/irc-logs / w3c / #html-wg / 2008-01-07 / end

Options:

  1. # Session Start: Mon Jan 07 00:00:00 2008
  2. # Session Ident: #html-wg
  3. # [00:07] * Quits: tH (Rob@87.102.20.204) (Quit: ChatZilla 0.9.79-rdmsoft [XULRunner 1.8.0.9/2006120508])
  4. # [00:18] * Quits: Lachy__ (Lachlan@84.215.54.100) (Ping timeout)
  5. # [00:24] * Joins: Lachy__ (Lachlan@88.91.105.112)
  6. # [00:34] * Quits: Lachy__ (Lachlan@88.91.105.112) (Ping timeout)
  7. # [00:38] * Quits: jgraham (james@81.86.215.9) (Quit: This computer has gone to sleep)
  8. # [00:40] * Joins: Lachy__ (Lachlan@84.215.54.100)
  9. # [00:48] * Joins: shepazu (schepers@128.30.52.30)
  10. # [00:51] * Quits: Jotschi (java@80.109.65.213) (Quit: Verlassend)
  11. # [01:00] * Quits: hasather (hasather@90.231.107.133) (Ping timeout)
  12. # [01:03] * Joins: hasather (hasather@90.231.107.133)
  13. # [01:57] * Joins: olivier (ot@128.30.52.30)
  14. # [02:12] * Quits: hasather (hasather@90.231.107.133) (Quit: leaving)
  15. # [02:16] * Joins: jgraham (james@81.86.215.9)
  16. # [02:16] * anne reads comments on http://tech.gtaero.net/2008/01/why-html-should-become-dead-language.html
  17. # [02:16] * anne wonders why XML 1.1 is considered a minor change...
  18. # [02:18] * Quits: jgraham (james@81.86.215.9) (Quit: This computer has gone to sleep)
  19. # [02:23] <Philip> anne: That's "by comparison", not an absolute
  20. # [02:24] <anne> XML 1.1 is completely incompatible with XML 1.0 where as "XML5" would be completely compatible with XML 1.0 (and XML 1.1)
  21. # [02:27] <Philip> It's still relatively minor in terms of specification and implementation differences from XML 1.0, regardless of how big a problem adoption is
  22. # [02:30] <Philip> (and in terms of design-philosophy differences)
  23. # [02:31] <anne> I'm not sure I see your point. Sure, the differences between XML 1.0 and 1.1 are not too big (in fact, most likely XML 1.0 will be changed by the XML Core WG and XML 1.1 will be obsoleted), but the versioning design stopped adoption. XML5 does not have that versioning design issue.
  24. # [02:34] <Philip> My point is just that XML 1.1 can be considered a minor change because there aren't many differences
  25. # [02:34] <Philip> which isn't a very interesting point, but it's what you seemed to be wondering :-)
  26. # [02:35] <anne> I see, time to go to bed
  27. # [02:35] * Quits: shepazu (schepers@128.30.52.30) (Client exited)
  28. # [02:35] <anne> g'night
  29. # [02:35] * Joins: shepazu (schepers@128.30.52.30)
  30. # [02:36] <Philip> Night
  31. # [03:12] * Quits: Yudai (Yudai@218.221.196.114) (Ping timeout)
  32. # [03:14] * Joins: Yudai (Yudai@218.221.196.114)
  33. # [04:13] * Joins: Zeros (Zeros-Elip@69.140.40.140)
  34. # [04:34] * Joins: aaronlev (chatzilla@209.6.168.245)
  35. # [05:05] * Quits: gavin_ (gavin@99.227.30.12) (Quit: gavin_)
  36. # [05:09] * Joins: gavin_ (gavin@99.227.30.12)
  37. # [05:09] * Quits: aaronlev (chatzilla@209.6.168.245) (Quit: ChatZilla 0.9.79 [Firefox 3.0b3pre/2008010605])
  38. # [06:05] * Quits: sbuluf (cer@200.49.132.97) (Ping timeout)
  39. # [06:37] * Quits: jmb (jmb@152.78.68.189) (Ping timeout)
  40. # [06:42] * Joins: jmb (jmb@152.78.68.189)
  41. # [07:13] * Quits: xover (xover@193.157.66.5) (Quit: Leaving)
  42. # [07:26] * Joins: xover (xover@193.157.66.5)
  43. # [07:30] * Quits: gavin_ (gavin@99.227.30.12) (Ping timeout)
  44. # [07:35] * Joins: gavin_ (gavin@99.227.30.12)
  45. # [07:52] * Quits: laplink (link@193.157.66.108) (Quit: This computer has gone to sleep)
  46. # [08:33] * Quits: olivier (ot@128.30.52.30) (Quit: Leaving)
  47. # [08:41] * Joins: laplink (link@193.157.66.108)
  48. # [08:55] * Joins: MikeSmith (MikeSmith@mcclure.w3.org)
  49. # [09:07] * Joins: billmason (billmason@67.169.196.81)
  50. # [09:08] * Quits: billmason (billmason@67.169.196.81) (Quit: billmason)
  51. # [09:08] * Joins: billmason (billmason@67.169.196.81)
  52. # [09:10] * Quits: billmason (billmason@67.169.196.81) (Quit: billmason)
  53. # [09:34] * Quits: MikeSmith (MikeSmith@mcclure.w3.org) (Quit: Less talk, more pimp walk.)
