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

Options:

  1. # Session Start: Tue May 05 00:00:00 2009
  2. # Session Ident: #whatwg
  3. # [00:02] * Quits: svl (n=me@ip565744a7.direct-adsl.nl) ("And back he spurred like a madman, shrieking a curse to the sky.")
  4. # [00:11] * Joins: nessy (n=nessy@115.129.12.79)
  5. # [00:17] * Quits: weinig_ (n=weinig@17.246.16.148)
  6. # [00:18] * Joins: doublec (n=doublec@118-93-168-138.dsl.dyn.ihug.co.nz)
  7. # [00:26] * Joins: weinig (n=weinig@17.246.16.148)
  8. # [00:44] * Quits: nessy (n=nessy@115.129.12.79) (Read error: 110 (Connection timed out))
  9. # [00:47] * aroben is now known as aroben|away
  10. # [00:47] <annevk5> Hixie, the attribute having the wrong name in the spec does seem like a problem
  11. # [00:48] <Hixie> yeah your e-mail is in the bucket
  12. # [00:48] <Hixie> i'm assuming there's something more wrong with it
  13. # [00:48] * Quits: virtuelv (n=virtuelv@084202133045.customer.alfanett.no) (Read error: 104 (Connection reset by peer))
  14. # [00:48] * Joins: weinig_ (n=weinig@nat/apple/x-26df8753156d4f30)
  15. # [00:49] * Parts: billmason (n=billmaso@ip28.unival.com)
  16. # [00:49] * Quits: cgriego (n=cgriego@out-02.hotels.com)
  17. # [00:54] * Quits: jwalden (n=waldo@63.245.220.241) ("->S")
  18. # [00:55] * Quits: dolske (n=dolske@firefox/developer/dolske)
  19. # [00:56] * Quits: weinig_ (n=weinig@nat/apple/x-26df8753156d4f30) (Read error: 104 (Connection reset by peer))
  20. # [00:57] * Quits: weinig (n=weinig@17.246.16.148) (Read error: 145 (Connection timed out))
  21. # [01:00] * Joins: weinig (n=weinig@nat/apple/x-571872bc49b3f22b)
  22. # [01:02] * Joins: dbaron (n=dbaron@corp-241.mountainview.mozilla.com)
  23. # [01:03] * aroben|away is now known as aroben
  24. # [01:13] * Joins: jwalden (n=waldo@corp-241.mountainview.mozilla.com)
  25. # [01:13] * Joins: dolske (n=dolske@corp-241.mountainview.mozilla.com)
  26. # [01:14] * Quits: taf2_ (n=taf2@38.99.201.242)
  27. # [01:16] * Joins: weinig_ (n=weinig@17.246.16.148)
  28. # [01:22] * Quits: karlcow (n=karl@nerval.la-grange.net) ("This computer has gone to sleep")
  29. # [01:24] * Quits: weinig (n=weinig@nat/apple/x-571872bc49b3f22b) (Read error: 145 (Connection timed out))
  30. # [01:27] * Quits: heycam (n=cam@203-217-72-53.dyn.iinet.net.au) ("bye")
  31. # [01:33] * Joins: olliej (n=oliver@17.246.16.102)
  32. # [01:37] * Joins: davidb (n=davidb@bas4-toronto06-1279310258.dsl.bell.ca)
  33. # [01:37] <hsivonen> Philip`: no, hadn't seen it. filed http://bugzilla.validator.nu/show_bug.cgi?id=503
  34. # [01:38] <hsivonen> Hixie: Re: \0 test, probably specs aren't ambiguous. the selector behavior vs. DOM Inspector is just weird
  35. # [01:39] <Hixie> ah
  36. # [01:40] <Hixie> certain aspects of that test looked wrong given html5's parsing rules fwiw
  37. # [01:40] * Joins: ojan (n=ojan@203.39.247.241)
  38. # [01:41] * Quits: akamike (n=mikerobi@92.3.35.15)
  39. # [01:42] <Hixie> " student is making a timeline of important events in Apple's history. As he reads Wikipedia entries on the topic, he clicks on dates and selects "add to timeline", which causes an entry to be added to his timeline"
  40. # [01:43] <Hixie> assuming that the timeline is implemented by a third-party app or web site
  41. # [01:43] <Hixie> and assuming that the browser doesn't have built-in knowledge of this app or site
  42. # [01:43] <Hixie> and assuming wikipedia doesn't either
  43. # [01:43] <hsivonen> Hixie: which parts of the test case looked wrong?
  44. # [01:43] <hsivonen> the test case is failing with the HTML5 parser
  45. # [01:44] * weinig_ is now known as weinig
  46. # [01:44] <Hixie> hsivonen: &#0; and ^@ in HTML5 become U+FFFD, not U+0000
  47. # [01:44] <Hixie> i guess if wikipedia can output an .ics file somehow, the third party site or app could register as a handler for text/calendar files...
  48. # [01:44] <hsivonen> Hixie: isn't that the point of the test?
  49. # [01:45] <hsivonen> Hixie: that it's a bug if any of the selectors matches
  50. # [01:45] <Hixie> hsivonen: oh i didn't look to see if the test expected the selectors to match, my bad
  51. # [01:45] <Hixie> which parts of the test case are failing?
  52. # [01:47] <hsivonen> Hixie: http://hsivonen.iki.fi/screen/228856-2.png
  53. # [01:47] <Hixie> test case url?
  54. # [01:48] <hsivonen> http://mxr.mozilla.org/mozilla-central/source/layout/reftests/bugs/228856-2.html
  55. # [01:51] <Hixie> is css also converting things to FFFD?
  56. # [01:51] <Hixie> but only at the start of a selector?
  57. # [01:51] <hsivonen> Hixie: I have no idea
  58. # [01:52] <Hixie> hah, there's a regexp search replace error in the text part of the test (doesn't affect results)
  59. # [01:52] <Hixie> &amp;. should be &amp;#
  60. # [01:53] * Quits: doublec (n=doublec@118-93-168-138.dsl.dyn.ihug.co.nz) (Read error: 113 (No route to host))
  61. # [01:54] <Hixie> hsivonen: i bet that the css parser is turning ^@ into U+FFFD
  62. # [01:54] <Hixie> hsivonen: (but not \0)
  63. # [02:11] * Joins: taf2_ (n=taf2@c-68-49-245-59.hsd1.dc.comcast.net)
  64. # [02:11] * Quits: onar_ (n=onar@17.244.69.252)
  65. # [02:31] * Joins: heycam (n=cam@130.194.73.110)
  66. # [02:31] * Quits: davidb (n=davidb@bas4-toronto06-1279310258.dsl.bell.ca)
  67. # [02:45] * Quits: roc (n=roc@guest-226.mountainview.mozilla.com)
  68. # [02:47] * Joins: slightlyoff (n=slightly@nat/google/x-9203959946faaf84)
  69. # [02:55] * Joins: onar_ (n=onar@c-98-234-65-251.hsd1.ca.comcast.net)
  70. # [02:56] * Quits: Amorphous (i=jan@unaffiliated/amorphous) (Read error: 110 (Connection timed out))
  71. # [02:58] * Joins: Amorphous (i=jan@unaffiliated/amorphous)
  72. # [03:09] * Joins: LHSilli (n=s@cm-84.208.110.159.getinternet.no)
  73. # [03:10] * Parts: LHSilli (n=s@cm-84.208.110.159.getinternet.no)
  74. # [03:20] * Quits: slightlyoff (n=slightly@nat/google/x-9203959946faaf84) (Read error: 110 (Connection timed out))
  75. # [03:20] * Joins: dave_levin (n=dave_lev@98.203.247.78)
  76. # [03:22] * Joins: nessy (n=nessy@115.129.26.147)
  77. # [03:23] * Joins: dave_levin_ (n=dave_lev@72.14.224.1)
  78. # [03:24] * Quits: onar_ (n=onar@c-98-234-65-251.hsd1.ca.comcast.net)
  79. # [03:32] * Quits: dave_levin (n=dave_lev@98.203.247.78) (Read error: 145 (Connection timed out))
  80. # [03:33] * Quits: heycam (n=cam@130.194.73.110) ("bye")
  81. # [03:43] * Joins: tantek (n=tantek@adsl-63-195-114-133.dsl.snfc21.pacbell.net)
  82. # [03:48] * Quits: nessy (n=nessy@115.129.26.147) (Read error: 110 (Connection timed out))
  83. # [03:53] * Quits: taf2_ (n=taf2@c-68-49-245-59.hsd1.dc.comcast.net)
  84. # [03:56] * Joins: onar_ (n=onar@c-98-234-65-251.hsd1.ca.comcast.net)
  85. # [04:14] * Quits: weinig (n=weinig@17.246.16.148)
  86. # [04:14] * Quits: annevk5 (n=annevk@53568A94.cable.casema.nl)
  87. # [04:16] * othree_ is now known as othree
  88. # [04:16] * Joins: aboodman4 (n=aboodman@67.218.102.211)
  89. # [04:22] * Quits: tantek (n=tantek@adsl-63-195-114-133.dsl.snfc21.pacbell.net)
  90. # [04:24] * Joins: doublec (n=doublec@118-93-168-138.dsl.dyn.ihug.co.nz)
  91. # [04:34] * Quits: aboodman4 (n=aboodman@67.218.102.211) (Read error: 60 (Operation timed out))
  92. # [04:39] * Quits: doublec (n=doublec@118-93-168-138.dsl.dyn.ihug.co.nz) ("Leaving")
  93. # [04:43] * aroben is now known as aroben|away
  94. # [04:45] * Joins: weinig (n=weinig@c-67-180-35-124.hsd1.ca.comcast.net)
  95. # [04:49] * Joins: aboodman4 (n=aboodman@64.81.72.99)
  96. # [04:49] * Quits: aboodman4 (n=aboodman@64.81.72.99) (Client Quit)
  97. # [04:52] * Joins: tantek (n=tantek@c-67-180-202-79.hsd1.ca.comcast.net)
  98. # [04:58] * Joins: doublec (n=doublec@118-93-168-138.dsl.dyn.ihug.co.nz)
  99. # [05:02] * Joins: doublec_ (n=doublec@118-93-168-138.dsl.dyn.ihug.co.nz)
  100. # [05:03] * Quits: doublec (n=doublec@118-93-168-138.dsl.dyn.ihug.co.nz) (Read error: 113 (No route to host))
  101. # [05:03] * doublec_ is now known as double
  102. # [05:04] * Quits: double (n=doublec@118-93-168-138.dsl.dyn.ihug.co.nz) (Client Quit)
  103. # [05:05] * Quits: tantek (n=tantek@c-67-180-202-79.hsd1.ca.comcast.net)
  104. # [05:08] * Joins: roc (n=roc@65.123.0.210)
  105. # [05:09] * Joins: doublec (n=doublec@118-93-168-138.dsl.dyn.ihug.co.nz)
  106. # [05:10] * Quits: dolske (n=dolske@firefox/developer/dolske)
  107. # [05:14] * Quits: dbaron (n=dbaron@corp-241.mountainview.mozilla.com) ("8403864 bytes have been tenured, next gc will be global.")
