again
# [21:56] <Hixie> Lachy: what you said about intranet pages is basically exactly what i said in my first post on that thread
# [21:57] <jgraham> There are several possible conclusions from this: a) this is an unusually bad example of a table summary b) navigating tables blind is so hard that even really bad table summaries are helpful c) the useulness of table summaries has been ven overstated since even when used correctly they are not that helpful
# [21:57] <jgraham> Those conclusions are not mutually exclusive
# [21:57] <gsnedders> Hixie: Is there any easy way to work out whether a cell is a header cell?
# [21:58] <beowulf> th?
# [21:58] <Hixie> you mean other than seeing if its tag name is "th"?
# [21:58] * Joins: tndH (n=Rob@cpc2-leed18-0-0-cust427.leed.cable.ntl.com)
# [21:58] <Dashiva> jgraham: d) the table is non-euclidean
# [21:58] * Quits: jennb (n=jennb@72.14.227.1) (Read error: 60 (Operation timed out))
# [22:01] * Joins: dave_levin (n=dave_lev@72.14.227.1)
# [22:02] <gsnedders> Hixie: Right
# [22:02] <gsnedders> Hixie: Oh, to be more exact: is there any easy way to work out whether a cell is a header cell, and for what dimensions it is a header well?
# [22:02] * gsnedders realized he forgot the vital latter part
# [22:03] <Hixie> i don't understand your question
# [22:03] <gsnedders> I mean whether it is a header for a column or a row.
# [22:03] <gsnedders> or both
# [22:03] <beowulf> http://software.hixie.ch/utilities/js/live-dom-viewer/saved/194 # 2nd try
# [22:04] <Hixie> gsnedders: i haven't work out which parts of the spec's algorithm you need to apply to a single cell to work that out, no
# [22:04] <gsnedders> Hixie: ok
# [22:04] <Hixie> gsnedders: at a minimum you'll need to check that there are no cells that specify this on in the headers="" attribute
# [22:04] <gsnedders> Yeah, that's what I was thinking. Is there anything else?
# [22:04] <Hixie> gsnedders: and examine scope="", and if that's absent, the surrounding elements, iirc
# [22:05] <gsnedders> @scope?
# [22:05] <Hixie> scope="".
# [22:05] <gsnedders> What does it do, though?
# [22:05] * Hixie points gsnedders at the spec :-)
# [22:05] <gsnedders> Where in the spec? :P
# [22:05] <gsnedders> Under th?
# [22:06] * Hixie introduces gsnedders to the "find in page" functionality of his browser
# [22:06] * gsnedders smacks Hixie
# [22:06] <Hixie> beowulf: that table is a great example of why table explanations need to be visible to everyone!
# [22:06] <gsnedders> Hixie: Why do I need "find in page" when I have you? :P
# [22:07] <Hixie> because i'm doing other things
# [22:07] <Lachy> jgraham, yours looks like a reasonable attempt at making some sense out of the summary, though I don't think it fits it entirely accurately
# [22:07] <Dashiva> beowulf: That's obviously wrong
# [22:07] <gsnedders> Hixie: I guess I just need to work out "Forming relationships between data cells and header cells"?
# [22:07] * Joins: jennb (n=jennb@72.14.227.1)
# [22:07] <Lachy> however, I expect it's just an unusually bad example since it appears to have been made up without actually having a table to create it from