  54. # [09:40] * Joins: MikeSmith (MikeSmith@mcclure.w3.org)
  55. # [09:43] * Quits: jmb (jmb@152.78.68.189) (Ping timeout)
  56. # [09:43] * Joins: jmb (jmb@152.78.68.189)
  57. # [09:59] * Joins: zcorpan (zcorpan@88.131.66.80)
  58. # [10:05] * Joins: jgraham (james@81.86.215.9)
  59. # [10:18] * Joins: ROBOd (robod@89.122.216.38)
  60. # [10:21] * Quits: Lachy__ (Lachlan@84.215.54.100) (Quit: This computer has gone to sleep)
  61. # [10:32] * Quits: Zeros (Zeros-Elip@69.140.40.140) (Quit: Leaving)
  62. # [10:38] * Quits: jgraham (james@81.86.215.9) (Quit: This computer has gone to sleep)
  63. # [10:39] * Joins: Lachy (Lachlan@213.236.208.22)
  64. # [10:42] * Quits: Lachy (Lachlan@213.236.208.22) (Quit: Leaving)
  65. # [10:42] * Joins: Lachy (Lachlan@213.236.208.22)
  66. # [11:46] * Quits: MikeSmith (MikeSmith@mcclure.w3.org) (Quit: Less talk, more pimp walk.)
  67. # [11:51] * Joins: jgraham (james@81.86.215.9)
  68. # [11:56] * Joins: myakura (myakura@220.104.127.20)
  69. # [12:52] <hsivonen> Philip: XML 1.1 is *not* a minor change from the implementation point of view
  70. # [12:53] <anne> according to the guy who implemented it for Opera it is
  71. # [12:53] <hsivonen> Philip: there's craziness like the meaning of XSD datatypes changing depending on the XML version. A proper implementation would need to attach an XML version to just about everything that flows through the XML tubes
  72. # [12:54] <hsivonen> anne: does Opera only change change what counts as a Name in the parser?
  73. # [12:55] <hsivonen> sure, doing only that makes it seem easy
  74. # [12:56] <hsivonen> anne: in Opera, the higher layers are presumably anything-goes anyway and don't need a flag
  75. # [12:56] <anne> I believe we follow the specification
  76. # [12:56] <hsivonen> anne: what if you have an XSD validator somewhere in there
  77. # [12:56] <hsivonen> anne: what happens if you parse an XML 1.1 document into DOM and later serialize it using XHR?
  78. # [12:56] <anne> I don't know. I also don't really care about XSD or XML 1.1 so I'll leave at this :)
  79. # [12:57] <hsivonen> ok :-)
  80. # [13:03] * Joins: olivier (ot@128.30.52.30)
  81. # [13:10] * Quits: laplink (link@193.157.66.108) (Quit: This computer has gone to sleep)
  82. # [13:19] <Philip> hsivonen: Ah, okay
  83. # [13:20] <Philip> Are all those problems avoided by XML5?