  108. # [05:15] * Quits: roc (n=roc@65.123.0.210)
  109. # [05:19] * Joins: karlcow (n=karl@128.30.54.58)
  110. # [05:21] * Quits: aroben|away (n=aroben@unaffiliated/aroben) (Read error: 60 (Operation timed out))
  111. # [05:31] * Joins: aroben|away (n=aroben@unaffiliated/aroben)
  112. # [05:35] * Joins: hdh (n=hdh@58.187.19.111)
  113. # [05:39] * Joins: dolske (n=dolske@c-76-103-40-203.hsd1.ca.comcast.net)
  114. # [05:47] * Joins: archtech (n=stanv@83.228.56.37)
  115. # [05:55] * Joins: heycam (n=cam@130.194.73.110)
  116. # [06:19] * weinig is now known as weinig|store
  117. # [06:20] * weinig|store is now known as weinig
  118. # [06:21] * Joins: nessy (n=nessy@115.129.27.31)
  119. # [06:24] * weinig is now known as weinig|store
  120. # [06:31] * Quits: dave_levin_ (n=dave_lev@72.14.224.1)
  121. # [06:35] * Quits: karlcow (n=karl@128.30.54.58) ("This computer has gone to sleep")
  122. # [06:39] * Joins: roc (n=roc@65.123.0.210)
  123. # [06:40] * aroben|away is now known as aroben
  124. # [06:43] * Joins: weinig (n=weinig@c-67-180-35-124.hsd1.ca.comcast.net)
  125. # [06:47] * Quits: nessy (n=nessy@115.129.27.31) ("This computer has gone to sleep")
  126. # [06:58] * Joins: dglazkov (n=dglazkov@98.207.88.44)
  127. # [06:59] * Quits: weinig|store (n=weinig@c-67-180-35-124.hsd1.ca.comcast.net) (Read error: 110 (Connection timed out))
  128. # [06:59] * Quits: dglazkov (n=dglazkov@98.207.88.44) (Client Quit)
  129. # [07:10] * Joins: karlcow (n=karl@128.30.54.58)
  130. # [07:15] * Quits: roc (n=roc@65.123.0.210)
  131. # [07:21] * Joins: zcorpan (n=zcorpan@c83-252-196-43.bredband.comhem.se)
  132. # [07:27] * Quits: zcorpan (n=zcorpan@c83-252-196-43.bredband.comhem.se) (Read error: 60 (Operation timed out))
  133. # [07:34] * Quits: doublec (n=doublec@118-93-168-138.dsl.dyn.ihug.co.nz) ("Leaving")
  134. # [07:35] * Quits: archtech (n=stanv@83.228.56.37)
  135. # [07:37] * Joins: zcorpan__ (n=zcorpan@c83-252-196-43.bredband.comhem.se)
  136. # [07:38] <zcorpan__> annevk42: hey, the php stopped working on simon.html5.org - i guess it's the same issue as you had? how did you fix it?
  137. # [07:38] * zcorpan__ is now known as zcorpan
  138. # [07:41] * Joins: harig (n=opera@59.90.71.35)
  139. # [07:45] <zcorpan> apparently i had to change AddHandler application/x-httpd-php .php to AddHandler php-cgi .php
  140. # [07:48] <zcorpan> although now indexing is broken instead
  141. # [07:49] <zcorpan> oh well
  142. # [07:50] <Hixie> christ microformats are complicated if you start looking at all value-excerption, value-class, etc
  143. # [07:50] * Hixie doesn't understand at all how you parse an hcard
  144. # [07:52] <takkaria> with logic
  145. # [07:52] * Joins: myakura (n=myakura@p1120-ipbf2303marunouchi.tokyo.ocn.ne.jp)
  146. # [07:52] <takkaria> and a good grasp of the essential unsatisfactoryness of existence
  147. # [07:59] <zcorpan> when i wrote http://simon.html5.org/sandbox/html/w3c-home-in-html5 back in 2007, it only rendered correctly in opera. now it renders correctly in safari and firefox, too
  148. # [08:02] <zcorpan> i wonder when it will render correctly in ie
  149. # [08:07] * Joins: pesla (n=retep@procurios.xs4all.nl)
  150. # [08:09] * Joins: archtech (n=stanv@83.228.56.37)
  151. # [08:12] * Joins: maikmerten (n=merten@ls5dhcp196.cs.uni-dortmund.de)
  152. # [08:13] * Quits: maikmerten (n=merten@ls5dhcp196.cs.uni-dortmund.de) (Remote closed the connection)
  153. # [08:14] * Joins: maikmerten (n=merten@ls5dhcp196.cs.uni-dortmund.de)
  154. # [08:15] * Quits: jwalden (n=waldo@corp-241.mountainview.mozilla.com) ("ChatZilla 0.9.82.1-rdmsoft [XULRunner 1.8.0.9/2006120508]")
  155. # [08:21] * Quits: aroben (n=aroben@unaffiliated/aroben) (Read error: 104 (Connection reset by peer))
  156. # [08:31] * Joins: tantek (n=tantek@adsl-63-195-114-133.dsl.snfc21.pacbell.net)
  157. # [08:34] * Quits: heycam (n=cam@130.194.73.110) ("bye")
  158. # [08:35] * Joins: zalan (n=kvirc@dsl54009DA0.pool.t-online.hu)
  159. # [08:40] * Joins: zdobersek (n=zan@cpe-92-37-68-224.dynamic.amis.net)
  160. # [08:41] * Joins: Maurice (n=ano@a80-101-46-164.adsl.xs4all.nl)
  161. # [08:50] <Hixie> man vcard is much more complex than it looks
  162. # [08:53] * Joins: ap (n=ap@194.154.88.47)
  163. # [09:01] * Joins: pesla\work (n=retep@procurios.xs4all.nl)
  164. # [09:04] * Hixie sees more e-mails with people suggesting i'm a microformats shill
  165. # [09:05] * Joins: heycam (n=cam@203-217-72-53.dyn.iinet.net.au)
  166. # [09:05] <Hixie> is it a good sign that the microformats people think i'm an rdfa shill while the rdfa people think i'm a microformats shill?
  167. # [09:08] * Quits: pesla (n=retep@procurios.xs4all.nl) (Read error: 110 (Connection timed out))
  168. # [09:12] * Quits: onar_ (n=onar@c-98-234-65-251.hsd1.ca.comcast.net)
  169. # [09:18] * Quits: olliej (n=oliver@17.246.16.102) (Remote closed the connection)
  170. # [09:18] * Joins: olliej (n=oliver@17.246.16.102)
  171. # [09:30] * Quits: tantek (n=tantek@adsl-63-195-114-133.dsl.snfc21.pacbell.net)
  172. # [09:31] * Joins: sid0_ (n=sid0@unaffiliated/sid0)
  173. # [09:32] * Joins: tantek (n=tantek@adsl-63-195-114-133.dsl.snfc21.pacbell.net)
  174. # [09:36] <takkaria> Hixie: I'd guess so
  175. # [09:36] * Joins: philipj (n=philipj@pat.se.opera.com)
  176. # [09:41] * Quits: zalan (n=kvirc@dsl54009DA0.pool.t-online.hu) ("KVIrc 3.4.0 Virgo http://www.kvirc.net/")
  177. # [09:41] <Philip`> Hixie: I don't think it's good for everybody to think you're a shill
  178. # [09:42] <Hixie> not everybody thinks i'm a shill, only the people who have vested interests in the current topic of discussion
  179. # [09:43] <Philip`> Ah, so everybody who's not apathetic thinks you're a shill? :-)
  180. # [09:46] <Hixie> no, there are people who are interested in a solution to microdata annotations who don't yet have a horse in the race
  181. # [09:46] * Hixie keeps finding bugs in the hcard spec
  182. # [09:49] * Quits: sid0 (n=sid0@unaffiliated/sid0) (Read error: 110 (Connection timed out))
  183. # [09:50] <Hixie> oh wait, this is actually not a bug in hcard, it's crazyness in vcard
  184. # [09:50] <Hixie> nevermind!
  185. # [09:51] * Quits: myakura (n=myakura@p1120-ipbf2303marunouchi.tokyo.ocn.ne.jp) ("Leaving...")
  186. # [10:02] * Quits: sid0_ (n=sid0@unaffiliated/sid0) ("No Ping reply in 90 seconds.")
  187. # [10:03] * Joins: sid0 (n=sid0@unaffiliated/sid0)
  188. # [10:09] * Quits: zdobersek (n=zan@cpe-92-37-68-224.dynamic.amis.net) ("Leaving.")
