/irc-logs / w3c / #html-wg / 2007-07-17 / end

Options:

  1. # Session Start: Tue Jul 17 00:00:00 2007
  2. # Session Ident: #html-wg
  3. # [00:02] * Parts: hasather (hasather@81.235.209.174)
  4. # [00:07] * Quits: hyatt (hyatt@17.203.15.144) (Quit: hyatt)
  5. # [00:08] * Joins: hyatt (hyatt@17.203.15.144)
  6. # [00:08] * Quits: gavin (gavin@74.103.208.221) (Ping timeout)
  7. # [00:09] * Quits: Hixie (ianh@129.241.93.37) (Ping timeout)
  8. # [00:10] * Joins: Hixie (ianh@129.241.93.37)
  9. # [00:13] * Joins: gavin (gavin@74.103.208.221)
  10. # [00:13] * Quits: xover (xover@193.157.66.5) (Ping timeout)
  11. # [00:26] <Philip`> Hixie: http://code.google.com/webstats/2005-12/tables.html says "If someone can explain why so many pages would use a <table> tag and then not put any cells in it, please let us know." - did you count <th> cells?
  12. # [00:27] <Philip`> In ~8K pages, I see 12 with <table> but no <td>. Of those, 7 do have <th> (and one has 173 of them). Of the other 5 with no td/th, four are fairly trivial pages and just have a little bit of content surrounded by a random set of tags, and one is http://www.la-grange.net/w3c/xhtml1/ so you could ask Karl
  13. # [00:29] * Quits: mjs (mjs@64.81.48.145) (Quit: mjs)
  14. # [01:09] * Parts: billmason (billmason@69.30.57.156)
  15. # [01:20] * Joins: sbuluf (dydqaaq@200.49.140.225)
  16. # [01:33] * Joins: inimino (inimino@75.71.88.233)
  17. # [01:34] * Quits: Sander (svl@86.87.68.167) (Quit: And back he spurred like a madman, shrieking a curse to the sky.)
  18. # [01:38] * Joins: mjs (mjs@17.255.96.53)
  19. # [02:04] * Joins: karl (karlcow@128.30.52.30)
  20. # [02:08] * Quits: tH (Rob@87.102.76.26) (Quit: ChatZilla 0.9.78.1-rdmsoft [XULRunner 1.8.0.9/2006120508])
  21. # [02:13] <Philip`> karl: Would you happen to know why http://www.la-grange.net/w3c/xhtml1/ has some empty <table>s? (I was just surprised to see that as an example of the "completely bogus" pages with tables but no cells, like in http://code.google.com/webstats/2005-12/tables.html :-) )
  22. # [02:15] * Quits: gavin (gavin@74.103.208.221) (Ping timeout)
  23. # [02:21] * Joins: gavin (gavin@74.103.208.221)
  24. # [02:22] * Quits: hyatt (hyatt@17.203.15.144) (Quit: hyatt)
  25. # [02:43] * Joins: Lionheart (robin@66.57.69.65)
  26. # [02:57] <karl> Philip`: I wonder if I screwed my markup at a point :) I can check
  27. # [03:01] <karl> indeed the source code was really weird. I have fixed it.
  28. # [03:01] <karl> thanks
  29. # [03:16] * Joins: olivier (ot@128.30.52.30)
  30. # [03:16] <karl> hhhmmm it seems they are working hard in the SYMM WG.
  31. # [03:16] <karl> http://www.w3.org/TR/2007/WD-SMIL3-20070713/
  32. # [03:17] <karl> Last Call
  33. # [04:22] * Quits: gavin (gavin@74.103.208.221) (Ping timeout)
  34. # [04:27] * Joins: gavin (gavin@74.103.208.221)
  35. # [04:56] <karl> http://lists.w3.org/Archives/Public/public-html/mboxes/
  36. # [05:00] * Quits: mjs (mjs@17.255.96.53) (Quit: mjs)
  37. # [05:02] * Joins: mjs (mjs@17.255.96.53)
  38. # [05:03] * Quits: Lachy (chatzilla@203.214.140.60) (Quit: ChatZilla 0.9.78.1 [Firefox 2.0.0.4/2007051502])
  39. # [05:06] * Quits: mjs (mjs@17.255.96.53) (Connection reset by peer)
  40. # [05:12] * Joins: mjs (mjs@17.255.96.53)
  41. # [05:25] * Joins: MikeSmith (MikeSmith@mcclure.w3.org)
  42. # [05:51] * Joins: mjs_ (mjs@17.203.15.221)
  43. # [05:53] * Quits: mjs_ (mjs@17.203.15.221) (Quit: mjs_)
  44. # [05:53] * Quits: mjs (mjs@17.255.96.53) (Ping timeout)
  45. # [06:41] * Joins: mjs (mjs@17.203.15.221)
  46. # [07:36] * Quits: gavin (gavin@74.103.208.221) (Ping timeout)
  47. # [07:41] * Joins: gavin (gavin@74.103.208.221)
  48. # [07:53] * Quits: MikeSmith (MikeSmith@mcclure.w3.org) (Quit: Less talk, more pimp walk.)
