/irc-logs / w3c / #html-wg / 2007-06-01 / end

Options:

  1. # Session Start: Fri Jun 01 00:00:00 2007
  2. # Session Ident: #html-wg
  3. # [00:01] * Quits: Sander (svl@71.57.109.108) (Quit: And back he spurred like a madman, shrieking a curse to the sky.)
  4. # [00:09] * Quits: edas (edaspet@88.191.34.123) (Ping timeout)
  5. # [00:20] * Quits: karl (karlcow@128.30.52.30) (Quit: Where dwelt Ymir, or wherein did he find sustenance?)
  6. # [00:20] * Quits: heycam (cam@203.214.95.190) (Ping timeout)
  7. # [00:24] * Quits: mjs (mjs@17.255.96.233) (Quit: mjs)
  8. # [00:33] * Quits: anne (annevk@81.68.67.12) (Ping timeout)
  9. # [00:36] * Quits: kingryan (rking3@208.66.64.47) (Quit: kingryan)
  10. # [00:36] * Joins: kingryan (rking3@208.66.64.47)
  11. # [00:41] * Joins: mjs (mjs@17.255.96.233)
  12. # [01:04] * Quits: gavin_ (gavin@74.103.208.221) (Ping timeout)
  13. # [01:09] * Joins: gavin_ (gavin@74.103.208.221)
  14. # [01:11] * Joins: heycam (cam@130.194.72.84)
  15. # [01:16] * Parts: hasather (hasather@81.235.209.174)
  16. # [01:16] * Quits: billmason (billmason@69.30.57.156) (Quit: .)
  17. # [01:22] <MikeSmith> cheers to mjs for pointing out that discussion in public-html thread about CSS should better take place somewhere other than on public-html
  18. # [01:23] * Joins: Zeros (Zeros-Elip@67.154.87.254)
  19. # [01:27] <DanC_lap> mjs is helping shape the discussion, yeah... I wonder if he and a few others are interested to do that regularly
  20. # [01:27] <mjs> what exactly would I be signing up for?
  21. # [01:28] <DanC_lap> good question... hmm...
  22. # [01:29] <DanC_lap> well, it's sort of a vacuous task as long as i can't figure out what the group should be doing
  23. # [01:30] <DanC_lap> if Hixie doesn't want to do review-by-email of html5 for a few months, I'm tempted to focus on testing/validation-tools... others have suggested going back to design principles.
  24. # [01:31] <mjs> he's not objecting to review-by-email
  25. # [01:32] <mjs> he just doesn't want an obligation to respond to review comments live as they come in
  26. # [01:32] <mjs> and would prefer to have discussion recorded
  27. # [01:32] <mjs> (in practice maybe he'd jump into some of the discussions anyway)
  28. # [01:32] <mjs> at least that is my understanding of his comments
  29. # [01:33] <DanC_lap> my experience of review-by-email is that comments are directed at the editor. Granted, the editor can only respond to so many comments... "best effort" was the way I phrased it that he didn't seem interested in.
  30. # [01:33] <Hixie> right now i'm replying to e-mails from 2005, 2006
  31. # [01:33] <Hixie> i simply don't have the bandwidth to make changes in response to e-mails in real time
  32. # [01:34] <Hixie> however, i would really love to see controversial proposals discussed, with opinions summarised
  33. # [01:34] <Hixie> e.g. in the wiki
  34. # [01:34] <Hixie> with each issue summarised in the way i described, so that hyatt and i can just go through each one when we get there
  35. # [01:35] * DanC_lap gets called back to the tag meeting...
  36. # [01:36] <mjs> DanC_lap: my experience with reviewing specs in the past is that multiple weeks before any feedback from the group was par for the course - internal review is different of course, but I think recording the issues is much more important than realtime editorial response
  37. # [01:37] <mjs> (actually, "multiple weeks" is the good case, many times it went months without even an indication that my feedback had been formally recorded anywhere)
  38. # [01:38] <DanC_lap> exactly, multiple weeks sucks... it's best to reward reviewers with feedback within a few days
  39. # [01:38] <Hixie> if it takes me one hour to respond to a piece of feedback (average)
  40. # [01:39] <DanC_lap> I'm not all that motivated to have people review the spec in detail if the editor isn't interested to lead discussion of (some of) the comments that come in
  41. # [01:39] <Hixie> and i spend 8 hours per day responding to feedback
  42. # [01:39] <Hixie> that means we can deal with 8 e-mails per day
  43. # [01:39] <DanC_lap> 8 per day would be plenty. 5 or 6 a week would be useful.
  44. # [01:39] <Hixie> many of which would be requesting contradictory things
  45. # [01:39] * Hixie would much rather deal with e-mails at the rate of ~50 per day
  46. # [01:39] <Hixie> taking blocks of related e-mails at once and responding to all of them
  47. # [01:39] <Hixie> as i do on the whatwg list to great effect
  48. # [01:40] * Hixie has received many comments to the effect that people really like the summary e-mails he sends
  49. # [01:40] <DanC_lap> right. I'd expect you to grab one or two threads a few times a week.