  189. # [10:10] * Joins: svl (n=chatzill@a194-109-2-36.dmn.xs4all.nl)
  190. # [10:19] * Joins: nessy (n=nessy@203.158.60.217)
  191. # [10:19] * Joins: pauld (n=pauld@host81-158-125-194.range81-158.btcentralplus.com)
  192. # [10:24] * Quits: ojan (n=ojan@203.39.247.241)
  193. # [10:31] * Joins: slightlyoff (n=slightly@204.14.154.244)
  194. # [10:31] * Quits: slightlyoff (n=slightly@204.14.154.244) (Remote closed the connection)
  195. # [10:41] * Quits: tantek (n=tantek@adsl-63-195-114-133.dsl.snfc21.pacbell.net)
  196. # [10:49] * Joins: annevk5 (n=annevk@53568A94.cable.casema.nl)
  197. # [10:52] * Quits: archtech (n=stanv@83.228.56.37) (No route to host)
  198. # [11:12] <Hixie> vcard is awesome
  199. # [11:12] <Hixie> some of its fields have two states, URI or inline binary
  200. # [11:12] <Hixie> but all of its fields have both a type and an encoding
  201. # [11:13] <Hixie> URI is uri type, text encoding
  202. # [11:13] <Hixie> inline binary in binary type, b encoding
  203. # [11:14] * Joins: ROBOd (n=robod@89.122.216.38)
  204. # [11:14] <Hixie> the default? binary type, text encoding
  205. # [11:14] <Hixie> so if you specify a uri, you must set the type to uri
  206. # [11:14] <Hixie> but if you inline the binary data, you must set the encoding to b
  207. # [11:17] * Quits: maikmerten (n=merten@ls5dhcp196.cs.uni-dortmund.de) (Remote closed the connection)
  208. # [11:20] <Philip`> They should solve the problem by making everything always be a URI
  209. # [11:20] * Joins: archtech (n=stanv@83.228.56.37)
  210. # [11:20] <Philip`> and then they can use data: URIs if they want inline binary data
  211. # [11:21] * jgraham discovers that writing if (condition); {do something} rather than if (condition) {do something) has rather surprising consequences
  212. # [11:21] <Philip`> jgraham: Is it surprisingly that randomly inserting incorrect characters makes code not work? You'd get the same problem if you wrote if (!condition) {do something} :-p
  213. # [11:22] <Philip`> *surprising
  214. # [11:22] <jgraham> Philip`: It is somewhat surprising if it is a character that is often optional at the end of a line
  215. # [11:22] * Joins: maikmerten (n=merten@ls5dhcp196.cs.uni-dortmund.de)
  216. # [11:23] <jgraham> Or rather it is unsurprising but not as easy to spot
  217. # [11:24] <Philip`> Oh, I suppose it's more surprising in a crazy language like JS where correct syntax is optional
  218. # [11:27] * Parts: zcorpan (n=zcorpan@c83-252-196-43.bredband.comhem.se)
  219. # [11:29] * Joins: zcorpan (n=zcorpan@c83-252-196-43.bredband.comhem.se)
  220. # [11:29] * Hixie is starting to get a headache from the vcard spec
  221. # [11:33] * annevk42 had that when he tried to figure out how browsing contexts worked for the first time
  222. # [11:34] <Hixie> heh
  223. # [11:35] <Hixie> the first time those were specced, they were screwed up
  224. # [11:35] <annevk42> :p
  225. # [11:51] * Joins: mat_t (n=mattomas@nat/canonical/x-16dd2055c766c7cd)
  226. # [11:55] * Quits: Lachy (n=Lachlan@85.196.122.246) ("This computer has gone to sleep")
  227. # [12:02] * Quits: hdh (n=hdh@58.187.19.111) (Remote closed the connection)
  228. # [12:03] * Joins: hdh (n=hdh@58.187.19.111)
  229. # [12:09] * Joins: Lachy (n=Lachlan@pat-tdc.opera.com)
  230. # [12:57] * Joins: myakura (n=myakura@p1120-ipbf2303marunouchi.tokyo.ocn.ne.jp)
  231. # [12:59] * Quits: zcorpan (n=zcorpan@c83-252-196-43.bredband.comhem.se) (Remote closed the connection)
  232. # [12:59] * Joins: zcorpan (n=zcorpan@c83-252-196-43.bredband.comhem.se)
  233. # [13:06] * Joins: zalan (n=kvirc@catv-89-132-200-147.catv.broadband.hu)
  234. # [13:13] * Joins: VeXocide (i=vexocide@snail.stack.nl)
  235. # [13:17] * Joins: danbri (n=danbri@s5590d015.adsl.wanadoo.nl)
  236. # [13:23] * Quits: zcorpan (n=zcorpan@c83-252-196-43.bredband.comhem.se)
  237. # [13:47] <olliej> who the heck is this giovanni guy who has randomly appeared?
  238. # [13:47] <annevk42> I wonder about that too sometimes, but decided not to care too much :)
  239. # [13:48] * Joins: taf2_ (n=taf2@38.99.201.242)
  240. # [13:48] <olliej> annevk42: hee
  241. # [13:49] <annevk42> yo!
  242. # [13:49] <annevk42> Safari for Ubuntu coming soon? :)
  243. # [13:49] <Philip`> He appeared about five months ago
  244. # [13:53] * Quits: maikmerten (n=merten@ls5dhcp196.cs.uni-dortmund.de) (Remote closed the connection)
  245. # [13:53] <jgraham> annevk42: With Chromium for Ubuntu, why would anyone still care about Safari ;)
  246. # [13:53] * Joins: maikmerten (n=merten@ls5dhcp196.cs.uni-dortmund.de)
  247. # [13:59] <annevk42> jgraham, presumably Safari for Ubuntu would have the Mac font settings and such
  248. # [14:00] <annevk42> and my current Chromium crashes on paste
  249. # [14:00] <annevk42> not fun
  250. # [14:00] * annevk42 uses Midori instead
  251. # [14:00] <Philip`> Just run Windows Safari in Wine
  252. # [14:02] <olliej> jgraham: because without safari how would chrome actually render anything :p
  253. # [14:03] <Philip`> It could still render the pipes screensaver
  254. # [14:04] <jgraham> olliej: Aren't you obliged to continually point out the different between Safari and WebKit?
  255. # [14:04] <olliej> jgraham: i gave that up when google started acting like they wrote their own browser
  256. # [14:04] <olliej> :D
  257. # [14:05] * jgraham likes the Chromium start up tab "This browser is not ready yet! This is a pre-alpha build of Chromium on Linux. It is woefully incomplete. It's ‘Chromium’, not ‘Google Chrome’"
  258. # [14:05] <jgraham> And then later
  259. # [14:05] * Quits: maikmerten (n=merten@ls5dhcp196.cs.uni-dortmund.de) (Remote closed the connection)
  260. # [14:05] <jgraham> Please don't file bugs, Blogging about it is not helpful
  261. # [14:06] * Joins: maikmerten (n=merten@ls5dhcp196.cs.uni-dortmund.de)
  262. # [14:13] * Joins: pmuellr (n=pmuellr@nat/ibm/x-b4988449c31dca5b)
  263. # [14:14] * Joins: zcorpan (n=zcorpan@pat.se.opera.com)
  264. # [14:21] * Quits: zcorpan (n=zcorpan@pat.se.opera.com) (Read error: 104 (Connection reset by peer))
  265. # [14:21] * Joins: zcorpan (n=zcorpan@pat.se.opera.com)
  266. # [14:22] <Philip`> http://en.wikipedia.org/wiki/File:BGP_FSM.svg - "W3C ✓ The source code of this SVG is valid" - "Errors found while checking this document as SVG 1.1 + IRI! Result: 16 Errors, 4 warning(s)"
  267. # [14:22] <Philip`> Validation badges are a bit useless when people use them for invalid files
  268. # [14:24] * Quits: archtech (n=stanv@83.228.56.37)
  269. # [14:24] * Joins: archtech (n=stanv@83.228.56.37)
  270. # [14:25] <jgraham> s/for invalid files//
  271. # [14:26] <Philip`> Hmm, I meant something more like "Validation badges are a bit useless given that people use them for invalid files"
  272. # [14:26] <Philip`> If they were a guarantee of validity, they'd be reasonably useful, because they'd save you the effort of running the validator when you want to see that a page is valid
  273. # [14:28] <jgraham> Philip`: Since you can't rely on people to do that, I maintain that my restatement is equivalent to the original
  274. # [14:30] <Philip`> jgraham: I assumed you were reading my statement as more like "Validation badges a bit useless in the specific cases in which people use them for invalid files", and your restatement was not equivalent to, and more correct than, the original
  275. # [14:32] <Philip`> (so I was clarifying that my statement was indeed intended to be equivalent to your restatement)
  276. # [14:32] <Philip`> (so, um, I think we agree?)
  277. # [14:36] <zcorpan> Hixie: should events that don't bubble still be dispatched on window when fired on document?
  278. # [14:39] * Quits: mat_t (n=mattomas@nat/canonical/x-16dd2055c766c7cd) ("This computer has gone to sleep")
  279. # [14:42] * Quits: annevk5 (n=annevk@53568A94.cable.casema.nl) (Read error: 110 (Connection timed out))
  280. # [14:42] * Quits: annevk42 (n=annevk@53568A94.cable.casema.nl) (Read error: 110 (Connection timed out))
  281. # [14:43] * Joins: mat_t (n=mattomas@nat/canonical/x-51952978b8a5c569)
  282. # [14:45] * zcorpan wonders whether the query string has been considered for the svg <param> discussion
  283. # [15:01] <heycam> zcorpan, it does (for the exposing of parameters to the parameter referencing mechanism)
  284. # [15:01] <heycam> but i see what you mean: maybe we don't need a <param> element; we can just use the query string when referencing documents
  285. # [15:02] <heycam> otoh, perhaps other referenceable things (applets?) would only work with <param> elements
  286. # [15:02] <heycam> on my third hand, maybe they wouldn't work without modification with <svg:param>s anyway
  287. # [15:03] * Joins: annevk42 (n=annevk@5ED2D22C.cable.ziggo.nl)
  288. # [15:16] * Quits: annevk42 (n=annevk@5ED2D22C.cable.ziggo.nl) (Remote closed the connection)
  289. # [15:27] * Joins: annevk42 (n=annevk@5ED2D22C.cable.ziggo.nl)
  290. # [15:29] * Quits: pauld (n=pauld@host81-158-125-194.range81-158.btcentralplus.com)
  291. # [15:34] * Joins: davidb (n=davidb@bas4-toronto06-1279277705.dsl.bell.ca)
  292. # [15:35] <zcorpan> heycam: you have three hands?
