/irc-logs / w3c / #html-wg / 2008-11-18 / end

Options:

  1. # Session Start: Tue Nov 18 00:00:00 2008
  2. # Session Ident: #html-wg
  3. # [00:21] * Joins: sryo (sryo@190.245.198.94)
  4. # [00:55] * Quits: tH (Rob@129.11.83.58) (Quit: ChatZilla 0.9.83-rdmsoft [XULRunner 1.9.0.1/2008072406])
  5. # [01:02] * Quits: aroben_ (aroben@71.58.97.175) (Connection reset by peer)
  6. # [01:14] * Quits: MikeSmith (MikeSmith@mcclure.w3.org) (Ping timeout)
  7. # [01:18] * Joins: MikeSmith (MikeSmith@mcclure.w3.org)
  8. # [01:32] * Quits: MichaelC (Michael@128.30.52.30) (Quit: ChatZilla 0.9.84 [Firefox 3.0.4/2008102920])
  9. # [02:18] * Joins: Lionheart (robin@66.57.69.65)
  10. # [02:27] * Quits: Lionheart (robin@66.57.69.65) (Quit: Leaving.)
  11. # [02:27] * Joins: Lionheart (robin@66.57.69.65)
  12. # [02:32] * Quits: Lionheart (robin@66.57.69.65) (Ping timeout)
  13. # [02:32] * Joins: Lionheart (robin@66.57.69.65)
  14. # [02:33] * Parts: billmason (bmason@69.30.57.41)
  15. # [03:06] * Quits: MikeSmith (MikeSmith@mcclure.w3.org) (Ping timeout)
  16. # [03:15] * Quits: maddiin (mc@87.185.236.129) (Quit: maddiin)
  17. # [03:20] * Joins: MikeSmith (MikeSmith@mcclure.w3.org)
  18. # [03:40] * Quits: Sander (svl@86.87.68.167) (Quit: And back he spurred like a madman, shrieking a curse to the sky.)
  19. # [03:43] * Quits: MikeSmith (MikeSmith@mcclure.w3.org) (Quit: sex break)
  20. # [04:27] * Joins: MikeSmith (MikeSmith@mcclure.w3.org)
  21. # [04:28] <Hixie> MikeSmith, you might want to help roy as he would like the wg to change the name of the spec and it's not clear that he has a clear understanding of how best to go about that (or if it is possible)
  22. # [04:31] * Quits: adele_ (adele@17.203.15.224) (Quit: adele_)
  23. # [04:31] <MikeSmith> Hixie: I'm not sure Roy would welcome my help. Anyway, he's not the only person who's suggested that the spec should have a different name. But I've not heard anyone yet suggest a more appropriate one.
  24. # [04:32] <Hixie> I suggestd "Web5" on #whatwg earlier
  25. # [04:32] <Hixie> then we can merge in SVG, MathML, the DOM specs, HTTP, URI, etc
  26. # [04:32] <Hixie> and make it a truly big spec
  27. # [04:33] <MikeSmith> I do wish that Roy would lay off of taking pot shots at browser vendors. It's not particular helpful.
  28. # [04:33] <Hixie> his latest e-mails indicate he really doesn't understand the realities of the web at the moment
  29. # [04:33] <Hixie> (these people complaining about big specs have obviously never seen ooxml and co, btw)
  30. # [04:34] <MikeSmith> yeah, true, it's not that big relative to many other specs
  31. # [04:34] <shepazu> Hixie: you aren't trying hard enough, then... surely you could be more verbose?
  32. # [04:34] <MikeSmith> ooxml thing is ridiculous though
  33. # [04:34] <MikeSmith> 5000 pages or something
  34. # [04:35] <MikeSmith> anyway, I got to grab lunch
  35. # [04:35] <MikeSmith> bbiab
  36. # [04:35] <Hixie> ooxml is tiny compared to what it SHOULD be to actually define the languages it defines
  37. # [04:35] <Hixie> ridiculously tiny
  38. # [04:35] <Hixie> it's like what HTML4 is to HTML5
  39. # [04:36] <shepazu> Part 1: Fundamentals and Markup Language Reference (5550 pages), Part 2: Open Packaging Conventions (128 pages), Part 3: Markup Compatibility and Extensibility (40 pages), Part 4: Transitional Migration Features (1455 pages)
  40. # [04:37] <Hixie> should probably be ten times that if it was to be actually useful
  41. # [04:37] <shepazu> 7173 pages
  42. # [04:43] * Quits: gavin_ (gavin@99.253.193.147) (Ping timeout)
  43. # [04:48] * Joins: gavin_ (gavin@99.253.193.147)
  44. # [04:53] * Quits: dbaron (dbaron@63.245.220.241) (Quit: 8403864 bytes have been tenured, next gc will be global.)