  84. # [13:24] * Joins: MikeSmith (MikeSmith@mcclure.w3.org)
  85. # [13:32] <hsivonen> Philip: that depends on whether XML5-ingesting apps go XML5 all the way or keep alive the capability to do XML 1.0 processing
  86. # [13:33] <hsivonen> Philip: the main questions are: 1) How will XML5 Name and NCName interact with XSD datatypes and xml:id and 2) will you ever want to serialize XML 1.0 (as opposed to XML5) from an XML5-capable pipeline
  87. # [13:34] <hsivonen> Philip: my answer to #2 is "yes"
  88. # [13:34] <hsivonen> and that's a problem
  89. # [13:50] * Joins: Julian (chatzilla@217.91.35.233)
  90. # [14:02] <Philip> http://www.joelonsoftware.com/items/2007/12/03.html - "If the spec defines precisely what a program will do, with enough detail that it can be used to generate the program itself, this just begs the question: how do you write the spec? Such a complete spec is just as hard to write as the underlying computer program, because just as many details have to be answered by spec writer as the programmer."
  91. # [14:03] <Philip> That seems untrue - spec writers don't have to worry about performance, which means they have many fewer decisions to make
  92. # [14:05] <anne> performance, memory usage, limits, etc.
  93. # [14:10] <Philip> Implementation is just an optimising transformation of the specification's algorithm
  94. # [14:11] <hsivonen> Philip: occasionally, one would hope spec writers worried about perf
  95. # [14:11] <hsivonen> Philip: consider XPath vs. Selectors
  96. # [14:12] <hsivonen> Philip: if you allow arbitrary XPath, you essentially throw away a large class of perf optimizations and make things so much harder that no one bothers to optimize in practice
  97. # [14:12] <hsivonen> Philip: for example, where's the theoretically possible streaming fail-fast Schematron validator?
  98. # [14:13] * Philip wonders why it's useful for a validator to be streaming
  99. # [14:14] <hsivonen> Philip: perf, memory usage and, most importantly, the ability to put the validator into the input pipeline of a real app so that the app internals can trust the validator failing before the app ever sees bogus input
  100. # [14:15] <hsivonen> or rather, the last point was about fail-fast
  101. # [14:15] <hsivonen> Philip: so perf and memory usage mainly
  102. # [14:17] * Quits: MikeSmith (MikeSmith@mcclure.w3.org) (Quit: Less talk, more pimp walk.)
  103. # [14:19] <Philip> Ah, that sounds reasonable
  104. # [14:27] * Joins: laplink (link@193.157.66.108)
  105. # [14:45] * Joins: MikeSmith (MikeSmith@mcclure.w3.org)
  106. # [15:09] * Quits: myakura (myakura@220.104.127.20) (Quit: Leaving...)
  107. # [15:46] * Philip is surprised to see someone in his research group using <xmp>
  108. # [16:12] * Quits: gsnedders (gsnedders@86.137.236.187) (Quit: Partying in teh intarwebs)
  109. # [16:36] * Quits: Lachy (Lachlan@213.236.208.22) (Quit: This computer has gone to sleep)
  110. # [16:38] * Joins: billmason (billmason@69.30.57.156)
  111. # [16:39] * Quits: billmason (billmason@69.30.57.156) (Quit: .)
  112. # [16:58] * Quits: Julian (chatzilla@217.91.35.233) (Quit: ChatZilla 0.9.79 [Firefox 2.0.0.11/2007112718])
  113. # [17:06] * Quits: shepazu (schepers@128.30.52.30) (Quit: shepazu)
  114. # [17:06] * Joins: tH_ (Rob@83.100.249.130)
  115. # [17:06] * tH_ is now known as tH
  116. # [17:10] * Quits: olivier (ot@128.30.52.30) (Quit: Leaving)
  117. # [17:13] * Joins: aaronlev (chatzilla@66.30.196.151)
  118. # [17:35] * Joins: matt (matt@128.30.52.30)
  119. # [17:37] * Joins: aroben (aroben@76.111.161.65)
  120. # [17:40] * Quits: heycam (cam@124.168.61.74) (Ping timeout)
  121. # [17:41] * Quits: matt (matt@128.30.52.30) (Quit: matt)
  122. # [17:43] * Quits: aroben (aroben@76.111.161.65) (Ping timeout)
  123. # [17:47] * Joins: heycam (cam@210.84.62.145)
  124. # [17:48] * Joins: aroben (aroben@76.111.161.65)
  125. # [18:10] <anne> Hixie, Philip, if you can get the HTML5 spec changed with respect to transforms and paths relatively fast I might be able to get Opera changed
  126. # [18:14] <Philip> anne: To match what Firefox and Safari do?
  127. # [18:14] * Philip wonders how non-trivial it is
  128. # [18:14] <anne> the claim was a couple of hours of work
  129. # [18:14] <Philip> (like when line widths are affected by transforms, and such)
  130. # [18:15] <anne> (if you could update your tests while you're at it, btw...)