  293. # [15:45] <Philip`> I think the hands may be largely metaphorical
  294. # [15:46] * Quits: mpt (n=mpt@canonical/launchpad/mpt) (Read error: 113 (No route to host))
  295. # [15:49] * Philip` reads a book on network routing design, and finds a bit saying "This design looks good on paper until you try to implement it. It's very difficult to justify all the AS numbers from American Registry for Internet Numbers (ARIN) or other Regional Internet Registries (RIRs) required for such a network. AS numbers are a finite resource that is quickly being depleted."
  296. # [15:49] <Philip`> It's both funny and sad that arbitrary integers are a scarce resource
  297. # [15:49] * jgraham decides to pick up a few spare when he is next out
  298. # [15:50] <Philip`> (AS numbers are kind of like IPv4 addresses, except there's 2^16 times fewer of them)
  299. # [15:52] <Philip`> (and just like NAT is a 'solution' to IPv4 scarcity, the book suggests using private AS numbers and then doing some ugly tricks to hide them and look like a single globally-unique AS number to the rest of the world)
  300. # [15:54] <Philip`> (and just like IPv6 is a better solution, there's a new extension for 4-byte AS numbers (which is more than anybody will ever need), with approximately zero adoption so far)
  301. # [15:56] <Philip`> (I suppose the lesson is that when you design a protocol with a globally-unique bounded number, decide how many bits you'll need if the protocol is wildly successfully and used by the entirety of human civilisation throughout our galaxy, and then double it)
  302. # [15:56] <jgraham> The other solution might be to not design protocols that rely on globally unique bounded numbers (but that may be impractical)
  303. # [15:58] <Philip`> Good luck convincing Cisco to route packets based on unbounded integers
  304. # [15:59] <Philip`> You could build a secretive file-sharing protocol which transfers megabytes of data in the source address of each packet
  305. # [16:01] * Quits: zcorpan (n=zcorpan@pat.se.opera.com)
  306. # [16:03] <Philip`> (...And it's kind of hard to avoid the "globally unique" part, since people really like having identifiers with a unique meaning)
  307. # [16:06] * Joins: mpt (n=mpt@canonical/launchpad/mpt)
  308. # [16:11] * Joins: sverrej (n=sverrej@59.90.71.35)
  309. # [16:22] * Quits: nessy (n=nessy@203.158.60.217) ("This computer has gone to sleep")
  310. # [16:23] * Joins: aroben (n=aroben@unaffiliated/aroben)
  311. # [16:24] * Quits: drostie (n=hopkins@5354256F.cable.casema.nl) (Read error: 104 (Connection reset by peer))
  312. # [16:26] * Joins: drostie (n=hopkins@5354256F.cable.casema.nl)
  313. # [16:30] * Joins: ZombieLoffe (n=e@unaffiliated/zombieloffe)
  314. # [16:35] * Quits: Simetrical (n=Simetric@wikipedia/simetrical) (Remote closed the connection)
  315. # [16:43] * Joins: tantek (n=tantek@adsl-63-195-114-133.dsl.snfc21.pacbell.net)
  316. # [16:46] * Joins: virtuelv (n=virtuelv@213.236.208.247)
  317. # [16:48] * Quits: sverrej (n=sverrej@59.90.71.35) (Read error: 145 (Connection timed out))
  318. # [16:50] * Joins: mstange (n=markus@buntes215.wohnheim.uni-kl.de)
  319. # [16:51] * Joins: Simetrical (n=Simetric@wikipedia/simetrical)
  320. # [16:53] * Joins: billmason (n=billmaso@69.30.57.28)
  321. # [16:56] * Quits: Lachy (n=Lachlan@pat-tdc.opera.com) ("This computer has gone to sleep")
  322. # [16:56] * Joins: pauld (n=pauld@host81-158-125-194.range81-158.btcentralplus.com)
  323. # [17:00] * Quits: maikmerten (n=merten@ls5dhcp196.cs.uni-dortmund.de) (Remote closed the connection)
  324. # [17:04] * Quits: Maurice (n=ano@a80-101-46-164.adsl.xs4all.nl) ("Disconnected...")
  325. # [17:04] * Quits: Simetrical (n=Simetric@wikipedia/simetrical) (Remote closed the connection)
  326. # [17:05] * Joins: Simetrical (n=Simetric@wikipedia/simetrical)
  327. # [17:06] * Joins: roc (n=roc@65.123.0.210)
  328. # [17:06] * Quits: pauld (n=pauld@host81-158-125-194.range81-158.btcentralplus.com)
  329. # [17:08] * Quits: annevk42 (n=annevk@5ED2D22C.cable.ziggo.nl) (Remote closed the connection)
  330. # [17:09] * Quits: virtuelv (n=virtuelv@213.236.208.247) (Read error: 113 (No route to host))
  331. # [17:30] * Joins: Lachy (n=Lachlan@85.196.122.246)
  332. # [17:32] * Joins: cgriego (n=cgriego@12.14.172.51)
  333. # [17:34] * Joins: sid0_ (n=sid0@unaffiliated/sid0)
  334. # [17:38] * Quits: mstange (n=markus@buntes215.wohnheim.uni-kl.de) ("ChatZilla 0.9.84-2009030221 [Firefox 3.6a1pre/20090426031353]")
  335. # [17:40] * Joins: bgalbraith (n=bgalbrai@63.245.220.241)
  336. # [17:42] * Joins: sid0__ (n=sid0@unaffiliated/sid0)
  337. # [17:43] * Quits: sid0 (n=sid0@unaffiliated/sid0) (Nick collision from services.)
  338. # [17:43] * Quits: sid0_ (n=sid0@unaffiliated/sid0) (Nick collision from services.)
  339. # [17:43] * sid0__ is now known as sid0
  340. # [17:43] * sid0 is now known as sid0____________
  341. # [17:44] * Joins: mstange (n=markus@buntes215.wohnheim.uni-kl.de)
  342. # [17:44] * sid0____________ is now known as sid0
  343. # [17:45] * Joins: archtec2 (n=stanv@83.228.56.37)
  344. # [17:45] * Quits: archtech (n=stanv@83.228.56.37) (pratchett.freenode.net irc.freenode.net)
  345. # [17:45] * Quits: olliej (n=oliver@17.246.16.102) (pratchett.freenode.net irc.freenode.net)
  346. # [17:46] * Joins: michaeln1 (n=michaeln@nat/google/x-ebe782c07d1696ee)
  347. # [17:46] * Quits: pesla\work (n=retep@procurios.xs4all.nl) ("( www.nnscript.com :: NoNameScript 4.21 :: www.esnation.com )")
  348. # [17:48] * Quits: mat_t (n=mattomas@nat/canonical/x-51952978b8a5c569) (pratchett.freenode.net irc.freenode.net)
  349. # [17:48] * Quits: gsnedders (n=gsnedder@host86-136-52-180.range86-136.btcentralplus.com) (pratchett.freenode.net irc.freenode.net)
  350. # [17:48] * Quits: michaeln (n=michaeln@nat/google/x-fff924095b33bc8c) (pratchett.freenode.net irc.freenode.net)
  351. # [17:50] * Joins: sid0_ (n=sid0@unaffiliated/sid0)
  352. # [17:50] * Joins: archtech (n=stanv@83.228.56.37)
  353. # [17:50] * Joins: olliej (n=oliver@17.246.16.102)
  354. # [17:50] * Quits: olliej (n=oliver@17.246.16.102) (Read error: 104 (Connection reset by peer))
  355. # [17:51] * Joins: olliej (n=oliver@17.246.16.102)
  356. # [17:51] * Quits: taf2_ (n=taf2@38.99.201.242)
  357. # [17:51] * Quits: harig (n=opera@59.90.71.35) (Read error: 145 (Connection timed out))
  358. # [17:52] * Joins: mat_t (n=mattomas@nat/canonical/x-6b01242241ccc9d4)
  359. # [17:53] * Joins: gsnedders (n=gsnedder@host86-136-52-180.range86-136.btcentralplus.com)
  360. # [18:00] * Joins: dave_levin (n=dave_lev@72.14.227.1)
  361. # [18:01] * Quits: mstange (n=markus@buntes215.wohnheim.uni-kl.de) ("ChatZilla 0.9.84-2009030221 [Firefox 3.6a1pre/20090505030940]")
  362. # [18:01] * Joins: sid0__ (n=sid0@unaffiliated/sid0)
  363. # [18:02] * Quits: archtech (n=stanv@83.228.56.37) (Connection timed out)
  364. # [18:03] * Quits: sid0_ (n=sid0@unaffiliated/sid0) (Connection timed out)
  365. # [18:07] * Joins: Maurice (n=copyman@5ED548D4.cable.ziggo.nl)
  366. # [18:07] * Quits: bgalbraith (n=bgalbrai@63.245.220.241)
  367. # [18:09] * Joins: bgalbraith (n=bgalbrai@corp-241.mountainview.mozilla.com)
  368. # [18:10] * Joins: zcorpan (n=zcorpan@c83-252-196-43.bredband.comhem.se)
  369. # [18:10] * Joins: annevk5 (n=annevk@5ED2D22C.cable.ziggo.nl)
  370. # [18:14] * archtec2 is now known as archtech
  371. # [18:15] * Quits: shepazu (n=schepers@adsl-144-137-232.rmo.bellsouth.net)
  372. # [18:19] * Quits: sid0 (n=sid0@unaffiliated/sid0) (Read error: 110 (Connection timed out))
  373. # [18:21] * Quits: mpt (n=mpt@canonical/launchpad/mpt) (No route to host)
  374. # [18:22] * Quits: svl (n=chatzill@a194-109-2-36.dmn.xs4all.nl) ("And back he spurred like a madman, shrieking a curse to the sky.")