  45. # [05:16] * Joins: adele (adele@24.7.125.179)
  46. # [05:17] * Joins: dbaron (dbaron@71.204.144.136)
  47. # [05:17] * MikeSmith returns from cafeteria
  48. # [05:17] <MikeSmith> beautiful day outside.. should be outside enjoying it while it lasts
  49. # [06:26] * DanC waves
  50. # [06:27] <DanC> tinkering with URI syntax. got ABNF parsing in javascript working well enough to read the URI spec. "use strict"
  51. # [06:27] <DanC> phpht
  52. # [06:27] <DanC> http://homer.w3.org/~connolly/projects/urlp/rev/747cc30bf331
  53. # [06:27] <pimpbot> Title: urlp: changeset 10:747cc30bf331 (at homer.w3.org)
  54. # [06:30] <DanC> not as self-documenting as it could be...
  55. # [06:35] * DanC wonders if the URI reference grammar is regular
  56. # [06:38] <DanC> ew... it seems to be ambiguous. abc: parses as relative, since pchar includes :
  57. # [06:39] <DanC> oops... no, that's 0<pchar>
  58. # [06:42] <MikeSmith> DanC: what's the planned product of the tinkering?
  59. # [06:43] <DanC> an editorial revision of the URL section that uses a different layering to achieve the same externally-visible result
  60. # [06:43] * Joins: Zeros (Zeros-Elip@129.2.175.74)
  61. # [06:45] <MikeSmith> DanC: cool
  62. # [06:47] <DanC> oh felgercarb. this rule uses the "prose-val" deely: path-empty = 0<pchar>
  63. # [06:47] * Quits: Zeros (Zeros-Elip@129.2.175.74) (Quit: Leaving)
  64. # [06:48] <DanC> surely that wasn't on purpose
  65. # [06:48] * DanC hopes to find a changelog for the URI spec...
  66. # [06:50] <DanC> roy mentions abnf2re in http://labs.apache.org/webarch/uri/rev-2002/issues.html
  67. # [06:50] <pimpbot> Title: Issues List for the URI specification revision (RFC 2396 to RFC 3986) (at labs.apache.org)
  68. # [06:59] * Quits: dbaron (dbaron@71.204.144.136) (Quit: g'night)
  69. # [07:07] <DanC> this looks promising... http://svn.apache.org/repos/asf/labs/webarch/tags/uri/draft-05/rev-2002/rfc2396bis.xml
  70. # [07:07] <pimpbot> Title: Uniform Resource Identifier (URI): Generic Syntax (at svn.apache.org)
  71. # [07:09] <DanC> 503091 fielding path-empty = 0&lt;pchar&gt;
  72. # [07:09] <DanC> r503091 | fielding | 2004-04-01 19:01:20 -0600 (Thu, 01 Apr 2004) | 2 lines
  73. # [07:09] <DanC> finish the last change and copy to collected bnf
  74. # [07:21] * Quits: rking3 (rking3@24.5.77.167) (Quit: rking3)
  75. # [07:27] * Quits: heycam (cam@130.194.72.84) (Quit: bye)
  76. # [08:30] * Joins: heycam (cam@210.84.56.87)
  77. # [08:31] <DanC> ok, now I think it's pretty self-documenting. http://homer.w3.org/~connolly/projects/urlp/raw-file/e9aa1c4d3dd6/tinker.html
  78. # [08:31] <pimpbot> Title: URI syntax tinkering (at homer.w3.org)
  79. # [08:39] <DanC> hmm... would need a copy of the RFC checked in to actually work. darn.
  80. # [08:52] <pimpbot> planet: Tor Arne Vestbø is a WebKit Reviewer <http://webkit.org/blog/300/tor-arne-vestb%c3%b8-is-a-webkit-reviewer/> ** Full Pass of Acid3 <http://webkit.org/blog/280/full-pass-of-acid-3/> ** Introducing SquirrelFish Extreme <http://webkit.org/blog/214/introducing-squirrelfish-extreme/>
  81. # [08:56] <DanC> ok, copy checked in. MikeSmith , wanna give it a try? http://homer.w3.org/~connolly/projects/urlp/raw-file/15debb758efb/tinker.html
  82. # [08:56] <pimpbot> Title: URI syntax tinkering (at homer.w3.org)
  83. # [08:56] * MikeSmith tries now
  84. # [08:59] <MikeSmith> DanC: it's producing output, but not sure what criteria one would use to evaluate it
  85. # [09:00] <DanC> well, all it does is parse the ABNF and dump concat as <ol> and alt as <ul> and such
  86. # [09:00] <DanC> so far
  87. # [09:01] <MikeSmith> OK
  88. # [09:02] <DanC> I hope the grammar is regular so I can compile it to a regex
  89. # [09:04] <MikeSmith> DanC: is there not a way to test that it's regular?