  131. # [18:15] <MikeSmith> hsivonen - about http://hsivonen.iki.fi/doctype/ -- doesn't the XHTML spec require that a conforming XHTML document must include a doctype declaration?
  132. # [18:16] * Philip wouldn't mind writing new tests and finding out how it should be defined
  133. # [18:17] <zcorpan> MikeSmith: the xhtml5 spec doesn't :)
  134. # [18:17] <hsivonen> MikeSmith: XHTML 1.0 required "strictly conforming" docs to have a doctype: http://www.w3.org/TR/xhtml1/#strict
  135. # [18:17] <Philip> (and it does seem like a sensible change, since it makes the drawing functions more powerful, e.g. it lets you make a path with a rotated arc segment in it, which you can't do with the current spec)
  136. # [18:18] <hsivonen> MikeSmith: "strictly" and the section immediately after imply that there can be documents that aren't "strictly" conforming
  137. # [18:18] <MikeSmith> OK
  138. # [18:19] <hsivonen> MikeSmith: in any case, one can get rid of the doctype and say the document tries to be XHTML5 (even if occasionally invalid)
  139. # [18:20] <MikeSmith> hsivonen - I see. but the current state of things is still that for HTML documents, any doctype-less document will trigger quirks mode?
  140. # [18:21] <anne> hsivonen, when are you going to update the validator with the new conformance model?
  141. # [18:21] <hsivonen> anne: "in due course". hopefully later this week
  142. # [18:21] <anne> that sounds pretty fast :)
  143. # [18:23] <zcorpan> MikeSmith: yes
  144. # [18:23] <hsivonen> MikeSmith: yes. my advice to get rid of the doctype applies to XML content types--not text/html
  145. # [18:25] <MikeSmith> zcorpan, hsivonen - thanks
  146. # [18:29] <Philip> http://software.hixie.ch/utilities/js/live-dom-viewer/?%3C!DOCTYPE%20html%3E%0A%3Ccanvas%20id%3Dc%3E%3C%2Fcanvas%3E%0A%3Cscript%3E%0Awindow.onload%20%3D%20function%20()%20%7B%0A%20%20var%20ctx%20%3D%20document.getElementById('c').getContext('2d')%3B%0A%20%20ctx.lineWidth%20%3D%2010%3B%0A%20%20ctx.moveTo(50%2C%2050)%3B%0A%20%20ctx.lineTo(200%2C%2050)%3B%0A%20%20ctx.lineTo(200%2C%20100)%3B%0A%20%20ctx.rotate(Math.PI%2F3)%3B%0A%20%20ctx.scale(1%2C%204)%3B%0A%20
  147. # [18:29] <Philip> Urgh
  148. # [18:29] * Joins: Sander (svl@86.87.68.167)
  149. # [18:30] <Philip> http://tinyurl.com/2bfvog - fun
  150. # [18:31] * Philip hopes Safari matches Firefox
  151. # [18:31] * Philip hopes Firefox matches Firefox too
  152. # [18:33] <MikeSmith> hsivonen - so the practice of "you need to include a doctype if you serve a document as text/html but not of you serve it as XHTML" seems like something that's not going to be intuitive to most authors and so probably not a practice we would have much success getting many to adopt
  153. # [18:33] <Philip> (Ah, good, looks like FF3 is the same)
  154. # [18:34] <MikeSmith> hsivonen - though I understand why you're recommending it
  155. # [18:36] <zcorpan> MikeSmith: if we want to give consistent advice, then <!DOCTYPE html> gives standards mode in text/html and is harmless in XML
  156. # [18:37] * Joins: timbl (timbl@128.30.5.98)
  157. # [18:44] * Quits: aroben (aroben@76.111.161.65) (Connection reset by peer)
  158. # [18:45] * Joins: aroben (aroben@76.111.161.65)
  159. # [18:55] * Quits: aroben (aroben@76.111.161.65) (Ping timeout)
  160. # [18:57] * Quits: gavin_ (gavin@99.227.30.12) (Ping timeout)
  161. # [18:58] * Joins: aroben (aroben@76.111.161.65)
  162. # [18:59] * Joins: gavin_ (gavin@99.227.30.12)
  163. # [19:01] * Quits: aroben (aroben@76.111.161.65) (Ping timeout)
  164. # [19:08] * Joins: edas (edaspet@82.233.238.50)
  165. # [19:08] * Joins: aroben (aroben@76.111.161.65)
  166. # [19:14] * Joins: gsnedders (gsnedders@86.137.236.187)
  167. # [19:20] * Quits: aroben (aroben@76.111.161.65) (Ping timeout)
  168. # [19:26] * Joins: edaspet (edaspet@82.233.238.50)
