/irc-logs / w3c / #html-wg / 2008-08-05 / end

Options:

  1. # Session Start: Tue Aug 05 00:00:00 2008
  2. # Session Ident: #html-wg
  3. # [00:15] * Quits: marcos (marcos@124.171.129.80) (Ping timeout)
  4. # [00:20] * Joins: marcos (marcos@124.171.203.244)
  5. # [00:29] * Quits: marcos (marcos@124.171.203.244) (Quit: marcos)
  6. # [00:30] * Joins: marcos (marcos@124.171.203.244)
  7. # [00:34] * Quits: marcos (marcos@124.171.203.244) (Quit: marcos)
  8. # [00:41] * Joins: mjs (mjs@17.203.15.230)
  9. # [00:41] * Joins: marcos (marcos@124.171.203.244)
  10. # [00:44] * Quits: marcos (marcos@124.171.203.244) (Quit: marcos)
  11. # [00:57] * Quits: aroben (aroben@71.58.56.76) (Connection reset by peer)
  12. # [01:10] * Quits: mjs (mjs@17.203.15.230) (Quit: mjs)
  13. # [01:14] * Joins: adele_ (adele@17.203.15.174)
  14. # [01:15] * Quits: adele (adele@17.203.14.172) (Ping timeout)
  15. # [01:17] * Quits: tlr (tlr@128.30.52.30) (Quit: tlr)
  16. # [01:22] * Joins: scotfl (scotfl@70.64.14.62)
  17. # [01:32] * Joins: mjs (mjs@17.203.15.230)
  18. # [02:08] * Quits: adele_ (adele@17.203.15.174) (Client exited)
  19. # [02:08] * Joins: adele (adele@17.203.14.172)
  20. # [02:16] * Quits: mjs (mjs@17.203.15.230) (Quit: mjs)
  21. # [02:19] * Joins: Zeros (Zeros-Elip@67.154.87.254)
  22. # [02:21] * Joins: marcos (marcos@131.181.148.227)
  23. # [02:24] * Quits: Zeros (Zeros-Elip@67.154.87.254) (Ping timeout)
  24. # [02:42] * Joins: mjs (mjs@17.255.97.30)
  25. # [02:42] * Joins: mjs_ (mjs@17.203.15.230)
  26. # [02:45] * Quits: mjs (mjs@17.255.97.30) (Ping timeout)
  27. # [02:47] * Quits: adele (adele@17.203.14.172) (Quit: adele)
  28. # [02:56] * Joins: adele (adele@17.203.14.172)
  29. # [03:36] * Quits: adele (adele@17.203.14.172) (Quit: adele)
  30. # [04:09] * Quits: dbaron (dbaron@63.245.220.241) (Quit: 8403864 bytes have been tenured, next gc will be global.)
  31. # [05:17] * Joins: adele (adele@24.7.125.179)
  32. # [05:18] * Quits: adele (adele@24.7.125.179) (Quit: adele)
  33. # [05:38] * Joins: Zeros (Zeros-Elip@69.140.40.140)
  34. # [05:46] * Joins: adele (adele@24.7.125.179)
  35. # [05:56] * Joins: Navarr (navarr@75.53.206.209)
  36. # [06:06] * Quits: adele (adele@24.7.125.179) (Quit: adele)
  37. # [06:10] * Quits: marcos (marcos@131.181.148.227) (Quit: marcos)
  38. # [06:18] * Quits: mjs_ (mjs@17.203.15.230) (Ping timeout)
  39. # [06:18] * Joins: mjs (mjs@17.255.97.30)
  40. # [06:18] * Quits: Navarr (navarr@75.53.206.209) (Quit: Navarr)
  41. # [06:30] * Quits: mjs (mjs@17.255.97.30) (Quit: mjs)
  42. # [06:31] * Joins: adele (adele@24.7.125.179)
  43. # [06:34] * Quits: adele (adele@24.7.125.179) (Quit: adele)
  44. # [06:42] * Joins: heycam (cam@130.194.72.84)
  45. # [06:55] * Quits: heycam (cam@130.194.72.84) (Quit: bye)
  46. # [07:18] * Quits: gavin (gavin@99.253.193.147) (Ping timeout)
  47. # [07:23] * Joins: gavin (gavin@99.253.193.147)
  48. # [07:23] * Joins: mjs (mjs@24.5.43.151)
  49. # [08:28] * Joins: marcos (marcos@131.181.148.227)
  50. # [08:43] * Joins: heycam (cam@124.168.12.194)
  51. # [09:05] * Quits: mjs (mjs@24.5.43.151) (Quit: mjs)
  52. # [09:11] * Quits: Zeros (Zeros-Elip@69.140.40.140) (Quit: Leaving)
  53. # [09:13] * Joins: mjs (mjs@24.5.43.151)