  90. # [09:05] <DanC> yes, but IIRC, it's pretty close to the compilation process
  91. # [09:05] <MikeSmith> OK
  92. # [09:05] <DanC> i.e. you look for loops
  93. # [09:05] <DanC> some loops you can factor out
  94. # [09:05] <Hixie> does anyone remember if we got an action item out of the http auth issue at the f2f?
  95. # [09:05] * DanC commits a little fix; notes http://homer.w3.org/~connolly/projects/urlp/raw-file/tip/tinker.html will always get the latest
  96. # [09:05] <pimpbot> Title: URI syntax tinkering (at homer.w3.org)
  97. # [09:06] <Hixie> or even a clear description of what we need to figure out to resolve the issue?
  98. # [09:06] * DanC reviews minutes of auth session http://www.w3.org/2008/10/23-html-wg-minutes.html#item08
  99. # [09:06] <pimpbot> Title: HTML WG face-to-face meeting -- 23 Oct 2008 (at www.w3.org)
  100. # [09:06] <MikeSmith> Hixie: no, I don't think there was any action, at least
  101. # [09:07] * MikeSmith re-read Julian's comment “IETF is waiting for W3C; it’s a user agent issue."
  102. # [09:08] <DanC> "No related actions" -- http://www.w3.org/html/wg/tracker/issues/13
  103. # [09:08] <pimpbot> Title: ISSUE-13 - HTML Weekly Tracker (at www.w3.org)
  104. # [09:08] <Hixie> pity
  105. # [09:08] <Hixie> i'm not sure what to do about it
  106. # [09:08] <Hixie> nobody has made a clear description of the problem as i understand it
  107. # [09:08] <Hixie> and so i can't really evaluate any of the proposals
  108. # [09:08] <MikeSmith> Yeah. I guess we'd need to ask Julian to take an action on it.
  109. # [09:09] <MikeSmith> I think he was away doing HTTPbis all of last week
  110. # [09:09] <DanC> it seems problematic that http servers use 200 ok responses to send auth challenge forms
  111. # [09:09] <Hixie> why?
  112. # [09:10] <Hixie> that's what julian said at the f2f but i didn't understand it then either
  113. # [09:11] <DanC> not sure about concrete consequences, but it sure feels wrong to have an auth challenge form sitting in a cache as though it's a successful response
  114. # [09:11] <Hixie> a lot feels wrong about html and the web
  115. # [09:13] <DanC> I said in June "before we make this a design issue, let's see if the WG agrees it's a requirement for HTML 5"; i.e. I can see leaving it out of our critical path
  116. # [09:13] <MikeSmith> yeah
  117. # [09:15] <MikeSmith> Jonas said at the f2f that it did not seem like a problem that we should try to be the ones to solve
  118. # [09:15] <Hixie> well the folder regarding this issue is one of the folders i have with the oldest feedback on html5
  119. # [09:15] <Hixie> so i should reply to it soon
  120. # [09:16] <Hixie> unless i hear otherwise, i'm likely to not make any changes based on the feedback
  121. # [09:16] <MikeSmith> OK, I'll ask Julian if he can try to write up a problem description
  122. # [09:17] <DanC> mnot says "When I last checked, all major browsers *do* display the 401 page *if* they
  123. # [09:17] <DanC> don't recognise the authentication scheme presented."
  124. # [09:17] <DanC> does HTML 5 say anything in that case?
  125. # [09:17] <Hixie> which case?
  126. # [09:17] <DanC> the case of a 401 response with an unknown auth scheme
  127. # [09:17] <DanC> (mnot's comment is https://bugzilla.mozilla.org/show_bug.cgi?id=271383#c9 )
  128. # [09:17] <pimpbot> Bug 271383: was not found.
  129. # [09:18] <Hixie> it basically says to do whatever http requires
  130. # [09:18] <Hixie> http://www.whatwg.org/specs/web-apps/current-work/#navigate
  131. # [09:18] <pimpbot> Title: HTML 5 (at www.whatwg.org)
  132. # [09:18] <DanC> works for me.