  49. # [08:01] * Joins: xover (xover@193.157.66.5)
  50. # [08:03] * Joins: mjs_ (mjs@17.203.15.221)
  51. # [08:03] * Quits: mjs (mjs@17.203.15.221) (Connection reset by peer)
  52. # [08:22] <karl> hmm interesting discussions ----> http://www.w3.org/2007/05/14-html-wg-irc#T21-24-33
  53. # [08:23] <karl> about parsing requirements for the charset for example
  54. # [08:23] <karl> We had the same discussion olivier and I about the validator.
  55. # [08:23] <karl> When and how to decide to stop preparsing for guessing the charset.
  56. # [08:23] <karl> or even content type information
  57. # [08:23] <karl> It raises another related question.
  58. # [08:25] <karl> If the specs says "stop after 512 bytes" for user agents, then it should say "when writing HTML documents, you have to put the charset information before the 512 bytes with the end of the charset no further than 512 bytes."
  59. # [09:12] <mjs_> I really shouldn't read HTML WG email, it's just depressing
  60. # [09:12] * mjs_ is now known as mjs
  61. # [09:14] <karl> mjs: ;)
  62. # [09:35] * Quits: olivier (ot@128.30.52.30) (Quit: Leaving)
  63. # [09:36] <karl> http://lists.w3.org/Archives/Public/public-html/2007Jul/0746
  64. # [09:37] <karl> HTML 5 parsers list by Henri Sivonen
  65. # [09:40] * Quits: karl (karlcow@128.30.52.30) (Quit: Where dwelt Ymir, or wherein did he find sustenance?)
  66. # [09:44] * Quits: gavin (gavin@74.103.208.221) (Ping timeout)
  67. # [09:49] * Joins: gavin (gavin@74.103.208.221)
  68. # [10:09] * Joins: MikeSmith (MikeSmith@mcclure.w3.org)
  69. # [10:11] * Quits: mjs (mjs@17.203.15.221) (Connection reset by peer)
  70. # [10:11] * Joins: mjs_ (mjs@17.203.15.221)
  71. # [10:23] * Joins: hasather (hasather@81.235.209.174)
  72. # [10:30] * Joins: Lachy (chatzilla@203.214.140.60)
  73. # [10:34] <Lachy> I just don't understand how some people can blatantly ignore evidence and make bogus claims about XHTML2's compatibility
  74. # [10:34] <schepers> I think it's a matter of having different criteria for compatibility
  75. # [10:36] <Lachy> the major criteria is whether or not a) legacy markup will still work in future UAs, and b) future markup will degrade gracefully in legacy UAs
  76. # [10:36] <Lachy> XHTML2 doesn't fit either of those
  77. # [10:36] <schepers> that's your major criteria, sure
  78. # [10:36] <Lachy> what else is there?