  169. # [19:27] * Quits: edas (edaspet@82.233.238.50) (Ping timeout)
  170. # [19:30] * Joins: dbaron (dbaron@63.245.220.241)
  171. # [19:39] * Joins: aroben (aroben@76.111.161.65)
  172. # [19:40] * Joins: adele (adele@17.203.15.207)
  173. # [19:44] * Quits: aroben (aroben@76.111.161.65) (Ping timeout)
  174. # [19:49] * Joins: aroben (aroben@76.111.161.65)
  175. # [19:52] * Quits: aroben (aroben@76.111.161.65) (Ping timeout)
  176. # [19:53] * Joins: aroben (aroben@76.111.161.65)
  177. # [19:59] * Quits: aroben (aroben@76.111.161.65) (Ping timeout)
  178. # [19:59] * Joins: aroben (aroben@76.111.161.65)
  179. # [20:09] <anne> http://lists.w3.org/Archives/Public/www-html/2008Jan/0003.html
  180. # [20:13] * Dashiva wonders why we can't just say they're aliases for each other
  181. # [20:14] <Philip> Like <i> and <em>, and <b> and <strong>?
  182. # [20:15] <Dashiva> No, those actually have useful differences (even though they may be non-existent in practice)
  183. # [20:15] <anne> I don't think his points are addressed by making <abbr> and <acronym> act the same way
  184. # [20:15] <Philip> If the differences are non-existent, doesn't that mean that actually they don't have useful differences?
  185. # [20:15] <hsivonen> differences but non-existent. hmm.. :-)
  186. # [20:16] <Dashiva> Philip: In a far future, the practice may have changed :)
  187. # [20:16] <anne> spec vs impl. hmm... :)
  188. # [20:16] <Dashiva> But we're never, ever going to get agreement on the whole abbreviation nightmare
  189. # [20:17] <Philip> Perhaps the important requirement is that a conscientious author should be able to easily work out which element to use in a given situation
  190. # [20:17] <Philip> (because otherwise they'll waste time arguing about which is best, since they're too conscientious to just pick one arbitrarily)
  191. # [20:18] <Dashiva> But it's equally important that the consumer of the information agrees the element is correct
  192. # [20:21] <Philip> Consumers will treat the different elements identically, because most authors are not conscientious and will pick one arbitrarily, and consumers can't detect how hard the author thought before deciding on an element, so they can't treat the elements differently
  193. # [20:21] * Parts: zcorpan (zcorpan@88.131.66.80)
  194. # [20:22] <Dashiva> Philip: True
  195. # [20:22] <Philip> so consumers don't care, and most authors don't care, so all we can do is make life a little easier for the authors who do care
  196. # [20:23] <Philip> (e.g. by keeping them out of arguments over whether SQL is an <acronym> or an <abbr>)
  197. # [20:29] <gsnedders> It depends if it is part of MySQL or note :P
  198. # [20:30] <Philip> Or SQLite
  199. # [20:31] <hsivonen> I can imagine that way back when someone started to split hairs about acronyms and abbreviations and someone else thought it would be easier to give in and have both than to deal with the complaints
  200. # [20:31] <hsivonen> and since then countless of hours have been wasted debating correct usage
  201. # [20:31] <Dashiva> I thought it was a netscape vs IE thing
  202. # [20:32] * Philip deduced part of the pronunciation from the official parts of the SQLite site saying "an SQLite database" and not "a SQLite database"
  203. # [20:32] <gsnedders> Peh. Netscape is dead.
  204. # [20:32] <Philip> (but the non-official parts (like the bug tracker) seemed to have a roughly equal mix of 'a' vs 'an')
  205. # [20:32] <gsnedders> an SQLite database?
  206. # [20:32] <gsnedders> how does that work out?
  207. # [20:33] <Philip> "an Ess Queue Light database", I think
  208. # [20:33] <Dashiva> ess kju
  209. # [20:33] <gsnedders> Ess. If you phonetically spell it, it makes sense :)
  210. # [20:34] <Dashiva> A/an is founded on pronounciation, after all
  211. # [20:34] <gsnedders> Disclaimer: I don't make sense.