  50. # [01:40] <Hixie> but then if we're going to wait a week, why can't someone else do the reading and summarising for me, letting me focus on the editing part?
  51. # [01:41] <Hixie> i mean, the group has 300 people, SOMEONE has to be ready to do work other than me and hyatt! :-)
  52. # [01:41] <DanC_lap> well, if someone else has most of the spec in their head, I suppose that's an idea
  53. # [01:41] * hyatt memorizes the spec.
  54. # [01:41] <Hixie> you don't even have to know the spec to summaries proposal
  55. # [01:41] <Hixie> s
  56. # [01:41] * hyatt is actually developing a moderately encyclopedic knowledge of parsing :)
  57. # [01:41] <Hixie> heh
  58. # [01:42] <mjs> DanC_lap: recording the issue, and discussion among working group members, is IMO highly valuable
  59. # [01:43] <DanC_lap> yes, but N^2 discussion is pretty inefficient. i'd like each discussion to have 1 or a few leaders.
  60. # [01:43] <Hixie> isn't the "leading" role you and chris' job? :-)
  61. # [01:44] <mjs> feedback from the editor is mainly useful if they want to request something from the commentor (like further info), since comments are input into the editing process, not output
  62. # [01:44] <DanC_lap> well, not for most of the discussion. I see my role as getting involved when discussion leader and the group can't find consensus.
  63. # [01:44] <mjs> I'm willing to help organize discussion, depending on topics, but I can't formally commit to recording all issues
  64. # [01:45] <DanC_lap> how about a few issues, or one or two sections, mjs?
  65. # [01:45] <Hixie> ok, so we have someone who'll lead discussion, now we just need someone to summarise discussions for the editors
  66. # [01:45] <mjs> DanC_lap: I have many time commitments, and recording something in an issue tracker well is a time-consuming task - I would likely do some of that for issues I care about, but I could not in good faith commit to doing it for all issues on a single topic even
  67. # [01:46] <mjs> I can think of lots of people who could do a good job recording issues, but I'm not sure if any of them are willing
  68. # [01:46] <mjs> I think maybe some could be convinced if it is a rotating responsibility, maybe based on week or month the message thread started
  69. # [01:46] <DanC_lap> "issues I care about" is a great filter. either the group gets the discussion leader to care about something, or it remains on the floor. (but if you can't do a whole section, I'll keep searching for some combination of people...)
  70. # [01:47] <mjs> DanC_lap: my personal areas of interest are not necessarily in 100% alignment with what is actually important to the group as a whole
  71. # [01:48] <DanC_lap> right, but it's a good 1st cut. I'd get involved if it became evident that you and must/most of the group disagreed.
  72. # [01:48] <DanC_lap> much/most
  73. # [01:49] <Hixie> http://esw.w3.org/topic/HTML/AbbrAcronym01 is the kind of thing i had in mind
  74. # [01:49] <Hixie> (though that isn't perfect)
  75. # [01:50] <DanC_lap> hmm... I would have thought that abbr/acronym would remain on the floor. hardly seems worth bothering about.
  76. # [01:50] <Hixie> i mean the structure
  77. # [01:50] <DanC_lap> ok
  78. # [01:50] <Hixie> i wasn't making any sort of comment on the issue itselfg
  79. # [01:50] <Hixie> the point being, there are people who are willing to make wiki pages that summaries the issues already
  80. # [01:50] <Hixie> i mean, they've done it for several issues
  81. # [01:51] <mjs> DanC_lap: I don't mind being involved in discussion, I am just being realistic that if it's my job to record issues, there will be a huge backlog because I have many time commitments already
  82. # [01:51] <DanC_lap> er... well, those wiki issue pages aren't among the most interesting comments I've seen on the spec, actually.
  83. # [01:51] <mjs> I think it would be fun to do, I just can't do it as a full time job
  84. # [01:52] <DanC_lap> "huge backlog" is something I mean to avoid by default; i.e. things don't get queued by default. things just sit there on the [archive] floor, by default.
  85. # [01:52] <DanC_lap> mjs, any particular section you're interested in?
  86. # [01:53] <DanC_lap> the wiki stuff seems to be an exercise in stucture for structure's sake. if the sections were more like "research into the current state of the web" and "estimated cost of change" and such, they'd be more interesting.
  87. # [01:54] <mjs> DanC_lap: personally? I'm especially interested in scripting APIs, multimedia, tokenizing/parsing (including attribute microsyntaxes) and the corresponding error handling
  88. # [01:54] <mjs> also hyperlinking mechanisms
  89. # [01:54] <mjs> and events
  90. # [01:55] <mjs> and forms, and generallly support for web apps (including offline web apps)
  91. # [01:55] <DanC_lap> you're the 3rd to express interest in tokenizing/parsing, my personal favorite.
  92. # [01:55] <mjs> I'm somewhat less interested in pure document issues like semantic markup for prose
  93. # [01:55] <DanC_lap> 4th, counting me
  94. # [01:56] * MikeSmith wonders whether Lachy might have time/interest in helping with summarizing discussions now and then
  95. # [01:56] <mjs> if I were to pick my favorite thing to start with, I'd start with the globally applicable APIs (HTMLElement, HTMLDocument, Window and related...)