  79. # [10:38] <schepers> well, Rob Burns seems very concerned with the semantics of the elements and attributes... to him, that's pretty important... I thought he'd made that clear
  80. # [10:38] * Joins: mjs (mjs@17.203.15.221)
  81. # [10:38] * Quits: mjs_ (mjs@17.203.15.221) (Ping timeout)
  82. # [10:38] <schepers> I'm not too concerned either way, myself
  83. # [10:39] <schepers> I have lots of opinions on it, but none of them have resolved to a single path
  84. # [10:39] <Lachy> yeah, but I haven't seen him describe any specific semantic incompatibilities (though, perhaps I missed it if he did)
  85. # [10:42] <schepers> anyway, this isn't actually a technical issue, but a political one
  86. # [10:42] <mjs> Lachy: Robert Burns has pretty clearly announced that he has different criteria for what constitutes "namespace integrity"
  87. # [10:42] <schepers> both groups want control of the XHTML "brand"
  88. # [10:42] <mjs> or whatever
  89. # [10:42] <mjs> I actually don't care about the brand, but the namespace URI is not negotiable
  90. # [10:43] <mjs> (the one that HTML5 will use, that is; don't really care what XHTML2 does)
  91. # [10:43] <schepers> well, you've actually argued about the name (the brand) before
  92. # [10:43] <mjs> I also don't think we have any reason to try to be compatible with XHTML2
  93. # [10:44] <schepers> again, it's my opinion that this isn't a technical argument per se
  94. # [10:44] <mjs> the namespace URI argument has technical impact
  95. # [10:44] <schepers> even if it's being fought with technical terms
  96. # [10:44] <schepers> yeah, but not as clear-cut as all that
  97. # [10:44] <Lachy> schepers: the namespace issue is entirely technical
  98. # [10:45] <Lachy> the branding issue is more political
  99. # [10:45] <schepers> I think it's part and parcel
  100. # [10:45] <Lachy> in what way is the namespace issue political?
  101. # [10:45] <schepers> the technical aspect of the ns issue is just a battle in the war of the brand, as far as I can see
  102. # [10:46] <mjs> it's pretty clear-cut to me, HTML5 has to use the XHTML1 namespace to support existing XHTML1 content, which is a goal
  103. # [10:46] <mjs> on the XHTML2 side it is political and as far as I can tell dishonest
  104. # [10:46] <mjs> the idea that the XHTML2 spec has a different namespace but that's just a placeholder smells of revisionist history
  105. # [10:47] <schepers> I think the same change could be leveled at both sides
  106. # [10:47] <schepers> which is why I'm getting increasingly bored with it
  107. # [10:47] <Lachy> yeah, the XHTML2 side hasen't put forth any technical argument for having the XHTML1 namespace, and have only been ignorant/dismissive of the technical arguments from the HTML5 side
  108. # [10:48] <schepers> including the feigned outrage
  109. # [10:48] <mjs> yeah, that's almost as much fun as being ostentatiously above-it-all
  110. # [10:48] <schepers> Lachy, I don't think the HTML5 side has any room to talk about being dismissive
  111. # [10:49] <Lachy> schepers: care to elaborate?
  112. # [10:49] <schepers> nah, being ostentatiously above-it-all is much more fun, since you can laugh at both sides
  113. # [10:49] <schepers> its the only way I can keep calm about the whole thing
  114. # [10:50] <mjs> well, I'm not quite ready to put Robert Burns in the Dmitry Turin bucket
  115. # [10:50] <schepers> Dmitry is entertaining
  116. # [10:51] <mjs> so I try to actually engage
  117. # [10:51] <schepers> I think Burns aluded to having some sort of product... not sure it's a browser, or authoring tool, or what
  118. # [10:51] <schepers> alluded?
  119. # [10:52] <schepers> I think Burns is being very straightforward
  120. # [10:52] <Lachy> Dmitry is more annoying than entertaining for me
  121. # [10:52] <schepers> well, I can understand that
  122. # [10:53] <schepers> I've deliberately taken a step back to ease my blood pressure... otherwise, reading public-html is like having road-rage for me
  123. # [10:53] <schepers> I'm only getting involved in issues I think are important
  124. # [10:53] <Lachy> yeah, that's probably a good idea
  125. # [10:54] <schepers> well, not exactly... a lot of it's important... but i'm only jumping in where I think the direction is going someplace I can't accept
  126. # [10:54] <Lachy> I think I'm going to have to stay right out of XHTML2 discussions from now on
  127. # [10:54] <schepers> the rest of it, I figure will work itself out somehow
  128. # [10:55] <mjs> I think I will limit comments on XHTML2 to mentioning that XHTML2 is off-topic, should that seem necessary
  129. # [10:56] <schepers> well, is it?
  130. # [10:56] <schepers> I mean, the whole NS issue isn't settled... nor is the name issue
  131. # [10:56] <schepers> or did I miss something?