# [22:07] <annevk2> Hixie, is it intentional that Web Sockets is not that strict on redirects?
# [22:07] <beowulf> Dashiva: which part was obvious?
# [22:08] <Lachy> Hixie, I don't see how what you said is the same as what I said
# [22:08] <annevk2> e.g. "blabha 301 fdjksf" also redirects
# [22:08] <annevk2> also, there's no difference in semantics between 301-307 within Web Sockets
# [22:08] * Quits: abarth (n=abarth@dhcp-45-156.EECS.Berkeley.EDU)
# [22:08] * Joins: jennb_ (n=jennb@72.14.227.1)
# [22:08] <Dashiva> beowulf: Well, maybe it's just the lack of data
# [22:09] <Dashiva> Try adding a few more trips and categories so I can verify
# [22:09] * Joins: abarth (n=abarth@dhcp-45-156.EECS.Berkeley.EDU)
# [22:09] <Hixie> Lachy: that we should take intranets into account, but that we can't have intranet support as a goal
# [22:10] <Hixie> annevk2: yes
# [22:12] * Joins: franksalim (n=frank@adsl-75-61-85-210.dsl.pltn13.sbcglobal.net)
# [22:12] <Dashiva> We need an oracle to help in the summary guesswork
# [22:12] <Lachy> I don't think that's quite what I said
# [22:13] <Lachy> well, the first part is. The second part isn't
# [22:13] * Parts: ojan (n=ojan@72.14.228.129)
# [22:13] * Quits: archtech (n=sv@83.228.56.37)
# [22:14] <Lachy> but reading your message, it sounds like you're saying that we don't need to take intranets into account because they don't have to use open standards
# [22:14] <beowulf> Dashiva: http://software.hixie.ch/utilities/js/live-dom-viewer/saved/195
# [22:14] <beowulf> Dashiva: fwiw, saying i was obviously wrong made me laugh :)
# [22:14] <Hixie> Lachy: i had hoped i had explained that my position was more subtle than that
# [22:14] <Hixie> oh well
# [22:15] <beowulf> like, the summary makes it crysal clear what the table should be!
# [22:16] * Quits: jennb (n=jennb@72.14.227.1) (Read error: 104 (Connection reset by peer))
# [22:16] <Hixie> annevk2: does html5 really require media="" to be treated as "all"?
# [22:16] * Joins: jennb__ (n=jennb@72.14.227.1)
# [22:16] <Hixie> annevk2: i thought it just deferred to MQ completely
# [22:17] <Xanthir> beowulf: Wow, hm, that's almost right. The first column doesn't have merged cells, though.
# [22:18] <beowulf> fwiw, i would guess that a lot of web authors have used this article as a tutorial on tables http://veerle.duoh.com/blog/comments/a_css_styled_table_version_2/
# [22:18] * Quits: mpt (n=mpt@canonical/launchpad/mpt) (Read error: 113 (No route to host))
# [22:18] <Xanthir> Fwiw, I'm with the "they just made up a summary without looking at a table" camp.
# [22:18] <beowulf> Xanthir: the first col does?
# [22:19] <Dashiva> beowulf: The grand total isn't grand, summing only one item
# [22:19] <jgraham> Lachy: Do you know this summary to be fiction?
# [22:19] <Xanthir> Oh, hrm, I suppose it does have merged cells. The blank cell in the upper corner is merged across two cells. ^_^
# [22:19] <beowulf> Dashiva: yeah i know, i can't think how you have a row with grand total though withuot it being, well, a sub-total
# [22:20] <Dashiva> The total columns are the subtotals, as I understand it
# [22:20] <beowulf> the summary says rows contain a grand total
# [22:20] <beowulf> so it must be a grand total for the row
# [22:21] <beowulf> like, if the total mileage is 8 miles at 0.50 a mile, the grand total for that row is 4
# [22:22] <Dashiva> But the rows don't (explicitly) contain any numbers to total
# [22:23] <jgraham> gsnedders: Maybe an OU maths course does make sense. I'm not really sure
# [22:23] <beowulf> Dashiva: they do if you consider rowspan, no?
# [22:24] <annevk2> Hixie, yeah, but we might change that bit of MQ
# [22:24] <annevk2> Hixie, so HTML5 would have to require that for compat
# [22:24] <beowulf> i should add scope in there i suppose
# [22:25] <Dashiva> beowulf: Sure, that's what I assumed too. But even then you have a grand total of one item (e.g. your food row)
# [22:25] * Quits: maikmerten (n=maikmert@Z86dc.z.pppool.de) (Read error: 54 (Connection reset by peer))
# [22:25] <Xanthir> Assuming a table with Euclidean geometry, like the nice one over on the link beowulf posted, we could easily imagine what the automated summary would be.