  96. # [01:56] <mjs> I dunno how well that corresponds to a section though
  97. # [01:56] <DanC_lap> Hixie, can you think of a useful filter for comments that are worth your while to consider? things like "at least 4 WG members want this change" come to mind, but popularity contests.... well... suck, most of the time.
  98. # [01:57] <mjs> I think getting things recorded on a wiki and bugzilla is sufficient, and we can ask the editors if we'd like them to consider some area sooner rather than later
  99. # [01:57] <Hixie> DanC_lap: i want to consider everything
  100. # [01:57] <mjs> in WHATWG Hixie has historically taken requests to look at some particular area sooner
  101. # [01:57] <Hixie> yeah
  102. # [01:57] <Hixie> especially from implementors
  103. # [01:58] * Joins: zcorpan_ (zcorpan@84.216.43.138)
  104. # [01:59] * DanC_lap noodles on using implementator attention as a priority mechanism...
  105. # [02:00] <Dashiva> When you have a feedback loop that goes idea -> implementation -> evaluation...
  106. # [02:00] <DanC_lap> mjs, a wiki could help with the N^2 problem, but training a group to use a wiki well is challenging...
  107. # [02:01] <mjs> I don't think a wiki would be the best choice, bugzilla is good for tracking and organizing
  108. # [02:01] <mjs> and for keeping a paper trail
  109. # [02:01] <mjs> but it's a bit of work to manage it
  110. # [02:02] <Hixie> http://esw.w3.org/topic/HTML/IssueTemplate
  111. # [02:02] * Joins: karl (karlcow@128.30.52.30)
  112. # [02:02] <DanC_lap> I'm trying to figure out getting reviewers to do careful reviews even when they won't get any response from the editor any time soon. I'd like to understand the feedback loop that involves changes to the spec better.
  113. # [02:04] <DanC_lap> hmm... http://esw.w3.org/topic/HTML/IssueTemplate is getting there...
  114. # [02:05] <DanC_lap> ... the esw wiki isn't a good home for the zillions of issues we'll have. we'll prolly need a new wiki. sigh.
  115. # [02:07] <mjs> I think just the feedback that "your comment is recorded as issue #xyz, here's the bugzilla link to track it" is pretty good
  116. # [02:07] <mjs> and better whan what you get in most WGs
  117. # [02:07] <mjs> but then you need people to be willing to file in bugzilla, and people willing to close the bugs as spec changes get made
  118. # [02:07] <DanC_lap> can you tell me more of the story? including the part where the spec text changes?
  119. # [02:08] * Joins: olivier (ot@128.30.52.30)
  120. # [02:10] <mjs> well, I'm hoping when the editors approach an area (perhaps by request of the chairs or of implementors or of enough of the working group) they look at the recorded issues, linked email threads, etc, and make appropriate changes which they hopefully announce in some appropriate way
  121. # [02:11] <mjs> this will presumably work better if group discussion has already happened and developed some rough consensus with good supporting evidence
  122. # [02:11] <DanC_lap> what's a likely frequency of such changes? it sounds like a ceremony that happens just a few times a month. am I getting the wrong picture?
  123. # [02:12] <Hixie> i'm planning on doing changes pretty much continuously
  124. # [02:12] <mjs> well, how the editors edit is best asked of the editors
  125. # [02:12] <mjs> my impression is that editing happens all the time, but right now is mostly addressing issues raised before the HTML WG was even formed
  126. # [02:12] <mjs> and also more recent issues on specific request
  127. # [02:12] <hyatt> i will be editing using a script hixie set up for me
  128. # [02:12] <hyatt> so checkins will still look like they come from hixie
  129. # [02:13] <Hixie> they'll be labelled "(hyatt)"
  130. # [02:13] <hyatt> it's going to say (hyatt) or something, right hixie?
  131. # [02:13] <hyatt> yeah
  132. # [02:13] <Hixie> i get an e-mail every time you run your commit script
  133. # [02:13] <Hixie> it adds the diff to my queue
  134. # [02:13] <Hixie> and i will write a script that basically just shows me the diff and commits it
  135. # [02:13] <Hixie> (after running through the gauntlet of preprocessors)
  136. # [02:14] <DanC_lap> er... are there any paths that don't go thru hixie?
  137. # [02:14] <Hixie> well, he could commit straight to the cvs repo
  138. # [02:14] <Hixie> but then it wouldn't get propagated to the whatwg version
  139. # [02:14] <Hixie> and that would be a pain
  140. # [02:14] <hyatt> yeah
  141. # [02:14] <Hixie> if i get run over by a bus, the whatwg version becomes somewhat academic anyway
  142. # [02:14] <DanC_lap> ok, that's available in a pinch.