  54. # [10:16] * Quits: Lachy (Lachlan@85.196.122.246) (Client exited)
  55. # [10:44] * Joins: Lachy (Lachlan@85.196.122.246)
  56. # [10:44] * Quits: Lachy (Lachlan@85.196.122.246) (Client exited)
  57. # [10:44] * Joins: Lachy (Lachlan@85.196.122.246)
  58. # [10:44] * Quits: Lachy (Lachlan@85.196.122.246) (Client exited)
  59. # [10:45] * Joins: Lachy (Lachlan@85.196.122.246)
  60. # [11:13] * Joins: zcorpan (zcorpan@88.131.66.80)
  61. # [11:18] * Joins: tlr (tlr@128.30.52.30)
  62. # [11:29] * Quits: mjs (mjs@24.5.43.151) (Quit: mjs)
  63. # [11:30] * Quits: Lachy (Lachlan@85.196.122.246) (Quit: This computer has gone to sleep)
  64. # [11:51] * Joins: Lachy (Lachlan@213.236.208.22)
  65. # [12:53] * Joins: myakura (myakura@118.8.102.216)
  66. # [13:12] * Quits: Philip (philip@80.177.163.133) (Ping timeout)
  67. # [13:30] <hsivonen> Julian: CURIEs take an anti-pattern and elaborate on it
  68. # [13:33] <Julian> hsivonen: that it is an anti-pattern is your opinion
  69. # [13:34] <Julian> as a matter of fact, the QName/CURIE pattern is not needed when the language directly allows extensibility, as in XHTML
  70. # [13:35] <hsivonen> Julian: is qnames-in-content an anti-pattern in your opinion?
  71. # [13:36] <hsivonen> Julian: have you suggested to the XHTML2 WG that they don't need CURIEs?
  72. # [13:36] <Julian> hsivonen: I try to avoid them. But sometimes it's hard not to use them, such as in XPath
  73. # [13:36] <Julian> hsivonen: where do they use CURIEs?
  74. # [13:37] <hsivonen> Julian: using XPath through a DOM API shows how qnames suck badly when removed from their XML parsing context
  75. # [13:37] <hsivonen> Julian: RDFa, IIRC
  76. # [13:37] <Julian> hsivonen: XPath -- but it would suck even more if you would need to expand the names in the XPath query
  77. # [13:38] <hsivonen> Julian: Selectors take an excellent an utterly devious approach and ignore the namespace URI in the easy case
  78. # [13:38] <Julian> hsivonen: RFDa wouldn't be needed if the language allowed arbitrary extensions as first-class citizens
  79. # [14:13] <gDashiva> So Julian, when you worry about clashing classnames, is the problem malicious or lazy authors?
  80. # [14:20] <Julian> lazy authors
  81. # [14:21] <zcorpan> the spec can require all sorts of things but that won't bite on lazy authors
  82. # [14:21] <hsivonen> Julian: can you imagine a random person out there putting "de.greenbytes" into a classname out of laziness?
  83. # [14:22] <zcorpan> hsivonen: yes if they copy it from somewhere else :)
  84. # [14:23] <hsivonen> zcorpan: can they also copy URIs?
  85. # [14:23] <zcorpan> perhaps they like the style that is attached to de.greenbytes but want to use it for something else
  86. # [14:23] <zcorpan> hsivonen: yes
  87. # [14:23] <gDashiva> I don't really see how lazy authors would care whether the classnames are URIs or not
  88. # [14:23] <gDashiva> Pasting is pasting
  89. # [14:24] <zcorpan> <span class="DONT_COPY_THIS_OR_I_WILL_SUE_YOU_greenbytes">
  90. # [14:25] <gDashiva> class="readonly:de.greenbytes"
  91. # [14:25] <Lachy> Julian, can you provide any significant evidence that demonstrates where any URI or DNS based naming scheme has resulted in a clash, even in a free form environment such as classnames?