# [22:25] <Xanthir> And honestly, it sounds relatively decent.
# [22:26] <Dashiva> Did anyone say what was wrong with my table? There was something, but I can't find it in scrollback.
# [22:26] <beowulf> you know, now that i've done that, even if a ua auto summarised a table in that way i find it hard to imagine it being useful to a blind person, less a person with learnign difficulties
# [22:27] * Quits: abarth (n=abarth@dhcp-45-156.EECS.Berkeley.EDU)
# [22:27] <Xanthir> "The table has four columns: Webhosting Home, Webhosting Home Plus, Webhosting Business, Webhosting Business Plus. It has six rows: Data Usage, MySQL Databases" etc.
# [22:27] * Joins: poe (n=poe@unaffiliated/xerox)
# [22:28] <Xanthir> If I was blind, that'd work for me.
# [22:28] <beowulf> Xanthir: that sounds more useful
# [22:28] <Xanthir> That's a basic table, nothing complex with merged cells or anything, but still.
# [22:28] <beowulf> the example summary is not a good one
# [22:28] <beowulf> very far from good in fact
# [22:29] * Joins: ap (n=ap@nat/apple/x-ervqisptzyvmznrz)
# [22:32] * Joins: abarth (n=abarth@dhcp-45-156.EECS.Berkeley.EDU)
# [22:36] <jgraham> beowulf: It would be interesting to repeat this exercise for some real table summaries. Sadly I don't know of any real, public, uses of the summary attribute that have structural information in
# [22:36] <Lachy> jgraham, I don't know for sure, but given that it's been asked about on the public-html, and no-one stepped forward to present the actual table that it's meant for, it seems likely that it doesn't exist.
# [22:36] * Quits: hober (n=ted@unaffiliated/hober) ("ERC Version 5.3 (IRC client for Emacs)")
# [22:37] * Quits: cying (n=cying@70.90.171.153)
# [22:37] * Joins: cying (n=cying@70.90.171.153)
# [22:38] * Quits: jennb_ (n=jennb@72.14.227.1) (Connection timed out)
# [22:39] * Quits: sbublava (n=stephan@77.119.68.106.wireless.dyn.drei.com)
# [22:39] * Quits: gsnedders (n=gsnedder@c83-252-197-103.bredband.comhem.se)
# [22:40] * Quits: roc_ (n=roc@121-72-178-80.dsl.telstraclear.net)
# [22:41] <beowulf> jgraham: yeah, i think if more people tried to do this there'd be a sudden realisation of the usefulness of @summary
# [22:43] <Xanthir> Well, not necessarily. Another possibility is the realization that we can generate useful automated summaries for simple tables, and that summaries are too difficult to generate even for humans for complex tables - blind people are just fucked in that case.
# [22:44] <Xanthir> In other words, in the first case machines can do summaries, so there's no reason to let humans do it, and in the second case humans can't do summaries, so letting them try is a fool's game.
# [22:44] <Dashiva> Well, some humans can do it
# [22:44] <beowulf> yeah, but how does a summary help a blind person or a person with learning difficulties if I can't even draw a table from a summary?
# [22:45] <beowulf> or do I miss a point somewhere?
# [22:45] <beowulf> perhaps I am a bad case study :)
# [22:45] <Dashiva> Perhaps this summary is a bad example
# [22:45] * Quits: ROBOd (n=robod@89.122.216.38) ("http://www.robodesign.ro")
# [22:46] <Lachy> This gives me another test that I can perform in my study :-)
# [22:46] * Philip` wonders what summary people are talking about
# [22:47] <Xanthir> beowulf: I'm saying that it may be true that @summary is never a useful attribute to expose. ^_^
# [22:47] <Lachy> If I take the summaries that people provide for the tables they are given, and ask others who haven't seen the tables to reconstruct the structure of the table and see how well they do
# [22:47] <beowulf> Xanthir: i think I agree
# [22:47] <Xanthir> When a summary would help a blind person, a machine can do it sufficiently well. When a summary wouldn't, not even a human can summarize it properly.
# [22:48] <Xanthir> There may be a middle ground of tables that are understandable through their summary, but are too complex to machine-generate a good one for.