  133. # [09:24] * DanC discovers http://srp.stanford.edu/ via https://bugzilla.mozilla.org/show_bug.cgi?id=356855
  134. # [09:24] <pimpbot> Bug 356855: was not found.
  135. # [09:26] * Joins: aaronlev_ (chatzilla@85.180.226.50)
  136. # [09:26] <MikeSmith> hmm, do any browsers actually yet support CDATA sections in text/html?
  137. # [09:26] * aaronlev_ is now known as aaronlev
  138. # [09:32] * DanC realizes he's only barely awake any more; puts self to bed before doing any serious damage...
  139. # [09:35] <MikeSmith> DanC: お疲れさま
  140. # [09:36] <Philip> MikeSmith: Opera sort of understands the <![CDATA[...]]> syntax in text/html (and has done for ages), but it does pretty crazy/stupid things with it
  141. # [09:36] <MikeSmith> Philip: I see
  142. # [09:37] <MikeSmith> so I guess that plan is that it'll be something that they support if/when the move to implementing HTML5-compliant parsers
  143. # [09:50] <Hixie> only if we keep the foreign content stuff
  144. # [09:50] <Hixie> not sure where we go from here for that
  145. # [09:50] <Hixie> since the f2f didn't conclude with anything
  146. # [09:54] <pimpbot> planet: Learn How To Write HTML 5 <http://www.w3.org/QA/2008/11/html5-howto.html> ** Fronteers HTML5 Presentation <http://annevankesteren.nl/2008/09/fronteers-html5-video> ** Anne van Kesteren: Fronteers HTML5 Presentation <http://annevankesteren.nl/2008/09/fronteers-html5-video>
  147. # [10:11] <MikeSmith> Hixie: so CDATA sections are to be allowed only in foreign content?
  148. # [10:20] <Hixie> yes
  149. # [10:20] * Quits: sryo (sryo@190.245.198.94) (Ping timeout)
  150. # [10:47] * Quits: MikeSmith (MikeSmith@mcclure.w3.org) (Quit: sex break)
  151. # [11:06] * Joins: ROBOd (robod@89.122.216.38)
  152. # [11:47] * Joins: MikeSmith (MikeSmith@mcclure.w3.org)
  153. # [11:54] <pimpbot> planet: Tor Arne Vestbø is a WebKit Reviewer <http://webkit.org/blog/300/tor-arne-vestb%c3%b8-is-a-webkit-reviewer/> ** Full Pass of Acid3 <http://webkit.org/blog/280/full-pass-of-acid-3/> ** Introducing SquirrelFish Extreme <http://webkit.org/blog/214/introducing-squirrelfish-extreme/>
  154. # [12:00] <hsivonen> the page count comparison OOXML and ODF has always been a bad comparison. ODF needs much more precision.
  155. # [12:01] <hsivonen> that is, even though I politically like ODF, reading the spec makes me think that it needs Hixie treatment
  156. # [12:02] <hsivonen> OTOH, the worse parts of OOXML are those that don't specify things properly but say that the application needs to imitate a particular version of Word or WordPerfect
  157. # [12:09] <hsivonen> if the ODF&OOXML folks had followed HTML5 principles, they'd have thoroughly specced .doc, .xls and .ppt without XML.
  158. # [12:29] * Quits: adele (adele@24.7.125.179) (Ping timeout)
  159. # [12:55] <pimpbot> planet: Re: two thousand twenty two <http://annevankesteren.nl/2008/09/re-two-thousand-twenty-two> ** Fronteers HTML5 Presentation <http://annevankesteren.nl/2008/09/fronteers-html5-video> ** Anne van Kesteren: Fronteers HTML5 Presentation <http://annevankesteren.nl/2008/09/fronteers-html5-video>
  160. # [13:15] * Quits: MikeSmith (MikeSmith@mcclure.w3.org) (Quit: sex break)
  161. # [13:48] * Joins: maddiin (mc@87.185.242.54)
  162. # [14:01] * Joins: Sander (svl@86.87.68.167)
  163. # [14:04] * Quits: Lachy (Lachlan@85.196.122.246) (Quit: This computer has gone to sleep)
  164. # [14:20] * Joins: Lachy (Lachlan@213.236.208.22)
  165. # [14:21] * Quits: Lachy (Lachlan@213.236.208.22) (Quit: Leaving)
  166. # [14:21] * Joins: Lachy (Lachlan@213.236.208.22)
  167. # [14:33] * Joins: MichaelC (Michael@128.30.52.30)
  168. # [14:37] * Joins: myakura (myakura@122.17.190.200)
  169. # [14:53] * Joins: aaronlev_ (chatzilla@85.180.226.59)
  170. # [14:54] * Quits: aaronlev (chatzilla@85.180.226.50) (Ping timeout)
  171. # [14:54] * aaronlev_ is now known as aaronlev
  172. # [15:54] * Quits: maddiin (mc@87.185.242.54) (Quit: maddiin)
  173. # [15:54] * Joins: maddiin (mc@87.185.242.54)
  174. # [16:04] * Quits: myakura (myakura@122.17.190.200) (Quit: Leaving...)
