# [06:57] <MikeSmith> issue: Should "HTML5: The Markup Language" draft contain parsing rules and DOM APIs?
# [06:57] * trackbot noticed an ISSUE. Trying to create it.
# [06:57] <trackbot> Created ISSUE-67 - Should \"HTML5: The Markup Language\" draft contain parsing rules and DOM APIs? ; please complete additional details at http://www.w3.org/html/wg/tracker/issues/67/edit .
# [07:03] <MikeSmith> action: Michael(tm) to add a note about issue-67 to the Status section of "HTML 5: The Markup Language"
# [07:03] * trackbot noticed an ACTION. Trying to create it.
# [07:03] <trackbot> Created ACTION-100 - Add a note about issue-67 to the Status section of \"HTML 5: The Markup Language\" [on Michael(tm) Smith - due 2009-02-04].
# [07:03] * trackbot getting information on ACTION-100
# [07:03] <trackbot> ACTION-100 -- Michael(tm) Smith to add a note about issue-67 to the Status section of \"HTML 5: The Markup Language\" -- due 2009-02-04 -- OPEN
# [08:56] <MikeSmith> Hixie: I don't think that was the most painful part of the f2f, at least
# [08:56] <Hixie> MikeSmith: i meant the tpac as a whole
# [08:56] <Hixie> i can't wait to see how long it takes for the htmlwg to decide to publish a last call draft, given how long it is taking for us to decide to publish a WD
# [09:00] <anne> I thought we'd just do it, but apparently the rules have changed again
# [09:00] <anne> maybe because some people do not want to publish that other draft
# [09:01] <Hixie> i'm fine with publishing the other draft, though mike's recent comments confused the heck out of me and now i no longer know who the audience is supposed to be any more
# [09:02] <hsivonen> I'd like to see anyone who object to publishing a new WD of "HTML 5" to cite specific changes to "HTML 5" that would remove the objection.
# [09:02] <MikeSmith> I'm frankly tired of discussing the audience non-issue.
# [09:04] <Hixie> i don't know how to review the draft if i don't know who it is supposed to be for
# [09:04] <Hixie> and since it's defining the same thing as the draft i'm editing, it seems like my review would be pretty important
# [09:06] <MikeSmith> as I've pointed out several times, we all had no trouble reviewing the HTML5 draft for the 4 years or whatever it existed before you added the Audience section
# [09:06] <Hixie> that's because everyone knew who it was for. browser vendors, tool vendors, language lawyer authors, and conformance checker authors.
# [09:06] <hsivonen> MikeSmith: when I reviewed "HTML 5" during that time, I thought I knew what it was for
# [09:07] <hsivonen> MikeSmith: incidentally, my conjecture of who it was for matches precisely what Hixie just said about it target audience
# [09:08] <MikeSmith> it's for people who need a definition of what a conformant document is
# [09:09] <Hixie> then it's woefully incomplete and not at all what larry wants.
# [09:09] <hsivonen> MikeSmith: of those set of people, at least the needs of validator developers are better addressed by "HTML 5"
# [09:11] <MikeSmith> well, I'd really love to spend a whole bunch more time discussing this right now, but I can't because I need to go do a presentation for a big portal site here about HTML5 and W3C stuff
# [09:11] <MikeSmith> I'll be back on later, we can pick it up then if you care to
# [14:15] <MikeSmith> takkaria: about the test-case stuff, I think once you get more time freed up (end of next month or March, I think you said), setting up something at github or wherever would be good
# [14:16] <MikeSmith> I don't know what other git hosting services there are, but github seems to be pretty good
# [14:33] <MikeSmith> Philip: I have some reason to suspect it would be git, but I would not totally rule out Mercurial
# [14:33] <MikeSmith> but I think it would be one of those two
# [14:37] <Philip> MikeSmith: Okay - I'm just guessing it'd be easier to migrate a repository if both sides are using the same VCS, but it doesn't seem a critical issue
# [14:45] <Philip> "these steps don’t seem to install the updated git man pages, as they’re still on 1.5.5. I’m working on figuring that one out."