# [22:48] <Xanthir> That'd be fun, Lach.
# [22:48] <Philip`> Given any particular algorithm for deriving summaries, I can't imagine it would be too hard to find an example which it doesn't handle but which can be trivially explained by a human
# [22:49] <Xanthir> Likely true. But further questions must be answered:
# [22:50] <Xanthir> 1) Can humans reliably explain tables well?
# [22:50] <Lachy> I'm not convinved about the ability for a computer to automatically generate a summary from a table in a way that would provide any more information to a user than they would obtain by manually scanning the table headers
# [22:50] <Xanthir> 2) If no, or at least partially no, is that 'middle class' of tables common enough to justify allowing bad summaries to float around for tables that *can* be machine-summarized?
# [22:51] <Xanthir> That's roughly what I'm talking about, Lachy.
# [22:51] <beowulf> do sighted people summarise tables?
# [22:51] <Xanthir> There's a clear class of simple tables where that is good enough, and while allowing a human to write a summary might produce a better/shorter one, it might also produce a *worse* one.
# [22:52] <Xanthir> So, some tests about people's ability to summarize tables well would be cool.
# [22:52] <beowulf> or do we just interogate them using axis as guides?
# [22:52] <Xanthir> Also: tests concerning current @summary use and usefulness.
# [22:52] <Lachy> Xanthir, I'm planning to do such a study
# [22:53] <Xanthir> Yeah, I know, I heard you a page ago. ^_^ I'm just making it clear what my point was.
# [22:53] <Lachy> I'm in the process of developing the methodology, but soon I will need to find or create some actual tables that I can get authors to mark up
# [22:54] <Xanthir> beowulf: that's a good question. If the prevalent methodology of sighted people is to just use them for a row/col lookup based on the headers, then that might indicate that a simple listing of row/column headers is rather useful.
# [22:54] <Xanthir> But if we do approach most tables more holistically, then perhaps that sort of machine-generated approach isn't so useful.
# [22:55] * Joins: jennb (n=jennb@72.14.227.1)
# [22:57] <beowulf> Xanthir: my guess is the former
# [22:58] <beowulf> but then if that's the case it's not a summary of a table you need for access, it's a whole new macguffin entirely, no?
# [22:58] <webben> jgraham: "real, public, uses of the summary attribute that have structural information" http://uk.biz.yahoo.com/mutualfunds/area/united_kingdom.html
# [22:58] <jgraham> webben: It's no good if you post the answer to everyone
# [22:59] <webben> jgraham: oh, you mean give the summary first then deduce the table?
# [22:59] <beowulf> if someone posts the summary, i haven't visited the link yet
# [22:59] <jgraham> webben: Yeah
# [23:00] <jgraham> The summary is "300 funds arranged by row in alphabetically ascending order. Each row gives the basic information for that fund. The first column group contains five columns. The first column contains the fund name, which is a link to the full details of the fund in our funds database. The next two columns contain Standard & Poor&8217;s ratings of the fund, if any. The fourth column contains a link to our partner Hargreaves Lansdown, where you c
# [23:00] <jgraham> (that probably got cut off)
# [23:00] <webben> it did
# [23:00] <jgraham> where?
# [23:00] <webben> ..."can buy the fund online. The last column in the first column group tells you the annual management charge that you would be charged. The second column group contains four columns summarizing the fund?s performance over one year and three years, and a fifth column giving a link to the full performance details for that fund."
# [23:00] <webben> should end at "for that fund."
# [23:01] <webben> http://pastebin.com/f227fa975
# [23:01] <Lachy> "Except that the DOCTYPE should change every time the specification changes, to allow for evolution during the development of the HTML5 spec." -- http://lists.w3.org/Archives/Public/public-html/2009Aug/0274.html
# [23:02] <Lachy> I guess that means Hixie should add the spec's SVN revision number to the DOCTYPE to meet that criteria. :-)