  375. # [18:24] <tantek> Hixie, feel free to discuss/mention any bugs you find in the hCard spec in #microformats
  376. # [18:25] <tantek> see of course http://microformats.org/wiki/hcard-faq and http://microformats.org/wiki/hcard-issues
  377. # [18:25] <tantek> (some are known/resolved already, feel free to add "+1s" to any issues you also encounter/agree with)
  378. # [18:30] * Joins: dglazkov (n=dglazkov@nat/google/x-2b4445293f3cb37a)
  379. # [18:30] * Quits: weinig (n=weinig@c-67-180-35-124.hsd1.ca.comcast.net)
  380. # [18:31] * Quits: roc (n=roc@65.123.0.210)
  381. # [18:42] <Philip`> http://www.cssquirrel.com/comic/?comic=16
  382. # [18:43] * sid0__ is now known as sid0
  383. # [18:50] * Joins: mlpug (n=mlpug@91.156.60.13)
  384. # [18:54] * Joins: taf2_ (n=taf2@38.99.201.242)
  385. # [18:55] * Parts: zcorpan (n=zcorpan@c83-252-196-43.bredband.comhem.se)
  386. # [18:57] <Philip`> Hixie: Tantek probably shouldn't be underlined in http://www.whatwg.org/specs/web-apps/current-work/multipage/acknowledgements.html
  387. # [19:01] <jgraham> Philip`: That's the secret symbol to indicate that tantek is the evil puppetmaster who exerts sole control of the HTML5 spec
  388. # [19:02] <jgraham> Although now I have said that, I will have to kill evenyone on the channel. Except tantek of course
  389. # [19:04] * Joins: mpt (n=mpt@canonical/launchpad/mpt)
  390. # [19:04] * Philip` watches tantek perform the Dance of Despair and Disillusionment
  391. # [19:06] * Joins: maikmerten (n=maikmert@Z893e.z.pppool.de)
  392. # [19:06] <gsnedders> Philip`: it's because it's he's a span element, and Hixie is assuming all span elements are incomplete xref
  393. # [19:18] * Joins: weinig (n=weinig@nat/apple/x-a4f713da61a7f4ef)
  394. # [19:24] <takkaria> Philip`: it does seem the RDFa people see Hixie as having strong bias toward microformats
  395. # [19:31] * gsnedders doesn't like either RDFa or µf
  396. # [19:32] <takkaria> I'm also not much liking the rampant paranoia people seem to have
  397. # [19:37] <tantek> takkaria, at least such rampant paranoia is itself revealed/archived publicly so you can prioritize such input appropriately.
  398. # [19:37] * gsnedders waits for that to be quoted by MLW
  399. # [19:38] * Joins: svl (n=me@ip565744a7.direct-adsl.nl)
  400. # [19:39] <tantek> takkaria, and regarding biases, Hixie tends to have a bias for productive pragmatism, and microformats certainly have a stated bias for productive pragmatism, so it should not be much of a surprise that some overlap in opinions is the result.
  401. # [19:39] <takkaria> sure
  402. # [19:40] <tantek> gsnedders, do you like semantic HTML?
  403. # [19:40] <gsnedders> taYes
  404. # [19:40] <gsnedders> tantek: Yes, even.
  405. # [19:40] <takkaria> I have no side in this debate, and some of the accusations being bandied around are just mad
  406. # [19:42] <tantek> takkaria, while some of the accusations could certainly be taken as "mad", they do serve two useful functions, 1. the above-noted prioritization enabler, and 2. sometimes even "mad" accusations may actually be a sign of some other issue (other than the one made in a "mad" way) that is worthy of consideration.
  407. # [19:43] <takkaria> well, it's quite clear that Shelley doesn't feel valued at all
  408. # [19:43] <takkaria> but I think it's a bit extreme to say that Hixie gave you vetting powers over the use cases list
  409. # [19:44] <tantek> I believe Hixie has made this point (I don't have a citation) that sometimes even trolls (or more precisely, input of a trolling nature) can be helpful with the issues that are unintentionally point out.
  410. # [19:45] <takkaria> yeah, I've seen him make it
  411. # [19:45] <tantek> takkaria, such a statement ("gave ... vetting power over") is both extreme, and has the flaw of being a deliberate exaggeration, likely to construct the basis of a strawman argument. http://en.wikipedia.org/wiki/Straw_man (logical fallacy)
  412. # [19:46] <tantek> such constructs are not uncommon when arguing theoretical points in (perhaps mostly) academic forums
  413. # [19:47] <takkaria> indeed, I do have something of a grasp of argumentative logic and fallacies. :)
  414. # [19:47] <tantek> the only logical steps to take are to: 1. point out the strawman fallacy (and any other logical fallacies being made in defense of positions that are typically indefensible with evidence) and 2. (de)prioritize/filter input from such sources accordingly over time.
  415. # [19:48] <tantek> takkaria - no lack of grasp was implied :)
  416. # [19:48] * Quits: hdh (n=hdh@58.187.19.111) (Read error: 104 (Connection reset by peer))
  417. # [19:50] <Hixie> tantek: the thing is, i find it an especially weird accusation given that most of our conversation was me whining about problems with microformats :-)
  418. # [19:50] * Quits: drostie (n=hopkins@5354256F.cable.casema.nl) (Remote closed the connection)
  419. # [19:50] <tantek> the 2. step is necessary in order to avoid or at least minimize a (perhaps unintentional, but still effective) DOPA attack from such sources. http://tr.im/dopa
  420. # [19:51] * Quits: dolske (n=dolske@firefox/developer/dolske)
  421. # [19:54] <tantek> Hixie, it is not that weird, in that the paranoid are quite skilled at inserting however many layers of abstraction / meaning are necessary in order to twist a quote to justifying/bolstering their claim.
  422. # [19:54] <tantek> Of course, if the very intent of the paranoid is to push for positions advocating for increasing layers of abstraction / meaning, then resultant accusations exhibit a level of meta-irony that itself has value along the axis of humor.
  423. # [19:57] * Quits: weinig (n=weinig@nat/apple/x-a4f713da61a7f4ef)
  424. # [19:58] <Hixie> i think it's important to distinguish between certain people who have clearly just been spurned (like whoever does lastweek, and probably shelley) and people who are taking what those people say as gospel (like a lot of other people)
  425. # [19:58] <Hixie> the latter group aren't "paranoid", they've just been misled
  426. # [19:58] <Hixie> imho
  427. # [20:02] <tantek> agreed, that's a useful distinction
  428. # [20:04] <Hixie> of course the most useful thing to work out here would be why those who got spurned feel like they got spurned
  429. # [20:05] <tantek> Hixie, not sure if that is necessarily "useful" per se, if by "useful" you mean productive, and if by productive, we mean the next highest quality/quantity yield next-actions you could take, or even close. econ / diminishing-returns etc.
  430. # [20:06] <Hixie> i think finding out why people feel spurned is important from a moral perspective and from a pragmatic "stop making people run away" perspective
  431. # [20:06] <tantek> sometimes you just have let different ideas compete in the market, and let some succeed, while others struggle, wither, and are abandoned.
  432. # [20:07] <Hixie> yeah, but as i don't want mine to wither (since that would mean my life was mostly wasted), i'd rather figure out why it is people are abandoning it, and fix that problem
  433. # [20:09] <Hixie> speaking of which, i spoke to yet another group of people working on microdata stuff yesterday, who have neither a horse in the mf space not the rdfa space, and heard yet again the same thing -- "we want to have a generic syntax so that people can mark up stuff that is then exposed to an API, without hard-coding specific class values into the parser"
  434. # [20:09] <tantek> Hixie, lest we forget history, it was the browser makers (and you and I included, at the time) who felt "spurned" by W3C staff and advocates for XHTML2+XForms+SVG+namespaces+(insert X-* specs here) at the Web Applications and Compound Documents workshop in 2004-04 in San Jose. http://www.w3.org/2004/04/webapps-cdf-ws/minutes-20040601.html
  435. # [20:09] <Hixie> apparently they are considering using a subset of rdfa instead
  436. # [20:10] <Hixie> though their faces when they said that indicated that they felt that was more of a situation they were being forced into than a good thing
  437. # [20:10] * Joins: dbaron (n=dbaron@guest-226.mountainview.mozilla.com)
  438. # [20:10] <tantek> right, a priori theoretical solutions rarely take root in bottom-up market-decided dynamics
  439. # [20:10] <Hixie> tantek: oh i have been spurned many times in the standards world, that doesn't mean it's ok for me to spurn others :-)
  440. # [20:11] <tantek> rather, they tend to be adopted in top-down forced into situations
  441. # [20:11] <tantek> large corporations etc.
  442. # [20:11] * Joins: onar_ (n=onar@17.244.69.252)
  443. # [20:11] * Quits: onar_ (n=onar@17.244.69.252) (Remote closed the connection)
  444. # [20:11] * Joins: dolske_ (n=dolske@corp-241.mountainview.mozilla.com)
  445. # [20:12] * Joins: onar_ (n=onar@17.226.23.135)
  446. # [20:12] <tantek> Hixie, the point is that both WHATWG and microformats efforts were created because you and I both thought we it was worth the time and effort to explore alternatives to what the W3C was saying was "the right thing".
  447. # [20:12] * Joins: dolske- (n=dolske@corp-241.mountainview.mozilla.com)
  448. # [20:12] <Hixie> yes... not sure what the relevance is here
  449. # [20:12] <Hixie> can you elaborate?
  450. # [20:12] <tantek> the market-testing of ideas
  451. # [20:13] <tantek> here we are almost 5 years after that infamous workshop (which perhaps we should refer to with some sort of dramatic name like, the Great Web Semantics Schism)
  452. # [20:13] <Hixie> the market-testing of ideas is telling me that people want something with the extensibility of rdfa with a syntax nicer than microformats.
  453. # [20:13] <tantek> "wants" are not market-testing
  454. # [20:14] * Quits: ap (n=ap@194.154.88.47)
  455. # [20:14] <tantek> "buys" (what people actually spend time developing and then publishing) are market-testing
  456. # [20:14] <Hixie> when people start writing code, i consider that market testing.
  457. # [20:14] <tantek> writing code is a start
  458. # [20:14] <tantek> look at the stats on the # of abandoned sourceforge projects
  459. # [20:14] <Philip`> "gsnedders doesn't like either RDFa or µf" - I'll go one step further, and not like data at all
  460. # [20:15] <tantek> "start writing code" is insufficient. actually shipping something, better. interoperating with another implementation, now we're talking.