  212. # [20:34] <Philip> (See http://www.sqlite.org/faq.html#q7 etc)
  213. # [20:35] <gsnedders> Dashiva: can we just use Latin, so we need no articles?
  214. # [20:35] <Philip> Incidentally, I think "an historic" is silly, unless you pronounce it like "an 'istoric"
  215. # [20:36] <Philip> and I don't think the Queen's English includes "'istoric"
  216. # [20:36] <gsnedders> what's that?
  217. # [20:36] <gsnedders> en-gb-x-queen?
  218. # [20:36] <Dashiva> gsnedders: Seven years ago. Nowadays I'd prefer Japanese
  219. # [20:37] * gsnedders wonders when he started doing Latin
  220. # [20:37] <Philip> gsnedders: That would be a bad language code, since there'd be all sorts of backward-compatibility problems when it's updated to en-gb-x-king
  221. # [20:37] <gsnedders> only 2003.
  222. # [20:37] <Dashiva> en-gb-x-royal
  223. # [20:37] * Joins: Lachy (Lachlan@88.91.108.157)
  224. # [20:38] <gsnedders> Dashiva: but William Wales doesn't speak like that!
  225. # [20:38] <Dashiva> Then he needs to fix his compliance issues
  226. # [20:38] <gsnedders> :D
  227. # [20:39] * Dashiva is reminded of portal. "All subjects (...) must be informed that they may be informed of applicable regulatory compliance issues. No further compliance information is required (...)"
  228. # [20:39] <gsnedders> Oh, can someone remind me to NOT watch Sense and Sensibility on BBC1 when it's next on?
  229. # [20:43] * Quits: dbaron (dbaron@63.245.220.241) (Quit: 8403864 bytes have been tenured, next gc will be global.)
  230. # [20:44] <Philip> Dashiva: I thought the second sentence was spoken in a somewhat different voice, kind of like in "On the beep, the time will be <other_voice>ten minutes past five o'clock</other_voice>", because the computer was overriding the warning announcement to hide what was going to happen
  231. # [20:45] <Philip> in which case your lack of markup is losing significant semantics
  232. # [20:45] <Philip> but I may have just been misinterpreting Portal :-)
  233. # [20:48] <Dashiva> There was some interesting markup in the closed captions
  234. # [20:50] <Philip> The closed captioning worked surprisingly well
  235. # [20:50] <Philip> (I played the whole thing with no audio at first, so I had to rely on them)
  236. # [20:57] * Quits: Lachy (Lachlan@88.91.108.157) (Ping timeout)
  237. # [21:00] * Quits: edaspet (edaspet@82.233.238.50) (Quit: Quitte)
  238. # [21:04] * Joins: Lachy (Lachlan@84.215.54.100)
  239. # [21:17] * Quits: jgraham (james@81.86.215.9) (Quit: This computer has gone to sleep)
  240. # [21:22] * Joins: Thezilch (fuz007@68.111.154.116)
  241. # [21:27] * Joins: jgraham (james@81.86.215.9)
  242. # [21:33] <gsnedders> <http://www.toledoblade.com/apps/pbcs.dll/section?Category=RSS01&mime=XML> — lovely content-type
  243. # [21:34] <gsnedders> You really do need to take the last one of any non-repeatable header and not try to merge them
  244. # [21:35] <hsivonen> gsnedders: Firefox 2 seems to seems to sniff that one as a feed quite happily
  245. # [21:36] <gsnedders> hsivonen: all the major browsers just take the last Content-Type
  246. # [21:45] * Quits: ROBOd (robod@89.122.216.38) (Quit: http://www.robodesign.ro )
  247. # [21:45] <hsivonen> hmm. <applet> still has not made a comeback in HTML5
  248. # [21:46] <hsivonen> last time I checked, if you ignored validation issues, <applet> was always better than or as good as <object> for Java applets
  249. # [21:49] * hsivonen doesn't like transparent content models
  250. # [21:51] <hsivonen> Hixie: is there a reason why embedded content is still called out separately instead of just being called phrasing content?