  132. # [10:58] <Lachy> schepers: IMHO, the name issue is settled in favour of calling it XHTML5 and ignoring the XHTML2 WG's objections, and the namespace issue is settled in favour of XHTML5 reusing the XHTML1 namespace, and letting the XHTML2 WG use whatever they like and ignoring them
  133. # [10:58] <mjs> a proposal to change the namespace wouldn't be off-topic, just ill-conceived
  134. # [10:58] <mjs> talking about compatibility with XHTML2 is off-topic
  135. # [10:58] <schepers> Lachy: I think that's an opinion, not a resolution
  136. # [10:58] <schepers> like I said, unless I missed something (which is very possible)
  137. # [10:59] <schepers> mjs: sure, I can see that
  138. # [10:59] <Lachy> well, it's resolved as far has the HTMLWG is concerned. Let the XHTML2 WG come up with whatever resolution they like for themselves, as long as their resolution doesn't affect us.
  139. # [11:00] <schepers> was there a formal resolution?
  140. # [11:00] <Lachy> not a formal one
  141. # [11:00] <schepers> then it's not resolved
  142. # [11:01] <Lachy> but there's no point, since there have been no solid technical arguments against that resolution
  143. # [11:02] <mjs> we didn't have an official group vote or anything
  144. # [11:02] <mjs> but also no one seems motivated enough to call for one
  145. # [11:03] <hsivonen> schepers: the MIME type and the namespace URI are not negotiable for the XML serialization of HTML5, but if those say "xhtml", not calling the thing "XHTML something" is silly
  146. # [11:08] <hsivonen> Philip`: is the zaynar.demon Philip Taylor you?
  147. # [11:29] * Joins: ROBOd (robod@86.34.246.154)
  148. # [11:41] * Joins: zcorpan_ (zcorpan@90.229.146.10)
  149. # [12:11] * Quits: sbuluf (dydqaaq@200.49.140.225) (Quit: sbuluf)
  150. # [12:12] * Quits: Lachy (chatzilla@203.214.140.60) (Quit: ChatZilla 0.9.78.1 [Firefox 2.0.0.4/2007051502])
  151. # [12:24] * Joins: Sander (svl@86.87.68.167)
  152. # [12:25] * Parts: gorme (gorm@213.236.208.22)
  153. # [12:25] * Joins: gorme (gorm@213.236.208.22)
  154. # [12:26] * Joins: mjs_ (mjs@17.203.15.221)
  155. # [12:26] * Quits: mjs (mjs@17.203.15.221) (Connection reset by peer)
  156. # [12:30] * Joins: myakura (myakura@58.88.37.26)
  157. # [12:37] * Quits: gsnedders (gsnedders@81.132.88.104) (Quit: gsnedders)
  158. # [12:40] * Joins: mjs (mjs@17.255.96.53)
  159. # [12:41] * Quits: mjs_ (mjs@17.203.15.221) (Ping timeout)
  160. # [12:53] * Joins: gsnedders (gsnedders@81.132.88.104)
  161. # [12:55] * Joins: tH (Rob@87.102.76.26)
  162. # [13:24] <Philip`> hsivonen: Yes
  163. # [13:25] <Philip`> (I probably should get a proper domain and email address at some point, rather than just rotating between all the ones I currently have access to...)
  164. # [13:32] * Quits: gavin (gavin@74.103.208.221) (Ping timeout)
  165. # [13:37] * Joins: gavin (gavin@74.103.208.221)
  166. # [14:15] * Quits: myakura (myakura@58.88.37.26) (Quit: Leaving...)
  167. # [14:24] <zcorpan_> Philip`: would it be possible to list the top values of attributes? like http://canvex.lazyilluminati.com/misc/stats/2/analyse.cgi/tagattr/a/target
  168. # [14:28] <Philip`> zcorpan_: I don't think I can really do that in general, since I'd have to store every attribute value in order to count them at the end, and attribute values make up about a third of all the bytes in pages, so that's too much data to store easily
  169. # [14:28] <Philip`> (or am I missing cleverer ways to count attribute value frequency?)