  461. # [20:15] <Philip`> I think we should remove all the data from web pages, and use the saved bandwidth for adding more colours
  462. # [20:16] <Hixie> i've spoken to multiple groups during this research, from big companies like yahoo to small companies like manu's, from individuals doing front-line work like livebrum.co.uk to researchers like tbl, and many of these are actively investing time and money into implementing solutions that do both mf and rdfa, and with the exception of people actively working on one or the other, they are complaining about both.
  463. # [20:16] * tantek wonders if we should plan a party for 2009-06-01 - the 5 year anniversary of the Great Web Semantics Schism.
  464. # [20:16] <Hixie> many of these are in fact shipping (e.g. yahoo, shipping support for both mf and rdfa, livebrum.co.uk, shipping mf support)
  465. # [20:16] * Joins: mstange (n=markus@buntes215.wohnheim.uni-kl.de)
  466. # [20:17] <tantek> Right, and five years ago, there were numerous vendors shipping interoperable implementations of XForms - I believe the XForms page on the W3C site has record of this.
  467. # [20:17] * Quits: mat_t (n=mattomas@nat/canonical/x-6b01242241ccc9d4) (Remote closed the connection)
  468. # [20:17] <Hixie> sure, there's nothing wrong with xforms itself.
  469. # [20:17] <tantek> And yet, you thought it was worth pursuing WebForms2 and I though it was worth pursuing microformats.
  470. # [20:17] <Hixie> it's just not good for web browsers.
  471. # [20:18] <tantek> you could say the same thing about RDF
  472. # [20:18] <Hixie> i am saying the same thing about rdf and microformats :-)
  473. # [20:18] <Hixie> rdfa, rather
  474. # [20:18] <Hixie> the people on the front lines that i've spoken to haven't mentioned rdf
  475. # [20:19] <Philip`> "Schism" sounds rude
  476. # [20:19] <Hixie> (not counting people who are actively working on rdfa, who do mention rdf)
  477. # [20:19] <tantek> except that microformats evolved from established well-worn modern web design practices of semantic HTML (later documented as POSH)
  478. # [20:19] <Hixie> (working on the spec, i mean)
  479. # [20:19] <Hixie> just because it evolved from POSH doesn't mean it's solving all the relevant problems people are having
  480. # [20:21] <tantek> Philip` - anything less than schism would be politically correct watering down. Nothing short of a schism happened that day almost five years ago.
  481. # [20:21] <tantek> When all the browser vendors and professional web designers/developers found themselves agreeing, and also disagreeing with (most of) the W3C staff and X-* markup language advocates - it became quite clear to everyone in the room that a schism had occured.
  482. # [20:22] <Hixie> the big indicator was that Microsoft and Sun were in agreement
  483. # [20:22] <Hixie> and that _never_ happens
  484. # [20:22] * Joins: pauld (n=pauld@host81-158-125-194.range81-158.btcentralplus.com)
  485. # [20:22] <tantek> and Opera, and Mozilla
  486. # [20:22] <tantek> and Redhat as well!
  487. # [20:25] <tantek> Hixie - I'll politely remind you that microformats both never deemed to solve "all" problems people are having (leaving out "relevant" as different people have different measures for that), and even made that a non-goal: http://microformats.org/wiki/microformats#microformats_are
  488. # [20:25] <Hixie> just because you explicitly set out to not solve all the problems people want solved doesn't mean people will like it :-)
  489. # [20:26] <Hixie> the point is everyone i speak to who is using microformats complains that it doesn't provide the flexibility they want and they end up implementing other things too (rdfa, or a subset thereof, typically)
  490. # [20:26] * Quits: pauld (n=pauld@host81-158-125-194.range81-158.btcentralplus.com) (Client Quit)
  491. # [20:26] <Hixie> in my mind, that tells me there is a problem i need fixing in html5 with a solution that isn't just "use microformats"
  492. # [20:27] <Hixie> (it's also not just "use rdfa" given that those same people complain about rdfa, e.g. saying things like "oh we'll ignore the xmlns attributes and just prebind certain prefixes that we recognise")
  493. # [20:27] <tantek> or you could punt on the problem altogether in html5
  494. # [20:27] <takkaria> HIXIE IN RDFA BIAS SHOCKER
  495. # [20:28] <Hixie> i could, but i think it's clear that it's not a minor theoretical problem at this point
  496. # [20:28] <tantek> is everything that is not a minor theoretical problem then considered in the scope of something you need to fix in HTML5?
  497. # [20:29] <Philip`> It's also clear that it's a difficult problem to solve, and it's not a problem with a clear solution whose usage is being blocked by limitations of the HTML spec
  498. # [20:29] <tantek> alternative, punt it from HTML5, and let the market continue to evolve a set of different approaches
  499. # [20:29] <tantek> Philip - agreed.
  500. # [20:30] <Philip`> On the other hand, many solutions *are* blocked by limitations of the HTML spec because they insist on being valid HTML
  501. # [20:31] <Philip`> so perhaps it does really need changes to the spec
  502. # [20:31] <Hixie> tantek: things that go into text/html documents are in scope, so i think this is in scope
  503. # [20:31] <Philip`> (though maybe just the changes that are needed to facilitate the solution, rather than providing a whole actual solution)
  504. # [20:31] <tantek> Hixie - not necessarily - it could just be an extensibility mechanism that you leave "as is".
  505. # [20:32] <Hixie> the existing extension mechanisms aren't enough.
  506. # [20:33] <tantek> are you requesting proposals for incrementally adding to existing extension mechanisms to fix specific pain points found through real world experience, or ...
  507. # [20:34] <tantek> proposals to completely replace existing extension mechanisms with brand new mechanisms based more on attempts to solve universal / global problems?
  508. # [20:34] <Hixie> neither, proposals are not something i'm short on right now :-/
  509. # [20:34] <tantek> heh
  510. # [20:34] <Hixie> for every one use case, people put forward fifteen solutions
  511. # [20:34] * Quits: bgalbraith (n=bgalbrai@corp-241.mountainview.mozilla.com)
  512. # [20:34] <tantek> I'm familiar with that problem, but perhaps not as many as 15, more like 3-5.
  513. # [20:35] <Hixie> i read 15,000 lines' worth of e-mails and got about 150 lines' worth of use cases out of it
  514. # [20:35] <Philip`> I think you should solve the problem by saying that all HTML documents are conforming, and conformance checkers must always return success, and then that'd let people do whatever solutions they want
  515. # [20:35] * Joins: bgalbraith (n=bgalbrai@corp-241.mountainview.mozilla.com)
  516. # [20:36] <tantek> getting back to - gsnedders - you said you like semantic HTML right?
  517. # [20:38] <Philip`> Alternative solution: All microdata should be centrally collated and processed and presented, like in Wolfram Alpha
  518. # [20:38] <Philip`> It's like the Semantic Web, but without the Web
  519. # [20:38] * Quits: bgalbraith (n=bgalbrai@corp-241.mountainview.mozilla.com) (Client Quit)
  520. # [20:38] <Philip`> which frees you from all the horrid problems of the web
  521. # [20:38] <Philip`> so it's bound to work
  522. # [20:39] <Hixie> hah
  523. # [20:39] <Hixie> one of the scenarios in the e-mail i sent yesterday was basically "i have a bunch of text and want to make wolfram alpha"
  524. # [20:40] <Hixie> so it's come up!
  525. # [20:41] * jgraham doesn't particularly like semantic markup, at least for some definition of semantic
  526. # [20:42] <Philip`> Problem: i have a bunch of text and want to make wolfram alpha. Solution: write five million lines of Mathematica. Easy!
  527. # [20:43] <Hixie> it'll take more than that
  528. # [20:43] <Hixie> it'll also take five years of painstaking data entry
  529. # [20:43] <jgraham> You have to write the right 5 million lines
  530. # [20:43] <Philip`> Well, that too
  531. # [20:44] <Philip`> jgraham: All you have to do is find the right cellular automaton that will generate those lines, so it's not that much effort really
  532. # [20:46] <Hixie> remind me not to put Philip` it charge of solving any important problems :-P
  533. # [20:54] * Philip` sees http://urn.kb.se/resolve?urn=urn:nbn:se:liu:diva-12095 describing some extensions to the <canvas> API
  534. # [20:54] <Philip`> (from Chapter 6)
  535. # [20:55] <Philip`> (in which lineWidth is basically a gradient with multiple stops, so the width varies along the line)
  536. # [20:55] <Philip`> (and lineCaps are basically path objects that get drawn automatically)
  537. # [20:56] * Joins: virtuelv (n=virtuelv@084202133045.customer.alfanett.no)
  538. # [20:58] * Joins: pauld (n=pauld@host81-158-125-194.range81-158.btcentralplus.com)
  539. # [20:59] * Joins: jwalden (n=waldo@corp-241.mountainview.mozilla.com)
  540. # [21:01] * Quits: mstange (n=markus@buntes215.wohnheim.uni-kl.de) (Remote closed the connection)
  541. # [21:02] * Quits: davidb (n=davidb@bas4-toronto06-1279277705.dsl.bell.ca)
  542. # [21:02] * Joins: mstange (n=markus@buntes215.wohnheim.uni-kl.de)
  543. # [21:02] * Joins: davidb (n=davidb@bas4-toronto06-1279277705.dsl.bell.ca)
  544. # [21:03] * Joins: weinig_ (n=weinig@nat/apple/x-d6ed8a51b8845986)
  545. # [21:03] <tantek> gsnedders, if you like semantic HTML, I'd be interested to know which step along this path you end up not liking things: semantic HTML -> POSH -> microformats as a subset of POSH
  546. # [21:04] * Quits: ZombieLoffe (n=e@unaffiliated/zombieloffe) (Read error: 60 (Operation timed out))
  547. # [21:16] * weinig_ is now known as weinig|away
  548. # [21:18] * Quits: myakura (n=myakura@p1120-ipbf2303marunouchi.tokyo.ocn.ne.jp) ("Leaving...")
  549. # [21:19] <jgraham> tantek: What do you mean by semantic HTML?
  550. # [21:20] <gsnedders> tantek: µf seem to be too case-specific, and RDF is too general, basically.