  92. # [14:26] <Lachy> I find it hard to believe that clashing namespaces hasn't been a problem in practice, yet somehow something even more specific than a prefix will inevitably clash and cause huge problems when it occurs
  93. # [14:27] <Lachy> Julian, it seems to me that you're not really trying to solve real problems, but rather addressing imagined problems baesd on the hypothetical scenario of clashing names causing unspecified problems
  94. # [14:27] <Julian> Lachy: when prefixes do clash, you notice and rewrite them
  95. # [14:27] <Julian> when identifiers clash, you've lost
  96. # [14:28] <Julian> finally, there's an issue with coming out with a complete new naming system, when the rest of the world already uses an URI based system
  97. # [14:29] <krijnh> Does it? I don't :$
  98. # [14:29] <Lachy> but if you chose an identifier based on a URI or reverse DNS, then the chances of a clash are so slim given that a) few people have a domain similar to anyone elses, let alone one that can be reversed in a way that ends up looking like someone elses (i.e. the ab.cd vs. cd.ab example you gave on the list)
  99. # [14:30] <gDashiva> Julian: You can use the old naming system if you don't want to invent a new one, so there's no tax on you.
  100. # [14:30] <Lachy> b) such class names intended for consumption by your own tools, which clearly won't be run on other sites outside of your own system, and so clashes aren't a problem in practive
  101. # [14:30] <Lachy> *practice
  102. # [14:31] <Lachy> c) for any class names intended for wider use among the community, they really should be brought forth and *discussed* with the community that would be expected to make use of them
  103. # [14:31] <Julian> Lachy: the class names may be intended for public use. Yes, I know that's a use case you don't want to address.
  104. # [14:31] <Lachy> microformats is addressing that use case quite well already
  105. # [14:31] <Julian> Lachy: yes. But that's not necessarily the W3C or the WhatWG,
  106. # [14:32] <Lachy> so I really don't understand what use case you're trying to address
  107. # [14:32] <Julian> Lachy: no, Microformats only work for very generic stuff.
  108. # [14:32] <Julian> The process doesn't scale.
  109. # [14:32] <Lachy> in what way?
  110. # [14:32] <Julian> People have been sent away because their use case was not "generic" enough.
  111. # [14:33] <Julian> The point is it doesn't need to be generic.
  112. # [14:33] <Lachy> if some community wanted to collaborate on developing a vocabulary for some vertical market, then let the people involved in that market work together on such a vocabulary. I don't see how that doesn't scale
  113. # [14:33] <Julian> It's sufficient if it's of interest to a few selected parties that want to collaborate.
  114. # [14:33] <gDashiva> You can still use the microformat process even if you don't register it as a microformats sanctioned microformat
  115. # [14:33] <Julian> Exactly; but how do you deploy that vocabulary without potential name collisions?
  116. # [14:33] <hsivonen> gDashiva: rejecting stuff is part of the microformat Process
  117. # [14:34] <gDashiva> Julian: You look around for a collision, then deploy when you don't find one
  118. # [14:34] <hsivonen> Julian: you google for the names your are about to pick until you find something that gets no results
  119. # [14:34] <gDashiva> If you worry a lot, you create microformats-non-generic and register it there
  120. # [14:36] <Julian> gdashiva: why would I trade something that has some guarantee of reliability against what you just proposed?
  121. # [14:36] <gDashiva> Julian: Because that guarantee doesn't exist
  122. # [14:36] <Julian> also, it requires not only for me to google for other's names, but also for them to google for mine.
  123. # [14:36] <gDashiva> Julian: Nothing can save you from lazy authors
  124. # [14:37] <Julian> gdashiva: so why does it work then in practice?
  125. # [14:37] <Lachy> Julian, so what? Who cares if some random web site happens to use the same class name as you for an entirely different purpose? What *REAL PROBLEMS* does that create?
  126. # [14:37] <gDashiva> Julian: Because the collisions you worry about are mythical beasts
  127. # [14:38] <hsivonen> Julian: do you mean working like the XHTML2 WG asserting Authority over a namespace URI, but Apple, Mozilla and Opera implementing stuff specced elsewhere with that namespace URI?