  143. # [02:15] <Hixie> http://esw.w3.org/topic/HTML
  144. # [02:16] * hyatt heads home
  145. # [02:16] * Quits: hyatt (hyatt@17.255.98.44) (Quit: hyatt)
  146. # [02:20] <zcorpan_> DanC_lap: i seem to recall an email from you (to public-html) wanting to discuss organization of test cases
  147. # [02:24] <mjs> I gotta go
  148. # [02:24] * Quits: mjs (mjs@17.255.96.233) (Quit: mjs)
  149. # [02:25] * Joins: mjs (mjs@17.255.96.233)
  150. # [02:27] * Quits: tH (Rob@87.102.91.218) (Quit: -3)
  151. # [02:27] <mjs> DanC_lap: if you want me to take a shot at leading discussion, I could try doing it after Apple's developer conference, i.e. starting the week of June 18th, and I could work on recruiting people to help w/ issue tracking and on coming up with a near-term review agenda as a background task
  152. # [02:28] <mjs> DanC_lap: I can't really do all issue tracking personally, but I might be able to persuade some people to help out
  153. # [02:29] <mjs> DanC_lap: anyway, I gotta go for now
  154. # [02:33] * DanC_lap peeks in again after some more meeting...
  155. # [02:40] * Hixie sends a mail to kickstart this
  156. # [02:47] <mjs> Hixie: the wiki is low-overhead to add to, but will be high-overhead to organize and query once there are a lot of issues -- but maybe it's ok to start with a very low-process tool and scale up when/if the need arises
  157. # [02:48] <Hixie> i'm very happy to move to better
  158. # [02:48] <Hixie> tools
  159. # [02:48] <mjs> Hixie: for example, if we got up to 2000 issues, it would be pretty hard to search to see if someone had already raised a given issue
  160. # [02:48] <mjs> I thought about it and the wiki is a fine starting point and better than nothing
  161. # [02:48] <Hixie> however, "bad tool (wiki)" is infinitely better than the status quo of "no tool"
  162. # [02:48] <Hixie> yeah
  163. # [02:48] <Hixie> what you said
  164. # [02:51] * Quits: dbaron (dbaron@63.245.220.242) (Quit: 8403864 bytes have been tenured, next gc will be global.)
  165. # [02:52] * Hixie wonders wtf document.dir is
  166. # [02:53] <zcorpan_> same as document.documentElement.dir ?
  167. # [02:53] * Joins: sbuluf (fiaqidz@200.49.140.207)
  168. # [02:53] <Hixie> nope
  169. # [02:53] <Hixie> http://software.hixie.ch/utilities/js/live-dom-viewer/?%3C%21DOCTYPE%20html%3E%3Cscript%3Ew%28document.dir%20%3D%20%27rtl%27%29%3C/script%3Etest
  170. # [02:56] <zcorpan_> opera adds a DIR="rtl" on <html>
  171. # [02:56] <zcorpan_> ie7 too
  172. # [02:56] * Hixie cracks out IE
  173. # [02:56] <Hixie> if IE does, then sweet
  174. # [02:56] <Hixie> defining it will be trivial
  175. # [02:57] <Hixie> indeed it does
  176. # [02:58] <Hixie> sweet
  177. # [02:58] <karl> dir defined as an element?
  178. # [03:00] <zcorpan_> karl: attribute (both DOM attribute and content attribute)
  179. # [03:00] <zcorpan_> Hixie: also, both ie7 and opera allows text nodes outside the root element, but they aren't rendered
  180. # [03:01] <Hixie> fun
  181. # [03:03] <zcorpan_> or well, when i try the same in ie7 from a local file, it says that it can't open the file
  182. # [03:03] * Quits: DanC_lap (connolly@128.30.52.30) (Ping timeout)
  183. # [03:05] * Hixie looks at trying to define document.lastModified
  184. # [03:07] <mjs> Hixie: I'm writing an email of possible near-term focus discussion topics - are there any that you'd be interested in?
  185. # [03:07] <mjs> (right now the list is somewhat biased towards my personal interests)
  186. # [03:07] <Hixie> hm
  187. # [03:07] <mjs> oops, gotta run
  188. # [03:07] <mjs> will ask again tonight
  189. # [03:08] <Hixie> k
  190. # [03:08] <karl> i knew the attribute.
  191. # [03:09] <zcorpan_> karl: oh, you were thinking of the <dir>ectory element?
  192. # [03:11] * Quits: gavin_ (gavin@74.103.208.221) (Ping timeout)
  193. # [03:12] * Quits: mjs (mjs@17.255.96.233) (Ping timeout)
  194. # [03:12] * Quits: sbuluf (fiaqidz@200.49.140.207) (Connection reset by peer)
  195. # [03:12] * Joins: sbuluf (jwj@200.49.140.207)
  196. # [03:16] * Quits: h3h (bfults@66.162.32.234) (Quit: |)
  197. # [03:17] * Joins: gavin_ (gavin@74.103.208.221)
  198. # [03:24] <karl> yes I was wondering why hixie was surprised by the dir element, or maybe was it just the dom construct
  199. # [03:25] <zcorpan_> karl: document.dir is not <dir>
  200. # [03:25] <zcorpan_> karl: it's <html dir="">
  201. # [03:27] * Joins: mjs (mjs@64.9.237.122)
  202. # [03:28] <karl> yes :) then it's a well-known attribute
  203. # [03:28] <Hixie> mjs: i couldn't really see any area that i'd like reviewed before any other
  204. # [03:28] <karl> http://www.w3.org/TR/html401/struct/dirlang.html#adef-dir
  205. # [03:29] <mjs> Hixie: ok, I'll just suggest some and see if anyone else on public-html has preferences
  206. # [03:29] <Hixie> k
  207. # [03:33] <zcorpan_> karl: right. so hixie referred to the attribute ("Hixie wonders wtf document.dir is")... or are you referring to something else he said?