  251. # [21:54] * Quits: adele (adele@17.203.15.207) (Quit: adele)
  252. # [21:55] * Quits: jgraham (james@81.86.215.9) (Quit: This computer has gone to sleep)
  253. # [21:55] <hsivonen> Hixie: now that what "Transparent" means has become so much simpler, it might be feasible and reader-friendly to say "prose content if the parent allows prose content and phrasing content otherwise"
  254. # [21:56] <hsivonen> Hixie: or do I misunderstand the current cases of transparency?
  255. # [21:58] * Joins: jgraham (james@81.86.215.9)
  256. # [22:00] * Quits: mjs_ (mjs@64.81.48.145) (Ping timeout)
  257. # [22:00] * Quits: aroben (aroben@76.111.161.65) (Connection reset by peer)
  258. # [22:02] * Joins: shepazu (schepers@128.30.52.30)
  259. # [22:03] * Joins: aroben (adamroben@76.111.161.65)
  260. # [22:03] * Quits: aroben (adamroben@76.111.161.65) (Client exited)
  261. # [22:04] * Joins: aroben (adamroben@76.111.161.65)
  262. # [22:10] * Joins: adele_ (adele@17.203.15.207)
  263. # [22:11] <Hixie> hsivonen: well you still have to deal with <del>
  264. # [22:12] <hsivonen> Hixie: how so?
  265. # [22:12] <Hixie> doesn't <del> have really weird rules?
  266. # [22:12] <Hixie> or did those become moot
  267. # [22:12] <hsivonen> I though those became the same as, say, <canvas> now that bimorphic is gone
  268. # [22:17] <Hixie> i'm not sure, i don't yet have it all in my head really
  269. # [22:17] <Hixie> i made the changes without optimising them
  270. # [22:17] <Hixie> i'll have to do a second optimisation pass
  271. # [22:17] <Hixie> one reason to have the term "embedded content" separately is to be able to refer to that kind of content elsewhere, e.g. in other specs
  272. # [22:18] <Hixie> we also need a better term, right now it's not obvious to people that "prose" allows "embedded content"
  273. # [22:20] <hsivonen> Hixie: indeed, if a reader merely looks up an element and sees "transparent" or "embedded content", the practical meaning in terms of prose and phrasing is non-obvious
  274. # [22:20] <hsivonen> so the spec isn't quite good for random access right now
  275. # [22:23] <Hixie> yeah
  276. # [22:23] <Hixie> anne: do you have a timescale for when you need that canvas change by?
  277. # [22:24] <Hixie> i do like having the "transparent" keyword
  278. # [22:24] <Hixie> it makes my life a lot easier
  279. # [22:24] <Hixie> but there are probably changes that can be made
  280. # [22:25] <hsivonen> Hixie: using "transparent" while you are editing may be a good optimization, but eventually, the spec will be read more than written, so once it is stable, it may well make sense to expand what it means
  281. # [22:26] <Hixie> sure
  282. # [22:28] <Hixie> bbiab
  283. # [22:47] <anne> Hixie, one/two weeks?
  284. # [22:51] <anne> http://www.w3.org/MarkUp/2008/WD-xhtml-access-20080107/ ...
  285. # [22:51] * Quits: aaronlev (chatzilla@66.30.196.151) (Ping timeout)
  286. # [22:53] <anne> the conformance requirements are pretty badly done...
  287. # [23:14] * Quits: gsnedders (gsnedders@86.137.236.187) (Quit: Partying in teh intarwebs)
  288. # [23:20] * Joins: sbuluf (yntg@200.49.132.101)
  289. # [23:29] * Joins: dbaron (dbaron@63.245.220.241)
  290. # [23:29] <Hixie> anne: k
  291. # [23:31] <anne> i mean, the sooner the better, I understand this is one of the last major compat problems and there's a fair chance we might fix it on time if the spec is accurate
  292. # [23:32] * Joins: edas (edaspet@82.233.238.50)
  293. # [23:33] <Hixie> right
  294. # [23:40] <jgraham> What problems is XHTML access supposed to solve? It doesn't seem clear to me what the <access> element design is supposed to achieve.
  295. # [23:40] * Quits: Lachy (Lachlan@84.215.54.100) (Quit: Leaving)
  296. # [23:41] * Joins: Lachy (Lachlan@84.215.54.100)
  297. # [23:52] * Quits: timbl (timbl@128.30.5.98) (Quit: timbl)
  298. # [23:53] * Joins: mjs (mjs@17.203.15.217)
  299. # Session Close: Tue Jan 08 00:00:00 2008

The end :)