  128. # [14:38] <Lachy> so far, there's been no explanation of the problem beyond the assumption that collisions are inevitable
  129. # [14:38] <Lachy> which I believe to be false, given the abillity to use a domain as a class name
  130. # [14:38] <Julian> hsivonen: that's a W3C process issue
  131. # [14:39] <hsivonen> Julian: perhaps, but it shows that your assumption about guarantee is untrue in a very conspicuous case
  132. # [14:39] * Quits: sryo (sryo@190.245.200.47) (Ping timeout)
  133. # [14:40] * Quits: zcorpan (zcorpan@88.131.66.80) (Ping timeout)
  134. # [14:40] <hsivonen> Julian: the point is that the need for compatibility overrides Authority
  135. # [14:40] * Quits: takkaria (takkaria@208.75.87.166) (Ping timeout)
  136. # [14:40] * Quits: Dashiva (noone@80.202.213.134) (Ping timeout)
  137. # [14:40] * Joins: sryo (sryo@190.245.200.47)
  138. # [14:41] * Joins: takkaria (takkaria@208.75.87.166)
  139. # [14:41] * Quits: gDashiva (magnusrk@195.18.164.170) (Ping timeout)
  140. # [14:41] * Joins: gDashiva (magnusrk@195.18.164.170)
  141. # [14:41] * Quits: myakura (myakura@118.8.102.216) (Quit: Leaving...)
  142. # [14:41] * Quits: gavin (gavin@99.253.193.147) (Ping timeout)
  143. # [14:41] * Quits: Julian (chatzilla@217.91.35.233) (Ping timeout)
  144. # [14:42] * Quits: jgraham (jgraham@74.53.238.210) (Ping timeout)
  145. # [14:42] * Joins: jgraham (jgraham@74.53.238.210)
  146. # [14:42] * Joins: gavin (gavin@99.253.193.147)
  147. # [14:43] * Joins: Dashiva (noone@80.202.213.134)
  148. # [14:43] <gDashiva> Lag ho
  149. # [14:45] * Quits: inimino (weechat@67.207.138.202) (Ping timeout)
  150. # [14:45] * Quits: deltab (deltab@82.36.30.34) (Ping timeout)
  151. # [14:45] * Joins: deltab (deltab@82.36.30.34)
  152. # [14:46] * Joins: inimino (weechat@67.207.138.202)
  153. # [14:47] * Joins: Julian (chatzilla@217.91.35.233)
  154. # [14:52] <Lachy> Julian, any chance of getting an answer to my question?
  155. # [14:53] <Julian> could you please repeat the question?
  156. # [14:54] <Lachy> <Lachy> Julian, so what? Who cares if some random web site happens to use the same class name as you for an entirely different purpose? What *REAL PROBLEMS* does that create?
  157. # [14:54] <Julian> That wouldn't create a real problem.
  158. # [14:55] <Lachy> then what is the problem with clashes?
  159. # [14:55] <Julian> Real problems occur when one wants to combine vocabularies that have name clashes.
  160. # [14:55] <Lachy> so these would be vocabularies designed for wider use, and thus would be ones that should be developed with community involvement?
  161. # [14:55] <Julian> That happens, for instance, when those names double as identifiers in RDF, and you're building a triple store.
  162. # [14:56] <Lachy> I don't care about RDF. we're not talking about RDF here
  163. # [14:56] <Julian> The problem is that when you start a new vocabulary, you do not always know in advance whether it will be a private one, or it ends up being useful for other people.
  164. # [14:56] <Julian> Lachy: yes, I know you don't care.