# [14:45] <MikeSmith> the sources for the git docs are in the kind of markdown-type thing called asciidoc
# [14:46] <MikeSmith> which I had nothing to do with
# [14:46] <Philip> "I compile git and its man pages myself and I just noticed that the man pages (invoked with "git help log", for example) have a typesetting problem. There are ".ft" commands here and there"
# [14:47] <MikeSmith> hmm, I think I since have fixed that bug
# [14:47] <MikeSmith> and we released a release with the fix
# [14:47] <MikeSmith> the source of that problem as actually asciidoc
# [14:48] <pimpbot> Title: A typesetting problem with git man pages | KernelTrap (at kerneltrap.org)
# [14:48] <MikeSmith> the asciidoc tool converts its source input to DocBook, and uses the man-page stylesheet from the DocBook XSL stylesheets to generate man pages from that
# [14:49] * Philip is not trying to be negative on purpose, that's just how the Google search came out ;-)
# [14:49] * MikeSmith gets all kinds of SSL warning messages when I try to go to that URL
# [14:50] <Philip> You should switch to a browser that doesn't give you so many scary warning messages
# [14:53] <MikeSmith> because first off, they all hate man pages and the man macros
# [14:53] <Philip> Lachy: It doesn't fail to notice the problem; it just doesn't warn me, and merely puts a grey question-mark box in a place I'll never notice, saying "Site not secure blah blah gobbledygook" if I click on it
# [15:48] <pimpbot> Title: The Web Developer’s Guide to HTML 5 (at dev.w3.org)
# [15:48] <MikeSmith> one great thing about visiting Opera Japan is that you can have a chance to meet actual customers -- companies that pay money for a browser they can install on mobile phones, Nintendo DS, Wii, Windows Mobile devices, other stuff
# [15:49] <Lachy> (the heading levels are a bit messed up, but that's either a bug in anolis or, more likely, I've messed up my source document during the major restructing attempt)
# [15:52] <MikeSmith> speaking of anolis, haven't see gsnedders around so much lately
# [15:52] <MikeSmith> or maybe I've just not been around as much when he is
# [16:13] <karl> jgraham: I don't think it is really confusing. There is a bit of filter effect here. Many foreigners who come to Japan *want* to see something strange and indeed they see it as a strange country.
# [16:13] <karl> As many countries, there are things which are similar and different.
# [16:13] <jgraham> karl: The Tokyo metro system is confusing. Or I am really dumb. Or both.
# [16:14] <MikeSmith> I agree that it's both awesome and that it's still sometimes weird and confusing as hell for me even after being here 7 years
# [16:14] <MikeSmith> metro system is just really complex
# [16:14] <karl> :) metro system is like english railways system :)
# [16:14] * Philip 's experience of Japan is limited to watching Lost in Translation
# [16:14] <karl> but I really prefer to take the subway in tokyo than in Paris, less less less confusing for me
# [16:15] <MikeSmith> myakura: I was showing your blog to some developers at goo earlier tonight
# [16:15] <jgraham> karl: We don't have the thing where there are multiple stations with different operators but the same name and where you can buy tickets that will allow you on one operator's trains but not on the others
# [16:15] <MikeSmith> I recommended to them that they subscribe to your feed, if they want to keep up with what's going on around W3C/Web-standards stuff