  170. # [14:29] <Philip`> But I could do it much more easily if there's just a small set of attributes that are worth counting
  171. # [14:29] * Quits: hasather (hasather@81.235.209.174) (Ping timeout)
  172. # [14:29] <zcorpan_> ok
  173. # [14:40] * Quits: gsnedders (gsnedders@81.132.88.104) (Quit: gsnedders)
  174. # [14:42] * Joins: hasather (hasather@81.235.209.174)
  175. # [14:44] * Joins: gsnedders (gsnedders@81.132.88.104)
  176. # [14:45] <zcorpan_> uses of <noscript>: ads (document.write+<noscript>), tracking (script + <noscript><img>), inclusion of flash (plugin detection with script + <noscript><object>)
  177. # [14:45] * Quits: hasather (hasather@81.235.209.174) (Ping timeout)
  178. # [14:45] <zcorpan_> from analysing the first 10 pages in http://canvex.lazyilluminati.com/misc/stats/2/analyse.cgi/tag/noscript
  179. # [14:49] <zcorpan_> the first doesn't work at all in xml. the second neither, since <img>s are always loaded (except in opera). the third might work
  180. # [14:49] <Philip`> Maybe I shouldn't sort the list of pages by the number of occurrences? The first ones in the list may not be very representative, since they're using the feature an abnormal number of times
  181. # [14:50] <zcorpan_> indeed
  182. # [15:01] * Joins: hasather (hasather@81.235.209.174)
  183. # [15:01] <zcorpan_> from the bottom 10 pages: normal content (document.write+<noscript>), tracking, inclusion of css (<noscript><style>), ads, inclusion of navigation (document.write+<noscript><table>)
  184. # [15:10] <Philip`> zcorpan_: http://canvex.lazyilluminati.com/misc/stats/2/analyse.cgi/pages/tag/noscript
  185. # [15:11] <Philip`> ...should provide a more uniform selection of pages
  186. # [15:20] <Philip`> It's a bit odd how I see different element distribution compared to http://triin.net/2006/06/12/HTML, when looking at the same set of pages
  187. # [15:21] <Philip`> since I think a sample of ~8K pages ought to be something like +/- 1% of the true mean, but there are bigger variations than that
  188. # [15:23] <Philip`> though that one was a year ago, so maybe that's the reason for the changes
  189. # [15:30] * Quits: hasather (hasather@81.235.209.174) (Ping timeout)
  190. # [15:34] <Philip`> It's slightly irritating that I can't find the actual numbers used to generate the graphs like http://triin.net/archive/kool/webstat/figure-13.png, but it looks like there's a notable increase in meta, link, script, span, div
  191. # [15:35] <Philip`> and a notable decrease in table, tr, td, font, b, center
  192. # [15:36] <Philip`> (By "notable" I mean something around 2-3%)
  193. # [15:38] <Philip`> The XHTML 1.0 Transitional doctype has gone up from ~5% to ~11%
  194. # [15:38] <Philip`> If things have changed that much in a year, maybe the web isn't actually a lost cause
  195. # [15:39] * Quits: gavin (gavin@74.103.208.221) (Ping timeout)
  196. # [15:40] <zcorpan_> looking through the first few in that list, i find an empty <noscript> and a <noscript><meta refresh>
  197. # [15:41] <zcorpan_> where the meta refresh goes to the same page but with a query string
  198. # [15:44] <zcorpan_> of the 30-40 pages i've looked at now, i haven't found one single saying "Seems like your browser doesn't support JS!" or anything similar
  199. # [15:44] * Joins: gavin (gavin@74.103.208.221)
  200. # [15:45] <zcorpan_> but perhaps that's more common on web apps
  201. # [15:47] * Quits: gavin (gavin@74.103.208.221) (Ping timeout)
  202. # [15:50] <zcorpan_> http://www.google.com/search?q=you-have-javascript-disabled
  203. # [15:55] * Quits: gavin_ (gavin@63.245.208.169) (Ping timeout)
  204. # [16:00] <Philip`> Eek - 5% of dmoz.org's data is CNN.com
  205. # [16:01] <Philip`> At least it looks like they don't change their old pages when they redesign the site, so they wouldn't single-handedly cause a 5% shift in all the measured statistics...
  206. # [16:01] * Joins: gavin (gavin@74.103.208.221)
  207. # [16:02] * Joins: gavin_ (gavin@63.245.208.169)
  208. # [16:14] * Joins: nickshanks (nicholas@195.137.85.17)
  209. # [16:21] * Joins: jdandrea (jdandrea@24.228.42.231)
  210. # [16:23] * Quits: xover (xover@193.157.66.5) (Ping timeout)
  211. # [16:24] * Joins: xover (xover@193.157.66.5)
  212. # [16:27] <zcorpan_> do i understand it correctly that DOMTokenList.add() appends a space and the token if the underlying string is empty?
  213. # [16:28] <zcorpan_> "4. Otherwise, if the last character of the DOMTokenList object's underlying string is not a space character, then append a U+0020 SPACE character to the end of that string."