  165. # [14:57] <Lachy> so if it turns out to be useful for others as well, then you can bring your previously developed private vocabulary up within a relevant community as a starting point
  166. # [14:58] <Lachy> and if you had used org.example.foo (based on your own domain), and wanted others to use that vocabulary, than coming up with a name more appropriate for wider use would be a good thing to do
  167. # [15:05] <gDashiva> And if it's private, changing the name won't be a problem
  168. # [15:08] <Lachy> Julian, something that might help would be if some sort of design pattern emerged within the wider community, for expressing unique class names. Then once we see substantial evidence of the pattern working, we can spec it
  169. # [15:08] <Lachy> but so far, there isn't even substantial evidence showing that such a design pattern is needed or desired by the wider community
  170. # [15:10] <Lachy> so, if you, Sam and anyone else interested in developing such a pattern, would work together to design such a pattern, and encourage others to use it, then come back to the group and present the results, it would be very useful
  171. # [15:11] <Lachy> but I really think it's time for the extensibility thread to die on public-html at this stage, since there's been no new evidence presented for a while and it's just going round in circles
  172. # [15:16] * Joins: aaronlev (chatzilla@92.228.81.34)
  173. # [15:21] * Quits: marcos (marcos@131.181.148.227) (Quit: ...and I'm gone. )
  174. # [15:56] * Quits: Lachy (Lachlan@213.236.208.22) (Client exited)
  175. # [15:57] * Joins: Lachy (Lachlan@213.236.208.22)
  176. # [16:22] * Quits: aaronlev (chatzilla@92.228.81.34) (Ping timeout)
  177. # [17:06] * Joins: gsnedders (gsnedders@212.181.225.173)
  178. # [17:35] * Quits: Lachy (Lachlan@213.236.208.22) (Quit: This computer has gone to sleep)
  179. # [17:43] * Joins: adele (adele@24.7.125.179)
  180. # [17:54] * Joins: Lachy (Lachlan@85.196.122.246)
  181. # [17:56] * Joins: aaronlev (chatzilla@92.228.81.34)
  182. # [17:57] * Joins: aaronlev_ (chatzilla@92.228.81.34)
  183. # [18:00] * Quits: aaronlev (chatzilla@92.228.81.34) (Ping timeout)
  184. # [18:00] * aaronlev_ is now known as aaronlev
  185. # [18:02] * Quits: Julian (chatzilla@217.91.35.233) (Quit: ChatZilla 0.9.83 [Firefox 3.0.1/2008070208])
  186. # [18:16] * Quits: aaronlev (chatzilla@92.228.81.34) (Quit: ChatZilla 0.9.83 [Firefox 3.1a2pre/2008073103])
  187. # [18:16] * Joins: aaronlev (chatzilla@92.228.81.34)
  188. # [18:34] * Quits: DanC (connolly@128.30.52.30) (Client exited)
  189. # [18:36] * Joins: DanC (connolly@128.30.52.30)
  190. # [18:43] * Quits: ChrisWilson (cwilso@131.107.0.101) (Ping timeout)
  191. # [18:48] * Joins: ChrisWilson (cwilso@131.107.0.72)
  192. # [18:53] * Quits: adele (adele@24.7.125.179) (Quit: adele)
  193. # [19:08] * Joins: ROBOd (robod@89.122.216.38)
  194. # [19:09] * Quits: gsnedders (gsnedders@212.181.225.173) (Ping timeout)
  195. # [19:22] * Joins: timeless (timeless@65.75.195.122)
  196. # [19:24] * Joins: adele (adele@17.203.15.174)
  197. # [19:37] * Joins: aroben (aroben@71.58.56.76)
  198. # [19:43] * Quits: adele (adele@17.203.15.174) (Quit: adele)
  199. # [19:47] * Quits: ChrisWilson (cwilso@131.107.0.72) (Ping timeout)
  200. # [20:22] * Joins: adele (adele@17.203.15.174)
  201. # [21:05] * Joins: gsnedders (gsnedders@212.181.225.173)
  202. # [21:11] * tlr is now known as tlr-off
  203. # [21:13] * Quits: adele (adele@17.203.15.174) (Quit: adele)
  204. # [21:15] * Joins: adele (adele@17.203.14.172)
  205. # [21:15] * Joins: mjs (mjs@24.5.43.151)
  206. # [21:32] * Quits: adele (adele@17.203.14.172) (Quit: adele)
  207. # [21:35] * Joins: dbaron (dbaron@63.245.220.241)
  208. # [22:47] * Quits: mjs (mjs@24.5.43.151) (Quit: mjs)
  209. # [22:56] * Joins: ChrisWilson (cwilso@131.107.0.77)
  210. # [23:21] * Quits: tlr-off (tlr@128.30.52.30) (Client exited)
  211. # [23:21] * Quits: ROBOd (robod@89.122.216.38) (Quit: http://www.robodesign.ro )
  212. # [23:22] * Joins: tlr (tlr@128.30.52.30)
  213. # [23:34] * Quits: gsnedders (gsnedders@212.181.225.173) (Ping timeout)
  214. # [23:46] * Joins: mjs (mjs@17.203.15.230)
  215. # Session Close: Wed Aug 06 00:00:00 2008

The end :)