  175. # [16:05] * Joins: MikeSmith (MikeSmith@mcclure.w3.org)
  176. # [16:09] * Quits: MikeSmith (MikeSmith@mcclure.w3.org) (Client exited)
  177. # [16:11] * Joins: MikeSmith (MikeSmith@mcclure.w3.org)
  178. # [16:39] * Joins: billmason (bmason@69.30.57.41)
  179. # [16:48] * Quits: Lionheart (robin@66.57.69.65) (Connection reset by peer)
  180. # [16:48] * Joins: aroben (aroben@71.58.97.175)
  181. # [16:53] * Joins: dbaron (dbaron@71.204.144.136)
  182. # [17:03] * Parts: billmason (bmason@69.30.57.41)
  183. # [17:07] * Joins: billmason (bmason@69.30.57.41)
  184. # [17:08] * Quits: @trackbot (trackbot@128.30.52.30) (Connection reset by peer)
  185. # [17:12] * Joins: trackbot (trackbot@128.30.52.30)
  186. # [17:33] * Quits: aaronlev (chatzilla@85.180.226.59) (Ping timeout)
  187. # [17:34] * Joins: tH (Rob@129.11.83.58)
  188. # [17:58] <pimpbot> planet: Tor Arne Vestbø is a WebKit Reviewer <http://webkit.org/blog/300/tor-arne-vestb%c3%b8-is-a-webkit-reviewer/> ** Full Pass of Acid3 <http://webkit.org/blog/280/full-pass-of-acid-3/> ** Introducing SquirrelFish Extreme <http://webkit.org/blog/214/introducing-squirrelfish-extreme/>
  189. # [18:07] <Philip> pimpbot, you've said that before, at least twice
  190. # [18:07] <pimpbot> Philip: Huh?
  191. # [18:08] <Philip> pimpbot, I said you've said that before, at least twice
  192. # [18:08] <pimpbot> Philip: Huh?
  193. # [18:08] <Philip> Stupid bot :-(
  194. # [18:08] <Dashiva> pimpbot is less than pimping
  195. # [18:08] <pimpbot> Dashiva: Huh?
  196. # [18:11] * Joins: ChrisWilson (cwilso@131.107.0.80)
  197. # [18:24] <Lachy> MikeSmith, yt?
  198. # [18:31] <MikeSmith> Lachy: yeah, for a bit
  199. # [18:31] <Lachy> MikeSmith, I have some issues with the new markup-spec, especially the fact that it attempts to be normative rather than informative.
  200. # [18:32] <MikeSmith> and?
  201. # [18:32] <Lachy> and, although I haven't had time to read the whole thing since I've been travelling so much lately, it seems to have some overlap with the authoring guide
  202. # [18:33] <Lachy> I don't see the purpose of making it normative at all, since it basically just redefines what's already in the HTML5 spec
  203. # [18:35] <Lachy> although, I suppose, for the authoring guide, I was going to explain the element content models in a more reader friendly manner, rather than using a formal grammar like you've done
  204. # [18:36] <MikeSmith> it's not normative at this point, since it doesn't have any official standing in the group
  205. # [18:37] <MikeSmith> neither does the draft of the authoring guide
  206. # [18:37] <Lachy> I said "*attempts* to be normative". The abstract clearly says "This specification normatively defines the fifth major version of the HTML vocabulary"
  207. # [18:38] <MikeSmith> well, it will continue to attempt to be normative
  208. # [18:38] <MikeSmith> as long as I am editing it
  209. # [18:38] <Lachy> but if it does get officially published, I really don't see why it should be normative
  210. # [18:39] <MikeSmith> whether or not it actually ever becomes normative will be ultimately be something for the group to decide
  211. # [18:39] <MikeSmith> I don't really see much point in having a discussion about whether it should be normative or not until then
  212. # [18:39] <Lachy> ok, well I will object to it ever being published as a normative document later then.