# [23:02] <Xanthir> So he's basically wanting revision counters in doctype?
# [23:02] <Xanthir> haha
# [23:02] <jgraham> webben: Did you write that?
# [23:03] * Joins: hober (n=ted@unaffiliated/hober)
# [23:03] <jgraham> (I think that information is pretty helpful to me in understanding the table albeit not essential)
# [23:03] <webben> jgraham: a while ago yes
# [23:03] <takkaria> I know, let's mark all the english documents we write with the revision of english that we're using so that future people now how to parse what we're saying
# [23:03] <webben> about two years ago
# [23:03] <Xanthir> Dude is stuck on the ancient idea that version numbers actually *mean* something.
# [23:04] * Joins: archtech (n=sv@83.228.56.37)
# [23:04] <hober> takkaria: heh
# [23:04] * Quits: dglazkov (n=dglazkov@nat/google/x-c11d202e2985cb80)
# [23:04] <hober> takkaria: well, aren't we all speaking en-GB-x-Hixie... should that be the default?
# [23:07] <Lachy> takkaria, a lot of things, though obviously not everything, get dated in some way when they are written, so historians can use that information to help determine the grammar in use
# [23:07] <Lachy> so in a way, the date acts like a version number in a constantly evolving language
# [23:08] * Lachy wonders how everyone will be speaking differently tomorrow?
# [23:09] <Xanthir> Ie se'eb geni'kaeep'seh eekileh siheteh.
# [23:09] <takkaria> Lachy: this is true, but when it comes to practical comprehension, no-one uses dates either
# [23:09] <Xanthir> All right, so, dude's premise:
# [23:10] <takkaria> natural language is as unversioned as HTML in practice is
# [23:10] <Xanthir> Conformance checkers need a version number to know what to check against. True, but this information doens't have to be in-band. Just specify the version when you active the checker.
# [23:11] <Xanthir> Browsers need a version number to know how to render the page. False, unless you're assuming that browsers carry around multiple engines frozen from each version.
# [23:11] <Lachy> takkaria, don't you suffix everything you say with the current date and time? (2009-08-05T22:05Z)
# [23:11] * Quits: jennb__ (n=jennb@72.14.227.1) (Read error: 110 (Connection timed out))
# [23:11] <takkaria> Lachy: not generally, I find it quite inconvinient
# [23:11] <Lachy> s/22:05/21:05/
# [23:12] <Xanthir> I need to join public-html so I can respond to stuff like that.
# [23:14] <beowulf> http://software.hixie.ch/utilities/js/live-dom-viewer/saved/196
# [23:14] <Lachy> anyway, I wish someone (not me) would inform larry that as far as versioning and conformance checkers are concerned, the interesting question to ask is not does this document conform to the version/profile/whatever that it specifiies itself, but rather does it conform to the version/profile/whatever that I'm interested in.
# [23:15] <Hixie> annevk2: k well let me know if that changes and i'll update the spec
# [23:16] <takkaria> Lachy: too practical for the überpedants, that
# [23:16] <beowulf> oops, thead not closed...
# [23:17] * Quits: poe (n=poe@unaffiliated/xerox)
# [23:17] <Lachy> ah, this is where I paraphrased that from http://about.validator.nu/#faq (in the What's wrong with DTD's section)
# [23:18] <beowulf> now that i've looked at the site, the summary would have been better had it just quoted the headers
# [23:18] <beowulf> if i was having it read to me i think i'd prefer to know detail rather than generalisations
# [23:21] * Joins: roc (n=roc@203-97-204-82.dsl.clear.net.nz)
# [23:21] <Lachy> beowulf, having the summary attribute merely list the headers is not at all useful
# [23:22] <Xanthir> Just in the half-hour or so I've been looking at things, it seems useful to me. It lets me preload the table structure into my head immediately.