  551. # [21:25] * Quits: pauld (n=pauld@host81-158-125-194.range81-158.btcentralplus.com)
  552. # [21:28] <tantek> jgraham - this is a decent summary: http://microformats.org/wiki/semantic-html
  553. # [21:28] * Quits: dbaron (n=dbaron@guest-226.mountainview.mozilla.com) ("8403864 bytes have been tenured, next gc will be global.")
  554. # [21:29] * Joins: dbaron (n=dbaron@corp-241.mountainview.mozilla.com)
  555. # [21:30] <tantek> gsnedders the only real difference between POSH, in particular semantic class names and rel values, and microformats is that microformats specify a shared vocabulary (developed via the principles and the process) where as with POSH, there is little or no sharing of vocabulary. Is it the sharing of vocabulary you think is too case-specific?
  556. # [21:31] * Joins: pauld (n=pauld@81.158.125.194)
  557. # [21:33] <gsnedders> tantek: No, well, sort of. I see being able to use a generic parser to be a good idea™, but also I think RDF's approach of just mapping everything as object -> property (or whatever the terminology is) is too broad, and that there need
  558. # [21:34] <gsnedders> *needs to be more, um, let me say "non-orthogonal parallel mapping at the same level", which is really clumsy but is the sort of thing I mean
  559. # [21:34] <gsnedders> Anyhow, I need to do things more useful :)
  560. # [21:34] * Quits: maikmerten (n=maikmert@Z893e.z.pppool.de) (Remote closed the connection)
  561. # [21:35] <tantek> gsnedders, there are lots of "good idea™"s, but in a resource constrained universe (say, a universe bound by the laws of physics), you end up having to prioritize some ideas over others, which leads one to realize, that that prioritization is actually what is important. And ideas that may seem "good" in the abstract (in a vacuum) actually seem not so essential when compared to other ideas that you'd rather spend your limited re
  562. # [21:36] <tantek> the notion of "making it really easy to write a parser" is one such seemingly good idea in the abstract.
  563. # [21:38] <jgraham> tantek: That page wasn't really very helpful. It seemed to be a list of html elements
  564. # [21:39] <jgraham> I really meant "what characteristics do you think markup has to have to be considered semantic" and, as a followup, "why is having those characteristics a good thing"?
  565. # [21:40] * Quits: virtuelv (n=virtuelv@084202133045.customer.alfanett.no) (Remote closed the connection)
  566. # [21:40] <jgraham> http://microformats.org/wiki/posh has some more clues
  567. # [21:44] <tantek> jgraham, I prefer to answers first with the immediate pragmatic answers ("what do you mean by semantic HTML") and only after those are discussed/understood, to raise the conversation to an abstract level ("characterists", what is "good")
  568. # [21:45] <tantek> in my experience, most people simply want the immediate pragmatic answers to the questions, and pragmatic answers form a better basis for abstract discussion than pure abstract discussion.
  569. # [21:45] <jgraham> tantek: Hence the second question being the followup to the first
  570. # [21:45] <tantek> precisely, and yes, the POSH page goes into more depth
  571. # [21:46] * Quits: pauld (n=pauld@81.158.125.194)
  572. # [21:47] * Joins: drostie (n=hopkins@5354256F.cable.casema.nl)
  573. # [21:48] <tantek> jgraham, again, using actual examples to illustrate further, take a look at slide 9 "... the Semantic Spectrum" here: http://tantek.com/presentations/2003SXSW/stylesheets.html
  574. # [21:52] <jgraham> So the rules seem to be a) validity, b) Replace some elements (<b>, <br) with other elements, c) use @id rather than <a name=""> d) Use some patterns (only one seems to be documented?) e) Use classnames that identify abstract concepts
  575. # [21:54] <tantek> jgraham - many semantic (X)HTML patterns have been documented across the web, and the POSH wiki page serves as an attempt to provide a collection and index of sorts
  576. # [21:54] <jgraham> oh and f) "..."
  577. # [21:56] <jgraham> tantek: OK, so coming back to my question, a) does not seem to be a question of semantics. It is likely ood for the producer of the page (since valid markup is often easier to debug) and consumers (since the HTML 5 parsing algorithm is not yet widely implemented)
  578. # [21:56] <jgraham> *good
  579. # [21:56] <jgraham> but not semantics
  580. # [21:57] <tantek> HTML4 validators typically help point out problems which often are problems with the semantics of the page
  581. # [21:57] <tantek> thus a first "easy" step towards semantic HTML (assuming that's your goal) is to use such validators (e.g. W3C) to find such problems in your code and fix them.
  582. # [21:58] <tantek> in theory validity and semantics are orthogonal
  583. # [21:58] <tantek> in practice validity helps increase semantics
  584. # [21:58] <jgraham> b) Is closer to the issue that I care about. It is very easy for me to make an argument for using certian elements "correctly" (<h1>-<h6>, <th>, <td>) because they are used consistently enough that UAs can process them in a useful way (and many do so)
  585. # [21:59] <jgraham> So if you use <td> when you mean <th> it actually hurts users
  586. # [22:00] <jgraham> But <strong> is used to mean a whole laundry-list of things and in practice there is no difference between <strong> and <b> in UAs. So what is the argument for caring?
  587. # [22:01] <jgraham> Similarly, what is the harm ifI make a bibliography and mark up all the titles of the works using <i>?
  588. # [22:01] <jgraham> It's not like any UA could do anything useful with the information if I marked it up in a differnt way
  589. # [22:02] <jgraham> Yet people spend hours debating the "right" markup to use in every tny situation
  590. # [22:05] <tantek> jgraham, there is some abuse of <strong> no doubt. my own personal anecdotal experience is that there a <strong> *is* a bit better used (less abuse) in practice than <b>. using more semantically named tags also helps reinforce a more semantic mindset when authoring (rather than the "HTML as a way to print pixels" mindset).
  591. # [22:05] <tantek> and I have seen some reasonable arguments for repurposing of <i> as a semantic element for bibliographic (and other such) /instances/
  592. # [22:06] <tantek> thus yes, there are points of semantic HTML development which are debated, and there are points that are well agreed upon (often as a result of such debate occuring)
  593. # [22:06] <tantek> these debates started occurring mostly on blogs in the early 2000s.
  594. # [22:07] <jgraham> tantek: I know. I was there :)
  595. # [22:09] * Quits: mlpug (n=mlpug@91.156.60.13) (Remote closed the connection)
  596. # [22:09] * Quits: sid0 (n=sid0@unaffiliated/sid0) (Read error: 110 (Connection timed out))
  597. # [22:09] <jgraham> tantek: If you could get enough people to agree on and use say <i> for a specific purpose so that UAs could actually do something other than guess it should be emphasised in some way then that might be worthwhile (hence <h1>-<h6> being good). But I don't see that as possible
  598. # [22:09] <tantek> jgraham, if you have blog posts or presentations on semantic HTML from those early debates/discussion, I encourage you to add links to them to http://microformats.org/wiki/posh#POSH_Blog_posts
  599. # [22:10] <tantek> jgraham, I tend to agree with your assessment of <i>.
  600. # [22:10] <jgraham> tantek: I think I just commented on other people's blogs (and had somewhat different opinions to the ones I have now)
  601. # [22:12] <Philip`> Blogs existed in the early 2000s?
  602. # [22:12] <Philip`> Time goes so fast :-(
  603. # [22:13] <tantek> jgraham - glad to see you're still interested in the subject :)
  604. # [22:14] <tantek> I just started this page a few days ago based on some discussion in #microformats and it could definitely use some (e.g. your) help with more: http://microformats.org/wiki/posh-patterns
  605. # [22:14] <tantek> it would be great to have a library of posh patterns that folks could simply pick up and re-use
  606. # [22:15] <tantek> because such posh patterns do exist - they just were mostly debated/created before wiki-documenting/listing/indexing such things became established practice
  607. # [22:15] <jgraham> tantek: Well I don't hang out in #whatwg to look cool :)
  608. # [22:17] * Philip` does
  609. # [22:17] <Philip`> (I don't think it works)
  610. # [22:18] * Joins: sid0 (n=sid0@unaffiliated/sid0)
  611. # [22:21] * Quits: pmuellr (n=pmuellr@nat/ibm/x-b4988449c31dca5b)
  612. # [22:22] <gsnedders> Hixie: wrt charset equivalence, do browsers really treat XML which is US-ASCII as Windows-1252?
  613. # [22:26] <Philip`> http://realtech.burningbird.net/semantic-web/rdf-and-rdfa/arbitrary-vocabularies-and-other-crufty-stuff - hmm, seems pretty clear to me that formats/protocols without a specific vocabulary still have some value, e.g. they can save you some effort when writing specific tools or learning a specific language
  614. # [22:28] <Philip`> and the issue is more about how they don't solve the entire problem, because you still need to write a specific tool or design a specific language on top of those things, and then the problem is whether that generic framework plus the extra task-specific work is more expensive than a different generic framework (e.g. one that's far more minimalist) plus the extra task-specific tools
  615. # [22:28] <Hixie> gsnedders: other than IE, i believe so, yes, but test if you want to find out for sure
  616. # [22:28] * Quits: mstange (n=markus@buntes215.wohnheim.uni-kl.de) ("ChatZilla 0.9.84-2009030221 [Firefox 3.6a1pre/20090505030940]")
  617. # [22:28] <gsnedders> Yuk.
  618. # [22:31] <Philip`> gsnedders: data:text/xml;charset=us-ascii,<html%20xmlns='http://www.w3.org/1999/xhtml'><script>alert('%80'.charCodeAt(0))</script></html> ?
  619. # [22:31] <jgraham> Philip`: One of the common themes in embedded metadata conversations is tehnically inclined people saying something equivalent to "and then we will autogenerate the UI"
  620. # [22:31] <tantek> Philip, the problems are not merely on the expense (cost) side, but in experience, more so on the results (benefits) side.
  621. # [22:31] <tantek> jgraham - indeed.
  622. # [22:32] <Philip`> Opera says 8364; Firefox says "Error: no element found / Source File: jar:file:///opt/firefox/chrome/toolkit.jar!/content/global/bindings/dialog.xml / Line: 1" (?)
  623. # [22:32] <gsnedders> Philip`: Odd.