# [16:15] <karl> funny I talked about that just yesterday
# [16:17] <MikeSmith> jgraham: we have IC-card-based things here now that work across all the trains and subways
# [16:17] <MikeSmith> myakura: yeah. I did a presentation there earlier tonight
# [16:17] <MikeSmith> they have a great crew there
# [16:17] <MikeSmith> young and smart and really interested in learning about new standards stuff
# [16:18] <myakura> MikeSmith: interesting. was that about html5 and webapps?
# [16:18] <jgraham> karl, MikeSmith: For the tourist it is confusing enough that I wasted about 30 minutes and got the wrong ticket trying to get to Tsjuki (sp?) in November
# [16:20] <MikeSmith> all I have to do is tell it my destination
# [16:20] * karl remembers one of the first dicussions I had with mike ;) about mobile phones
# [16:21] <MikeSmith> mobiles all have GPS here, so device and apps on they can already know where you are
# [16:21] <MikeSmith> and you just need to tell them where you want to go
# [16:22] <karl> MikeSmith: plus there is a little green guy with a helmet which follows you everywhere with a timer to optimize your work. That is Japan.
# [16:48] <jgraham> takkaria: FWIW my input on DVCS is that I know Mercurial a little and therefore have a preference for using it. But I guess Git is not likely to be so hard to learn
# [16:50] <jgraham> (and bitbucket.org seems to be the GitHub equivalent if that is an issue)
# [16:53] * Philip wonders why IEBlog has no comments on the recent posts, and also why it pops up the browser's Username/Password dialog box every so often
# [16:56] * Philip has used Mercurial a bit too (mainly because Mozilla uses it), but doesn't think that counts for much since he hasn't done anything non-trivial with it and has no way to compare it against Git
# [17:04] <Philip> takkaria: You missed an opportunity to CC public-html-testsuite and increase its lifetime post count by 20% :-(
# [17:04] <jgraham> Yeah, I haven't used Git either so I don't have a strong preference
# [17:09] <Philip> But you must have heard the horror stories
# [17:09] <anne> I wonder when we get the first bikeshed discussion about whether <section> implies </p> or some other such trivial detail
# [17:11] <jgraham> gsnedders: I once (like >10 years ago) worked with some people who were using Visual SourceSafe. IIRC they often had to ask each other to unlock the files in order to edit them
# [17:13] <Philip> I think SourceSafe worked by storing the repository on a shared network drive and giving everyone write permissions, so a single bad client could destroy the entire repository
# [17:13] <Philip> (and that's if you remember to run the repair process every night so the natural corruption didn't destroy the repository first)
# [17:14] <jgraham> Quite an ironic name for a product, really
# [17:15] <takkaria> Philip: I was not aware public-html-testsuite existed
# [17:15] * Philip knows someone who recently started a job and has to use something that's basically RCS plus a load of shell scripts that don't work half the time
# [17:16] <takkaria> most people seem to say they prefer hg over git, I have no real preference
# [17:16] <Philip> People who use Git seem to prefer it over Mercurial
# [17:24] <jgraham> takkaria: I guess it depends howw strong the chance is that W3C will pick git over hg; if it is close to 50:50 then maybe we should just make a choice and hope that tips the balance. If it is like 90:10 then I guess we should go with the 90
# [17:25] <takkaria> hmm, html5lib has a php port now
# [17:27] * takkaria should really update hubbub to latest spec soon
# [17:27] <karl> hmm larry is saying the survey is biaised but so far most of the answer I have seen are in favor of markup spec. So the first draft effect seems to not be an issue
# [17:28] <Philip> He's also complaining about the question being biased, and seemingly not minding that asking a few people's Twitter followers is going to introduce more huge bias
# [17:29] <Philip> (and also his reworded question seems to be indicating that people aren't already able to review the document exactly the same as if it was a normative published draft)