  208. # [03:35] <karl> scratch scracth
  209. # [03:35] <karl> zcorpan_: I was just saying that I was surprised that hixie didn't know about dir attribute.
  210. # [03:36] <zcorpan_> ah
  211. # [03:37] <zcorpan_> ok. thing was that the Document node can't really have a dir="" attribute
  212. # [03:37] <zcorpan_> so document.dir doesn't make much sense
  213. # [03:38] * Quits: mjs (mjs@64.9.237.122) (Ping timeout)
  214. # [03:39] * Joins: mjs (mjs@64.9.237.122)
  215. # [03:42] * Quits: mjs (mjs@64.9.237.122) (Ping timeout)
  216. # [03:50] * Quits: Zeros (Zeros-Elip@67.154.87.254) (Quit: Leaving)
  217. # [03:51] * Quits: kingryan (rking3@208.66.64.47) (Quit: kingryan)
  218. # [03:53] * Joins: Sander (svl@71.57.109.108)
  219. # [04:13] <karl> MikeSmith: do you some browser developers (not working on Mozilla, Safari, IE, Opera) who are NOT part of the HTML WG?
  220. # [04:13] <karl> do you know
  221. # [04:19] <MikeSmith> yeah. there are a number of browser projects with other engines that aren't yet represented in the group
  222. # [04:20] <MikeSmith> karl ↑
  223. # [04:31] <karl> MikeSmith: see your mail ;)
  224. # [04:36] <MikeSmith> karl - kthx (^з^)-☆Chu!!
  225. # [04:38] * MikeSmith adds new "chuu" macro to his list of annoying 顔文字 kao-moji macros
  226. # [04:40] <karl> ahaha
  227. # [04:41] <karl> MikeSmith: you know what is Chu in chinese?
  228. # [04:43] <karl> when you want to say to someone else that he is being a pig - chu (4th tone) - 豖 http://www.cojak.org/index.php?function=code_lookup&term=8C56
  229. # [04:51] * Quits: mw22 (chatzilla@84.41.169.151) (Quit: Chatzilla 0.9.75-rdmsoft [XULRunner 1.8.0.4/2006060814])
  230. # [04:55] <karl> http://www.w3schools.com/browsers/browsers_stats.asp
  231. # [04:55] <karl> 2007 IE7 IE6 IE5 Fx Moz S O
  232. # [04:55] <karl> April 19.2% 37.3% 1.7% 32.9% 1.3% 1.7% 1.7%
  233. # [05:11] * Quits: zcorpan_ (zcorpan@84.216.43.138) (Ping timeout)
  234. # [05:15] * Joins: hober (ted@68.107.112.172)
  235. # [05:40] * Quits: gavin_ (gavin@74.103.208.221) (Ping timeout)
  236. # [05:45] * Joins: gavin_ (gavin@74.103.208.221)
  237. # [06:20] * Quits: spillner (spillner@141.76.40.118) (Ping timeout)
  238. # [06:21] * Joins: spillner (spillner@141.76.40.118)
  239. # [06:41] * Quits: Sander (svl@71.57.109.108) (Quit: And back he spurred like a madman, shrieking a curse to the sky.)
  240. # [06:53] <MikeSmith> NetSurf after 1.0 – where to from here? (John-Mark Bell) - http://www.drobe.co.uk/extra/NSWTFH.pdf
  241. # [06:54] <MikeSmith> see especially the proposed feature schedule in section 3.5.1
  242. # [06:55] <MikeSmith> [[NetSurf’s current engine is coming close to the end of its design life. This
  243. # [06:55] <MikeSmith> can be seen in the way that dynamic CSS changes simply don’t work, except
  244. # [06:55] <MikeSmith> on a very constrained level. This obviously extends to the impact scripting
  245. # [06:55] <MikeSmith> might have.
  246. # [06:56] <MikeSmith> Section 3.5.1 reflects my own priorities for future development. Stated
  247. # [06:56] <MikeSmith> explicitly, they are; HTML, XHTML, CSS, DOM, JavaScript, and the re-
  248. # [06:56] <MikeSmith> quired layout engine changes.
  249. # [06:56] <MikeSmith> ]]
  250. # [06:57] * Joins: mjs (mjs@64.81.48.145)
  251. # [07:04] <MikeSmith> jmb - nice writeup
  252. # [07:06] * Joins: hyatt (hyatt@24.6.91.161)
  253. # [07:13] <MikeSmith> hmm, I guess linking to the article instead of directly to that PDF is better
  254. # [07:13] <MikeSmith> "Could NetSurf 2.0 support JavaScript?" - http://www.drobe.co.uk/riscos/artifact1909.html
  255. # [07:14] <MikeSmith> the "NetSurf after 1.0 – where to from here?" PDF is the "his memo to fellow NetSurf coders" link in the article
  256. # [07:15] * MikeSmith wonders why jmb sent it out in PDF form instead of HTML ...