  214. # [16:28] * zcorpan_ wonders if that is actually intentional
  215. # [16:34] * Joins: billmason (billmason@69.30.57.156)
  216. # [16:41] * Quits: jdandrea (jdandrea@24.228.42.231) (Quit: ciao)
  217. # [17:11] * Joins: NiColasS (nicolas@213.7.50.213)
  218. # [17:11] <NiColasS> Hi! markup language is a language that gives instructions to a web browser how to display the pages ?
  219. # [17:13] <NiColasS> is it correct ?
  220. # [17:13] <zcorpan_> gsnedders: hmm. i wouldn't want xhtml5 to disallow any element when used in compound documents... if another spec says that element X may contain html:Y as child, then that's what it is...
  221. # [17:13] <zcorpan_> NiColasS: not really
  222. # [17:14] <zcorpan_> NiColasS: does http://en.wikipedia.org/wiki/Markup_language help?
  223. # [17:15] <NiColasS> I have read it but i couldn't understand a lot...i think because I don't know english very well and i can't understand some words
  224. # [17:16] <zcorpan_> NiColasS: a markup language doesn't necessarily have anything to do with instructions, browsers, presentation or web pages
  225. # [17:17] <NiColasS> well I will try to learn this article again having a dictionary next to me :P
  226. # [17:17] <zcorpan_> NiColasS: think of it as a kind of rich text -- text with extra information about the text, where the extra information is the markup
  227. # [17:18] <NiColasS> what markup means ? i couldn't find it in the dictionary :P
  228. # [17:19] <zcorpan_> "Computer Science. The collection of tags that describe the specifications of an electronic document, as for formatting." -- http://www.answers.com/markup?cat=biz-fin&nafid=3
  229. # [17:26] <NiColasS> thanks! i will read wikipedia and see:P
  230. # [17:28] <zcorpan_> http://simon.html5.org/test/html/dom/reflecting/DOMTokenList/ is now done
  231. # [17:30] <NiColasS> xhtml is a recommendation ? :P
  232. # [17:30] <NiColasS> w3.org : XHTML Basic
  233. # [17:30] <NiColasS> 1.1 Is a Candidate Recommendation
  234. # [17:31] <gsnedders> hmm… /TR/xhtml11/ links to a WD of the Second Edition of XHTML 1.1
  235. # [17:35] <zcorpan_> i think i have reviewed everything in section 2 now, except for .innerHTML in HTML and document.write(), which i don't intend to review now
  236. # [17:37] <zcorpan_> and i'm not done with all members of the HTMLDocument and HTMLElement interfaces
  237. # [17:37] <zcorpan_> perhaps i should continue with those now
  238. # [17:38] * Quits: nickshanks (nicholas@195.137.85.17) (Quit: nickshanks)
  239. # [17:40] <zcorpan_> "All Document objects (in user agents implementing this specification) must also implement the HTMLDocument interface" don't have tests for that yet either
  240. # [17:42] <zcorpan_> i might also want to go through all dom attributes in browsers to see if they are of the right type etc
  241. # [17:50] * Philip` makes his HTML-survey-reporting script five times faster, by using PostgreSQL instead of SQLite
  242. # [17:50] * Joins: polin8 (polin8@209.176.7.3)
  243. # [17:50] * Quits: polin8 (polin8@209.176.7.3) (Quit: :wq)
  244. # [17:56] * tH wishes there was a robots.txt5 spec
  245. # [18:00] <Philip`> robots.xml
  246. # [18:01] <Sander> robots.rdf!
  247. # [18:03] * Quits: gavin (gavin@74.103.208.221) (Ping timeout)
  248. # [18:03] <Philip`> robots.docx
  249. # [18:03] * Joins: dbaron (dbaron@71.198.189.81)
  250. # [18:05] <Sander> the scary thing is that if I google for robots.rdf, I get 90 hits, many of which appear to be seriously proposing it
  251. # [18:08] * Joins: gavin (gavin@74.103.208.221)
  252. # [18:10] * Quits: NiColasS (nicolas@213.7.50.213) (Quit: NiColasS)
  253. # [18:38] <zcorpan_> http://simon.html5.org/test/html/dom/interfaces/Document/
  254. # [19:30] * Quits: MikeSmith (MikeSmith@mcclure.w3.org) (Quit: Less talk, more pimp walk.)