# [17:29] <jgraham> karl: It's not clear that it matters
# [17:30] <jgraham> (although I guess it will bias it toward yes)
# [17:30] <karl> jgraham: you mean compared to the bias of the no of techies?
# [17:31] <Philip> karl: It's a very selective sample, and unlikely to be representative of any larger group of people, so there's no value in counting the number of yes/no votes
# [17:31] <Philip> (though there is value in reading specific feedback from those people)
# [17:32] <karl> that is funny. because people start to wonder about the "representativeness" (ouch) when there was a question of doing things differently
# [17:33] * jgraham wonders if karl has turned into a sheep
# [17:34] <karl> jgraham: that would make my life easier ;)
# [17:34] <Philip> ((Twitter probably isn't the ideal medium for substantive feedback, but at least it's better than nothing))
# [17:36] <Philip> "The IETF even has a rather amusing way of solving this via humming." - that would clearly violate our charter, since humming is only likely to work in teleconferences and therefore will exclude mere mailing list participants, unless we can attach audio files of humming
# [17:36] <jgraham> Philip: It is unclear to me that's true
# [17:37] <Philip> jgraham: It's unclear that it's better than nothing?
# [17:37] <karl> geez people like meta discussions :) no no in fact they are craving for it
# [17:40] <jgraham> Philip: Yes. I'm not really sure what the value of Twitter feedback is, but it is not clear that the value must be positive
# [17:43] <pimpbot> Title: jd/adobe: A Video Tale (at blogs.adobe.com)
# [17:44] <Philip> jgraham: Regardless of the feedback, the other part (feedforward?) of making people aware of the Markup Language document seems useful since it makes them more likely to read it and send feedback through a higher-bandwidth medium
# [17:44] * karl has to find a way to have less tabs opened. right now 124 tabs
# [17:45] <jgraham> karl: Do you have a cunning way to count or just by hand?
# [17:46] <Philip> "No one's disenfranchised" except, say, anyone on a platform that Flash doesn't support
# [17:47] <Philip> and anyone who has to indirectly pay their platform vendor for having added Flash support
# [17:48] <Lachy> karl, how do you possible manage 124 tabs?
# [17:49] <Lachy> (I know someone who's had 3 or 4 times that amount, but still.)
# [17:49] <Philip> (And also me, because I can't get the Flash plugin to work in Opera without crashing every few days)
# [17:49] <Lachy> I can't manage with more than a dozen tabs open, and I consider more than 6 too many
# [17:50] <karl> Lachy: that's my issue. I tried in different browsers. On my macbook, the best for low cpu consumptions is opera, the best for navigating the tabs at the chrome level is camino.
# [17:51] * jgraham has about 90 tabs spead over 3 browsers but most of those could be closed
# [17:55] <karl> Another possibility for me would be a system where I can keep a copy of the pages in an app at a time t. So I can go back to it later. The fact that I have so many tabs opened is because I want to talk or write about it later on. Either on a blog or mailing list, etc. sometimes it takes month before I write about it
# [17:57] <karl> 2. It is not saving the page as it is. (when I access to it again there will be a reload and that matters sometimes)
# [17:57] <Philip> Would Opera's Sessions work better? They save the state of all tabs, though I don't know if that's just URLs or includes form content too
# [17:57] <karl> but I'm using bookmarking systems too but for other stuff :) http://blogmarks.net/ is my main tool for it
# [17:58] <pimpbot> Title: Blogmarks.net : Public marks (at blogmarks.net)
# [18:00] <karl> Philip: yes I'm using Opera for the last two months. Good things for CPU (specifically with a lot of tabs), awful navigation for tabs, awful UI on the mac. But cool feature also like dragonfly, or the block content feature or the custom preferences adjustement by site. A bit dumb for setting/removing cookies. It doesn't really work.
# [18:02] * Philip hates how browsers all default to ordering tabs by recently-usedness when switching, rather than being consistent and matching the tab bar
# [18:03] * karl wishes that in fact it would be possible to have your tabs sessions be outside of the browser. So that you can reuse them from any browser
# [18:04] <karl> you quit Opera, open Camino… and hop you get the same session
# [18:05] <Philip> Why would you ever want to do that? :-)
# [18:10] <karl> Philip: the world is full of surprise
# [18:10] <karl> ok. I have to do something more important now… going to lunch.
# [18:12] * Quits: Lachy (Lachlan@213.236.208.22) (Quit: This computer has gone to sleep)