  257. # [07:17] <mjs> what's NetSurf?
  258. # [07:19] <MikeSmith> mjs - browser for RISC OS and ported to other platforms
  259. # [07:20] <MikeSmith> and GTK
  260. # [07:21] <MikeSmith> if you want to know more, ask jmb about it when he wakes up
  261. # [07:21] * Quits: sbuluf (jwj@200.49.140.207) (Ping timeout)
  262. # [07:21] <MikeSmith> he's one of the developers
  263. # [07:21] <MikeSmith> http://www.netsurf-browser.org/
  264. # [07:21] <MikeSmith> they just released v1.0 of it
  265. # [07:24] <MikeSmith> it lacks DOM/Javascript support and jmb's doc suggest that its HTML and CSS engines basically need to be rewritten
  266. # [07:26] * Joins: sbuluf (rribn@200.49.140.207)
  267. # [07:27] <MikeSmith> his proposed roadmap for moving it forward is (as far as standards support): New HTML parser + DOM core/html -> New CSS parser (if not ready for 1.2) + CSSOM -> DOM events -> JavaScript
  268. # [07:27] <mjs> what's RISC OS?
  269. # [07:27] <hsivonen> mjs: it is an OS originally developed by Acorn
  270. # [07:27] <mjs> DOM Events don't really make sense without JavaScript so that's kinda weird
  271. # [07:27] <mjs> oh, right
  272. # [07:28] <hsivonen> but how does one get hardware that run RISC OS?
  273. # [07:28] <hsivonen> runs evenn
  274. # [07:28] <mjs> he should just port WebKit, it would save him the 100-200 man-years of effort needed to write a modern real-world compatible browser engine
  275. # [07:42] * MikeSmith drops off to head into Keio SFC
  276. # [07:42] * Quits: MikeSmith (MikeSmith@mcclure.w3.org) (Quit: Get thee behind me, satan.)
  277. # [07:46] * Quits: gavin_ (gavin@74.103.208.221) (Ping timeout)
  278. # [07:51] * Joins: gavin_ (gavin@74.103.208.221)
  279. # [07:58] <karl> http://lists.w3.org/Archives/Public/www-validator/2007Jun/0002.html
  280. # [07:58] <karl> == Beta test for the W3C Markup Validator - version 0.8.0 beta 2 ==
  281. # [07:58] <karl> http://validator-test.w3.org/
  282. # [08:23] * Joins: loic (loic@90.29.111.249)
  283. # [08:25] * Quits: loic (loic@90.29.111.249) (Quit: hoopa rules)
  284. # [08:36] * Joins: mjs_ (mjs@64.81.48.145)
  285. # [08:36] * Quits: mjs (mjs@64.81.48.145) (Connection reset by peer)
  286. # [08:39] * Quits: heycam (cam@130.194.72.84) (Quit: bye)
  287. # [08:47] * Joins: anne (annevk@81.68.67.12)
  288. # [08:48] * Joins: hasather (hasather@81.235.209.174)
  289. # [09:21] * Parts: citoyen (eira@195.139.204.228)
  290. # [09:36] * Joins: heycam (cam@203.214.95.190)
  291. # [09:42] * Joins: tH_ (Rob@87.102.91.218)
  292. # [09:42] * tH_ is now known as tH
  293. # [09:47] * Quits: olivier (ot@128.30.52.30) (Quit: Leaving)
  294. # [09:52] * Joins: edas (edaspet@88.191.34.123)
  295. # [09:54] * Quits: gavin_ (gavin@74.103.208.221) (Ping timeout)
  296. # [09:59] * Joins: gavin_ (gavin@74.103.208.221)
  297. # [10:01] * Joins: icaaq (icaaaq@85.228.55.162)
  298. # [10:18] * Quits: karl (karlcow@128.30.52.30) (Quit: Where dwelt Ymir, or wherein did he find sustenance?)
  299. # [10:30] * Quits: inimino (inimino@75.71.88.233) (Ping timeout)
  300. # [10:36] <anne> hsivonen, I get different results for http://hsivonen.iki.fi/test/iso8859/ISO-8859-11.htm in Opera and IE7...
  301. # [10:38] <hsivonen> anne: does Opera match Firefox, though?
  302. # [10:39] <anne> yeah, it does
  303. # [10:39] <hsivonen> anne: what does IE 7 do with the C1 range?
  304. # [10:39] <anne> the IE7 result seems better
  305. # [10:39] <hsivonen> I wonder what better means in this case :-)
  306. # [10:40] <anne> actually, the byte range gives ‡ for quite a lot of characters
  307. # [10:41] <anne> nm
  308. # [10:41] <anne> it seems Firefox and Opera are identical to IE7 (not thoroughly checked)
  309. # [10:41] <anne> IE7 uses a different replacement character
  310. # [10:41] <anne> one that looks like "<" over here
  311. # [11:09] <hsivonen> anne: so IE maps -11 to Windows-874 as well?