  255. # [19:31] <Philip`> http://canvex.lazyilluminati.com/survey/2007-07-17/analyse.cgi/index - same data but hopefully faster at loading
  256. # [19:32] <zcorpan_> Philip`: is it regenerated each time, or cached?
  257. # [19:32] <Philip`> Still regenerated each time
  258. # [19:33] <Philip`> since caching seems not entirely trivial when there are so many possible pages
  259. # [19:33] <zcorpan_> ok
  260. # [19:34] <Philip`> and it's sufficiently non-slow that I don't feel particularly compelled to make it faster yet :-)
  261. # [19:35] <Philip`> http://canvex.lazyilluminati.com/svn/survey/trunk/ + http://canvex.lazyilluminati.com/svn/tokeniser/ is the relevant code I used, though it's not especially good or easy to use and people may want to find much better ways to do this
  262. # [19:37] * Philip` will try sending some information to public-html
  263. # [19:38] <Philip`> Oh, and I guess I should work out how to stop Googlebot traversing hundreds of thousands of pages and stealing all my CPU cycles
  264. # [19:39] <zcorpan_> "index, nofollow"?
  265. # [19:40] <zcorpan_> <meta name=robots content=index,nofollow>
  266. # [19:42] <Philip`> Do as many people follow that as robots.txt?
  267. # [19:42] <zcorpan_> people?
  268. # [19:43] <Philip`> Well, robotic people
  269. # [19:43] <zcorpan_> google does, at least
  270. # [20:11] * Quits: gavin (gavin@74.103.208.221) (Ping timeout)
  271. # [20:16] * Joins: gavin (gavin@74.103.208.221)
  272. # [20:40] * Quits: dbaron (dbaron@71.198.189.81) (Quit: 8403864 bytes have been tenured, next gc will be global.)
  273. # [20:44] * Quits: Sander (svl@86.87.68.167) (Quit: And back he spurred like a madman, shrieking a curse to the sky.)
  274. # [20:58] <DanC> anybody seen mjs? I'm trying to sync on design principles.
  275. # [21:00] <mjs> DanC: I'm here, about to go to lunch, will check in with you after lunch
  276. # [21:00] <DanC> ah. good.
  277. # [21:23] * Joins: Sander (svl@86.87.68.167)
  278. # [21:41] * Joins: dbaron (dbaron@63.245.220.241)
  279. # [21:59] * Joins: hyatt (hyatt@24.6.91.161)
  280. # [22:18] * Quits: gavin (gavin@74.103.208.221) (Ping timeout)
  281. # [22:23] * Joins: gavin (gavin@74.103.208.221)
  282. # [22:32] * Quits: zcorpan_ (zcorpan@90.229.146.10) (Ping timeout)
  283. # [22:49] * Joins: zcorpan_ (zcorpan@90.229.146.10)
  284. # [22:49] <zcorpan_> hm. on boolean attributes
  285. # [22:49] <zcorpan_> "If the attribute is present, its value must either be the empty string or the attribute's canonical name, exactly, with no leading or trailing whitespace, and in lowercase."
  286. # [22:50] <zcorpan_> it was case insensitive in html4
  287. # [22:53] <mjs> doesn't seem that important for non-lowercase versions to be conforming
  288. # [22:54] <zcorpan_> keywords in enumerated attributes in html5 may use any mix of lowercase and uppercase
  289. # [22:54] <zcorpan_> i don't see why boolean attributes should be different from enumerated attributes
  290. # [22:57] * Quits: Sander (svl@86.87.68.167) (Quit: And back he spurred like a madman, shrieking a curse to the sky.)
  291. # [22:57] * Quits: hyatt (hyatt@24.6.91.161) (Client exited)
  292. # [22:58] * Joins: hyatt (hyatt@24.6.91.161)
  293. # [23:13] * Quits: ROBOd (robod@86.34.246.154) (Quit: http://www.robodesign.ro )
  294. # [23:20] * Joins: Sander (svl@86.87.68.167)
  295. # [23:24] * Quits: zcorpan_ (zcorpan@90.229.146.10) (Ping timeout)
  296. # [23:34] * Quits: mjs (mjs@17.255.96.53) (Quit: mjs)
  297. # [23:35] * Joins: mjs (mjs@17.255.96.53)
  298. # Session Close: Wed Jul 18 00:00:00 2007

The end :)