  624. # [22:32] <gsnedders> Saf says 8364
  625. # [22:33] * Philip` doesn't know why Firefox gives that error message
  626. # [22:33] <jgraham> Philip`: Which tends to suggest that people underestimate the extra effort that you have to put in after you have a parser to do something usable.
  627. # [22:33] <tantek> one of the benefits that sought but often unsaid (unfortunately) is accurate data, and it turns out, the more human-usable the data (both authoring and viewing) the more accurate the data tends to be.
  628. # [22:33] <tantek> jgraham - precisely correct.
  629. # [22:34] <jgraham> On the other hand there is value in not having to continually reinvent the lower layers just because the high layer stuff is non-trivial
  630. # [22:35] <tantek> you could broaden that aforementioned "common theme" to technically inclined people tending to design for machines (e.g. parsing the content) at the expense of humans (e.g. authoring the content)
  631. # [22:35] <Philip`> As someone who's spent substantial amounts of time reverse-engineering weird custom binary file formats for some games, I'm certainly happy when they just use XML, because it saves a load of that low-level work
  632. # [22:36] <Philip`> (Of course they then compress the XML in a custom binary format, but at least that's just one format and not half a dozen for all the different types of data)
  633. # [22:36] <tantek> sure, having hints at the apparent boundaries and types of data helps a lot
  634. # [22:36] * Quits: taf2_ (n=taf2@38.99.201.242)
  635. # [22:36] <Philip`> It's much easier to process data when it's in a tree of elements/attributes than when it's in a stream of bytes
  636. # [22:37] * Joins: virtuelv (n=virtuelv@084202133045.customer.alfanett.no)
  637. # [22:37] <Philip`> Presumably the RDFa idea is it's much easier to process data when it's in a graph of subject-predicate-object triples than when it's in a tree of elements/attributes
  638. # [22:38] <tantek> it also turns out to be often more easy/robust to author data in a tree of elements/attributes that's displayed in some readily viewable tree-like form rather than a stream of bytes.
  639. # [22:38] <tantek> so the move from bytes -> tree of data made sense on both the cost and benefit side of the equation
  640. # [22:40] <Philip`> You don't author data in a tree, you author it in a stream of bytes that represents a serialised tree
  641. # [22:40] <tantek> but at some point, the requirements to add structure actually cost more than humans are willing to do to author it.
  642. # [22:40] <Philip`> (Well, I suppose you could have a fancy XML editor that lets you edit in a tree-like UI)
  643. # [22:40] <Philip`> (and I suppose some people must use such tools)
  644. # [22:41] <tantek> people author outlines (trees) all the time. varying levels of headings and paragraphs etc. are another form of display trees
  645. # [22:41] <tantek> so yes, people *do* author in visual tree-like structures beyond just simple flat streams
  646. # [22:41] <tantek> where-ever such tree-like structures have already long evolved in human communications
  647. # [22:42] <tantek> thus it's easy for people to do, because it's part of human culture to do so
  648. # [22:42] <tantek> however, as soon as you talk about attempting to author things as a "graph" (and you're not talking about charts), all of that breaks down. authoring cost goes straight out the window, resulting in much less content. simple economics.
  649. # [22:43] * Quits: drostie (n=hopkins@5354256F.cable.casema.nl) (Remote closed the connection)
  650. # [22:43] <Philip`> Anyway, I guess I was trying to say that generic frameworks like XML and RDF aren't useless, so it's a more complex issue of tradeoffs
  651. # [22:44] <Philip`> (and other non-vocabulary-specific generic frameworks might have better tradeoffs for various use cases)
  652. # [22:44] <Philip`> and I'll let someone else work out the actual details :-)
  653. # [22:45] * Joins: sid0_ (n=sid0@unaffiliated/sid0)
  654. # [22:46] <tantek> Philip - it's precisely that complexity of the tradeoffs that is all too often ignored, and specifically, the harder to quantify human aspects (ease of authoring, mapping to existing understood and practiced modes of human communication etc.)
  655. # [22:46] <tantek> and so you get people debating about generic parsers instead, while the human related problems actually have a much bigger impact on both the overall cost and the benefits from any such system
  656. # [22:47] * Joins: drostie (n=hopkins@5354256F.cable.casema.nl)
  657. # [22:48] <tantek> it's precisely because of this (technically minded folks ignoring human aspects because such aspects are often much harder to understand and design for), that "What are microformats?" starts with "Designed for humans first and machines second" - http://microformats.org/
  658. # [22:48] <jgraham> I can certianly agree that one of the big problems wih RDF(a) is the ifficulty of thinking in the right mode to author it. Proponents seem to say things like "it's just four attributes" whilst failing to understand that the underlying data model is rather abstract and complex
  659. # [22:48] * Quits: drostie (n=hopkins@5354256F.cable.casema.nl) (Read error: 104 (Connection reset by peer))
  660. # [22:49] <Philip`> Yeah, it's a danger if you just consider the generic framework in isolation, rather than as part of a solution that includes some vocabulary and some vocabulary-specific processing in order to solve particular use cases, because you'll be optimising half of the costs and ignoring the rest
  661. # [22:50] * Joins: drostie (n=hopkins@5354256F.cable.casema.nl)
  662. # [22:50] <Philip`> jgraham: The underlying data model is trivial - it's just a big bag of subject-predicate-object triples, each of which is a URI
  663. # [22:51] <Philip`> The problem seems to be that the mapping between the underlying data model and the serialised byte stream is incredibly complex
  664. # [22:51] <jgraham> For some definition of trivial which means "comprehensible if you are the sort of person who could complete a degree in mathematics, linguistics, or a similar subject)
  665. # [22:52] <Philip`> s/incredibly/very/ perhaps
  666. # [22:52] <tantek> jgraham - agreed.
  667. # [22:52] <Philip`> and also the mapping between the underlying data model and the specific problem domain is often complex
  668. # [22:53] * Joins: franksalim (n=frank@adsl-75-61-86-233.dsl.pltn13.sbcglobal.net)
  669. # [22:54] <tantek> Philip not just mapping to/from the serialized byte stream, but mapping to/from structures long established/evolved in human cultures/communications, and indeed, specific problem domains, and thus specific use cases.
  670. # [22:54] * Joins: sid0__ (n=sid0@unaffiliated/sid0)
  671. # [22:55] <Philip`> jgraham: I remember being taught about nouns and verbs when I was in primary school and it wasn't really that complex, and these triples are basically the same thing :-p (except with URIs, which I wasn't taught about in primary school)
  672. # [22:55] * gsnedders wonders why Philip` wasn't taught URIs in primary school
  673. # [22:56] * Quits: sid0_ (n=sid0@unaffiliated/sid0) (Read error: 60 (Operation timed out))
  674. # [22:58] <Philip`> gsnedders: Because I'm not a young whippersnapper like you and we didn't have internets in school back then
  675. # [23:00] <Philip`> The highlight of our computing classes was modifying the Minesweeper high scores in Windows 3.1
  676. # [23:01] * Philip` also cunningly inserted a QuickBASIC program into autoexec.bat so it would play the Teletubbies tune after booting
  677. # [23:01] <gsnedders> Man, you guys had Teletubbies!?
  678. # [23:02] <gsnedders> It only started in 1997 according to Wikipedia!
  679. # [23:02] * Quits: sid0 (n=sid0@unaffiliated/sid0) (Read error: 110 (Connection timed out))
  680. # [23:02] <Philip`> Okay, maybe it wasn't actually in primary school, it was in the one that comes after primary school and before secondary school
  681. # [23:03] <jgraham> Philip`: I'm not that sure primary school level linguisics helps with RDF much
  682. # [23:06] * Joins: sid0 (n=sid0@unaffiliated/sid0)
  683. # [23:12] * Quits: ROBOd (n=robod@89.122.216.38) ("http://www.robodesign.ro")
  684. # [23:12] * Quits: sid0__ (n=sid0@unaffiliated/sid0) (Read error: 60 (Operation timed out))
  685. # [23:17] * Quits: archtech (n=stanv@83.228.56.37) (Read error: 113 (No route to host))
  686. # [23:23] * Joins: sid0_ (n=sid0@unaffiliated/sid0)
  687. # [23:24] * Quits: Maurice (n=copyman@5ED548D4.cable.ziggo.nl) ("Disconnected...")
  688. # [23:31] <franksalim> Has this happened yet? <sayrer> maybe we should ask for a new WG [http://krijnhoetmer.nl/irc-logs/whatwg/20090414#l-64]
  689. # [23:33] <Hixie> not to my knowledge
  690. # [23:33] <Hixie> other than the hybi list
  691. # [23:34] * Quits: sid0_ (n=sid0@unaffiliated/sid0) (Read error: 104 (Connection reset by peer))
  692. # [23:35] * Quits: Simetrical (n=Simetric@wikipedia/simetrical) (Remote closed the connection)
  693. # [23:36] * Joins: Simetrical (n=Simetric@wikipedia/simetrical)
  694. # [23:37] * Quits: weinig|away (n=weinig@nat/apple/x-d6ed8a51b8845986) (Read error: 104 (Connection reset by peer))
  695. # [23:37] * Joins: weinig (n=weinig@nat/apple/x-b5f5e9045a2ef4e9)
  696. # [23:38] * Quits: Simetrical (n=Simetric@wikipedia/simetrical) (Remote closed the connection)
  697. # [23:38] * Joins: Simetrical (n=Simetric@wikipedia/simetrical)
  698. # [23:39] * Quits: Simetrical (n=Simetric@wikipedia/simetrical) (Read error: 131 (Connection reset by peer))
  699. # [23:39] * Joins: Simetrical (n=Simetric@wikipedia/simetrical)
  700. # [23:39] * Joins: sid0_ (n=sid0@unaffiliated/sid0)
  701. # [23:39] * Quits: sid0 (n=sid0@unaffiliated/sid0) (Read error: 110 (Connection timed out))
  702. # [23:41] * Joins: taf2_ (n=taf2@c-68-49-245-59.hsd1.dc.comcast.net)
  703. # [23:49] * Quits: dglazkov (n=dglazkov@nat/google/x-2b4445293f3cb37a)
  704. # Session Close: Wed May 06 00:00:00 2009

The end :)