  312. # [11:10] <anne> I suppose
  313. # [11:10] * anne isn't sure what the criteria are
  314. # [11:11] <anne> as in, these are not actual testcases
  315. # [11:11] <anne> more like demos
  316. # [11:11] <hsivonen> anne: right.
  317. # [11:12] <hsivonen> anne: http://www.microsoft.com/globaldev/reference/sbcs/874.mspx for reference
  318. # [11:13] * Quits: hyatt (hyatt@24.6.91.161) (Quit: hyatt)
  319. # [11:20] * Quits: hasather (hasather@81.235.209.174) (Ping timeout)
  320. # [11:27] * Joins: hasather (hasather@81.235.209.174)
  321. # [11:37] * Parts: asbjornu (asbjorn@84.48.116.134)
  322. # [11:40] * Quits: anne (annevk@81.68.67.12) (Ping timeout)
  323. # [11:41] * Joins: ROBOd (robod@86.34.246.154)
  324. # [12:01] * Quits: gavin_ (gavin@74.103.208.221) (Ping timeout)
  325. # [12:06] * Joins: gavin_ (gavin@74.103.208.221)
  326. # [12:11] <beowulf> is there a primer or somesuch for n00bs like me on a group like this? i read all the docs i was asked to in the signup process but still feel like there's a big chuck of missing information
  327. # [12:11] * Joins: asbjornu (asbjorn@84.48.116.134)
  328. # [12:26] * Parts: hasather (hasather@81.235.209.174)
  329. # [12:49] * Joins: pmg (dkgzs@200.49.140.187)
  330. # [12:50] * Quits: pmg (dkgzs@200.49.140.187) (Quit: pmg)
  331. # [12:51] * Quits: sbuluf (rribn@200.49.140.207) (Ping timeout)
  332. # [13:15] * Joins: frippz (fredrikfro@193.11.209.47)
  333. # [14:07] * Quits: gavin_ (gavin@74.103.208.221) (Ping timeout)
  334. # [14:12] * Joins: gavin_ (gavin@74.103.208.221)
  335. # [14:31] * Parts: frippz (fredrikfro@193.11.209.47)
  336. # [14:46] * Parts: beowulf (carisenda@91.84.50.132)
  337. # [15:36] * Joins: zcorpan_ (zcorpan@84.216.43.138)
  338. # [15:42] * Quits: spillner (spillner@141.76.40.118) (Ping timeout)
  339. # [15:54] * Quits: asbjornu (asbjorn@84.48.116.134) (Quit: asbjornu)
  340. # [16:28] * Joins: billmason (billmason@69.30.57.156)
  341. # [16:29] * Joins: anne (annevk@81.68.67.12)
  342. # [16:40] * Quits: gavin_ (gavin@74.103.208.221) (Ping timeout)
  343. # [16:46] * Joins: gavin_ (gavin@74.103.208.221)
  344. # [16:55] * Joins: hasather (hasather@81.235.209.174)
  345. # [16:58] * Parts: anne (annevk@81.68.67.12)
  346. # [17:25] * Quits: edas (edaspet@88.191.34.123) (Quit: http://eric.daspet.name/ et l'édition 2007 de http://www.paris-web.fr/ )
  347. # [17:46] * Joins: Sander (svl@71.57.109.108)
  348. # [17:46] * Joins: anne (annevk@83.82.206.111)
  349. # [17:48] * Joins: dbaron (dbaron@71.198.189.81)
  350. # [17:51] * Quits: mjs_ (mjs@64.81.48.145) (Quit: mjs_)
  351. # [18:04] <anne> Cisco Systems joined the HTML WG
  352. # [18:17] * Joins: DanC_lap (connolly@128.30.52.30)
  353. # [18:41] * Parts: anne (annevk@83.82.206.111)
  354. # [18:41] * Joins: anne (annevk@83.82.206.111)
  355. # [19:02] * Joins: h3h (bfults@66.162.32.234)
  356. # [19:02] * Quits: hasather (hasather@81.235.209.174) (Ping timeout)
  357. # [19:12] * Joins: kingryan (rking3@208.66.64.47)
  358. # [19:17] * Joins: hasather (hasather@81.235.209.174)
  359. # [19:18] * Joins: mjs (mjs@17.255.100.150)
  360. # [19:19] <anne> http://lists.w3.org/Archives/Member/w3c-ac-forum/2007AprJun/0058.html lol
  361. # [19:21] * mjs fears what the XML Core WG may be up to
  362. # [19:22] * Joins: mw22 (chatzilla@84.41.169.151)
  363. # [19:31] <anne> wow, someone is actually reading through the byte by byte algorithm and commenting on it
  364. # [19:34] * Joins: asbjornu (asbjorn@84.48.116.134)
  365. # [19:50] <mjs> where?
  366. # [19:50] <hsivonen> anne: I'm been reading through it, too. And converting it it a derministic algorithm that unreads at most one byte
  367. # [19:50] <hsivonen> I don't like the way the spec assumes a non-deterministic/lookahead decisions
  368. # [19:51] <anne> mjs, http://lists.w3.org/Archives/Public/public-html/2007Jun/0010.html
  369. # [19:51] <hsivonen> I just hadn't quite hit the attribute issue, yet
  370. # [19:51] <anne> hsivonen, it doesn't really assume that, does it?