# [23:22] * Joins: dglazkov (n=dglazkov@nat/google/x-lhzlgsaaepzaizdn)
# [23:23] <Xanthir> You'd probably be right if you says "not always useful".
# [23:23] <beowulf> i'd agree with not always useful
# [23:23] * Quits: abarth (n=abarth@dhcp-45-156.EECS.Berkeley.EDU)
# [23:23] <beowulf> but if i wanted to know what i could get from a table knowing the table headers would be a good start
# [23:24] <beowulf> it at least lets me ask the table questions
# [23:25] <beowulf> and if it's not useful i'd like to know what a useful summary is, if there's a vote
# [23:25] * Joins: abarth (n=abarth@dhcp-45-156.EECS.Berkeley.EDU)
# [23:27] * Quits: Rik|work (n=Rik|work@fw01d.skyrock.net)
# [23:27] * Joins: Rik|work (n=Rik|work@fw01d.skyrock.net)
# [23:27] * Quits: Rik|work (n=Rik|work@fw01d.skyrock.net) (Remote closed the connection)
# [23:28] <Xanthir> Do I need to pay anything to get on the public-html mailing list? This w3c form is scaring me.
# [23:28] <Hixie> you do not
# [23:29] <annevk2> Hixie, yeah, I'll file a bug once we update MQ
# [23:29] <Hixie> thanks
# [23:31] <Xanthir> Hm, the htmlwg page (http://www.w3.org/html/wg/) points me to a form that implies i have to become an invited expert, which also strongly implies that money is involved in this process. Am I doing things correctly, or is there a simpler way?
# [23:32] <annevk2> there's no money involved in getting an invited expert status
# [23:33] * Quits: stevepit (i=sjstrutt@naga.ircii.org) (Read error: 104 (Connection reset by peer))
# [23:33] <Xanthir> Ok, so at the relevant points I should just say "No, my employer doesn't want to pay any money, stop asking me."?
# [23:33] * Joins: stevepit (i=sjstrutt@naga.ircii.org)
# [23:33] * Quits: webben (n=benh@nat/yahoo/x-a13c2451db14dc3f) (Read error: 110 (Connection timed out))
# [23:36] <annevk2> Xanthir, http://blog.whatwg.org/w3c-restarts-html-effort
# [23:36] <annevk2> (took me ages to find, Google has removed blog.whatwg.org from their index)
# [23:37] * Quits: abarth (n=abarth@dhcp-45-156.EECS.Berkeley.EDU)
# [23:37] <Xanthir> ah, i remember seeing that a long time ago. it's no longer accessible from the "Mailing List" section of whatwg.org, which is where i think i saw it before.
# [23:38] <Xanthir> Or, dur, i could just scroll down a little bit.
# [23:38] * Quits: cying (n=cying@70.90.171.153) (Connection timed out)
# [23:38] <Xanthir> I should probably go put that on the wiki page.
# [23:39] <annevk2> it's was initially an email yes
# [23:39] <annevk2> not sure why it should also be on the wiki? did something change?
# [23:40] <Xanthir> The "Mailing List" option on whatwg.org takes you to a wiki page, which says nothing about the htmlwg.
# [23:41] <annevk2> that's not a wiki
# [23:41] <Xanthir> ...
# [23:42] <Xanthir> indeed it is not.
# [23:42] <Xanthir> the tabs threw me off.
# [23:43] <GPHemsley> Um... is it just me, or are reference links broken?
# [23:43] * Joins: abarth (n=abarth@dhcp-45-156.EECS.Berkeley.EDU)
# [23:43] <Hixie> not just you
# [23:43] <Hixie> i haven't done them yet
# [23:43] <GPHemsley> ah
# [23:43] <GPHemsley> ok
# [23:44] * Joins: weinig_ (n=weinig@17.246.18.83)
# [23:45] * Quits: weinig_ (n=weinig@17.246.18.83) (Remote closed the connection)
# [23:45] * Quits: weinig (n=weinig@nat/apple/x-5636aaa30ef45490) (Read error: 104 (Connection reset by peer))
# [23:45] * Joins: weinig (n=weinig@nat/apple/x-yrvnmnhgkcezrmch)
# [23:45] * Joins: weinig_ (n=weinig@17.246.18.83)
# [23:46] * Joins: Rik|work (n=Rik|work@fw01d.skyrock.net)
# [23:47] * Quits: weinig (n=weinig@nat/apple/x-yrvnmnhgkcezrmch) (Read error: 60 (Operation timed out))
# [23:50] * Quits: stevepit (i=sjstrutt@naga.ircii.org) (Remote closed the connection)
# [23:50] * Joins: stevepit (n=steve@naga.ircii.org)
# Session Close: Thu Aug 06 00:00:00 2009

The end :)