  371. # [19:51] <anne> hsivonen, it's just written in that way to simplify things I suppose
  372. # [19:52] <hsivonen> anne: I guess, but implementors who don't want to go back and forth in the stream need to rearrange some states
  373. # [19:52] * Quits: hasather (hasather@81.235.209.174) (Client exited)
  374. # [19:53] <hsivonen> anne: as written, it does assume lookahead
  375. # [19:53] * Joins: hasather (hasather@81.235.209.174)
  376. # [19:53] <mjs> unbounded lookahead?
  377. # [19:53] <hsivonen> mjs: no
  378. # [19:55] <hsivonen> but if one does states byte by byte, seeing < means the state can be in one of six cases in the spec
  379. # [19:56] <hsivonen> of course, it isn't hard to implement it differently--just that the main parsing algorithm maps closer to an implementation
  380. # [19:57] * Joins: edas (edaspet@88.191.34.123)
  381. # [19:57] <anne> if the better algorithm is about the same size...
  382. # [19:57] <anne> maybe we should change it
  383. # [20:01] * Quits: dbaron (dbaron@71.198.189.81) (Quit: ->office)
  384. # [20:08] * Joins: NiColasS (for@87.228.148.119)
  385. # [20:08] <NiColasS> Is it necessary to learn CSS 3 ?
  386. # [20:08] <anne> Wrong channel
  387. # [20:08] <NiColasS> thanks
  388. # [20:13] <NiColasS> bye
  389. # [20:13] * Quits: NiColasS (for@87.228.148.119) (Quit: NiColasS)
  390. # [20:14] <mjs> well that was surprisingly effective
  391. # [20:15] <anne> he asked in #css after that
  392. # [20:19] * Quits: edas (edaspet@88.191.34.123) (Ping timeout)
  393. # [20:19] * Joins: dbaron (dbaron@63.245.220.242)
  394. # [20:23] * Parts: hasather (hasather@81.235.209.174)
  395. # [20:24] * Joins: Zeros (Zeros-Elip@67.154.87.254)
  396. # [20:25] * Joins: hasather (hasather@81.235.209.174)
  397. # [20:27] * Joins: tH_ (Rob@87.102.93.94)
  398. # [20:27] * Quits: tH (Rob@87.102.91.218) (Connection reset by peer)
  399. # [20:27] * tH_ is now known as tH
  400. # [20:36] * Quits: DanC_lap (connolly@128.30.52.30) (Client exited)
  401. # [20:36] * Joins: DanC_lap (connolly@128.30.52.30)
  402. # [20:46] <anne> hsivonen, maybe it would indeed be better if charset sniffing was defined as a state machine too
  403. # [20:46] <anne> hsivonen, to make it easier to spot errors
  404. # [20:46] <anne> (and to more easily write a correct version)
  405. # [20:48] * Quits: DanC_lap (connolly@128.30.52.30) (Ping timeout)
  406. # [20:53] * Joins: DanC_lap (connolly@128.30.52.30)
  407. # [21:04] * Quits: gavin_ (gavin@74.103.208.221) (Ping timeout)
  408. # [21:09] * Joins: gavin_ (gavin@74.103.208.221)
  409. # [21:13] * Quits: DanC_lap (connolly@128.30.52.30) (Ping timeout)
  410. # [21:27] * Joins: laplink (link@193.157.66.93)
  411. # [21:39] * Quits: Lachy (chatzilla@203.217.34.167) (Quit: ChatZilla 0.9.78.1 [Firefox 2.0.0.3/2007030919])
  412. # [21:41] * Joins: Lachy (chatzilla@203.217.34.167)
  413. # [21:43] * Parts: hasather (hasather@81.235.209.174)
  414. # [21:44] * Joins: hasather (hasather@81.235.209.174)
  415. # [21:48] * Joins: hyatt (hyatt@24.6.91.161)
  416. # [21:50] * Joins: inimino (inimino@75.71.88.233)
  417. # [22:08] * Quits: ROBOd (robod@86.34.246.154) (Quit: http://www.robodesign.ro )
  418. # [22:16] * Parts: asbjornu (asbjorn@84.48.116.134)
  419. # [22:58] * Parts: hasather (hasather@81.235.209.174)
  420. # [23:03] * Joins: asbjornu (asbjorn@84.48.116.134)
  421. # [23:03] * Joins: hasather (hasather@81.235.209.174)
  422. # [23:12] * Quits: hyatt (hyatt@24.6.91.161) (Connection reset by peer)
  423. # [23:13] * Joins: hyatt (hyatt@24.6.91.161)
  424. # [23:32] * Quits: gavin_ (gavin@74.103.208.221) (Ping timeout)
  425. # [23:37] * Joins: gavin_ (gavin@74.103.208.221)
  426. # [23:51] * Joins: DanC_lap (connolly@128.30.52.30)
  427. # [23:55] * Quits: hyatt (hyatt@24.6.91.161) (Quit: hyatt)
  428. # Session Close: Sat Jun 02 00:00:00 2007

The end :)