  213. # [18:40] <MikeSmith> noted
  214. # [18:40] * Philip supposes the normativity is sort of bike-shedding, when the more important thing is the existence and usefulness of the document
  215. # [18:41] <DanC> MikeSmith, as Lachy is a member of the WG, you might practice trying to convince him, rather than just stiff-arming him
  216. # [18:42] <DanC> yeah, I don't have a strong feeling on normativity either, Philip
  217. # [18:43] <Lachy> Philip, because the normativeness of the document very much affects the purpose and usefulness of the document
  218. # [18:45] <Philip> Lachy: Would you be happier if it stayed exactly the same but removed the "this document is normative" and "some keywords are to be interpreted as defined in RFC2119" sentences? Or is the real concern with the level of detail and target audience and suchlike?
  219. # [18:45] <Lachy> because by attempting to be normative, then it's really just unnecssarily dupicating the HTML5 spec. Whereas if it's informative, then it's purpose is more clearly meant to explain the HTML5 spec
  220. # [18:47] <Philip> It's not necessarily unnecessary since it could be temporary duplication, pending removal of the equivalent text from the HTML5 spec if some experimentation and evaluation indicates that that's going to be a reasonable thing to do
  221. # [18:47] <Lachy> Philip, mildly. But then I would still have concerns about using a complicated, formal grammar to to informatively explain things
  222. # [18:50] <Lachy> In either case, the way it's written can be improved. But how exactly is very much dependent upon the intended audience and purpose of the document, which isn't very clear
  223. # [18:55] * Quits: Lachy (Lachlan@213.236.208.22) (Quit: This computer has gone to sleep)
  224. # [18:58] <pimpbot> planet: Re: two thousand twenty two <http://annevankesteren.nl/2008/09/re-two-thousand-twenty-two> ** Fronteers HTML5 Presentation <http://annevankesteren.nl/2008/09/fronteers-html5-video> ** Anne van Kesteren: Fronteers HTML5 Presentation <http://annevankesteren.nl/2008/09/fronteers-html5-video>
  225. # [19:01] <Dashiva> Why is pimpbot linking things from september
  226. # [19:14] <gsnedders> Dashiva: Because it's pimpin'.
  227. # [19:16] <Philip> Someone needs to write a pimplbot, obviously in C++
  228. # [19:17] <gsnedders> Philip: No, hoebot.
  229. # [19:17] * Joins: Lachy (Lachlan@85.196.122.246)
  230. # [19:19] <Philip> gsnedders: No, backhoebot
  231. # [19:22] <Sander> /ignore pimpbot
  232. # [19:24] <Philip> Sander: That would remove the opportunity to moan about it, which would be no fun
  233. # [19:25] <Sander> I'll just take my cue from the moaning of others. :)
  234. # [19:44] * Joins: adele (adele@17.203.15.224)
  235. # [19:51] * Quits: dbaron (dbaron@71.204.144.136) (Quit: 8403864 bytes have been tenured, next gc will be global.)
  236. # [20:04] * Quits: adele (adele@17.203.15.224) (Client exited)
  237. # [20:05] * Joins: adele (adele@17.203.15.146)
  238. # [20:08] * Joins: dbaron (dbaron@63.245.220.241)
  239. # [20:18] * Quits: adele (adele@17.203.15.146) (Quit: adele)
  240. # [20:22] * Joins: adele (adele@17.203.15.146)
  241. # [20:30] * Quits: adele (adele@17.203.15.146) (Quit: adele)
  242. # [20:32] * Joins: rking3 (rking3@24.5.77.167)
  243. # [20:32] * Joins: adele (adele@17.203.15.146)
  244. # [20:37] * Joins: Julian (chatzilla@80.143.192.242)
  245. # [20:39] <Julian> Looking at <http://krijnhoetmer.nl/irc-logs/html-wg/20081118#l-110>
  246. # [20:39] <pimpbot> Title: IRC logs: w3c / #html-wg / 20081118 (at krijnhoetmer.nl)
  247. # [20:39] <Julian> # # [09:09] <DanC> it seems problematic that http servers use 200 ok responses to send auth challenge forms
  248. # [20:39] <Julian> # # [09:09] <Hixie> why?
  249. # [20:39] <Julian> That's a joke, right?
  250. # [20:39] <DanC> no, I don't think so
  251. # [20:39] <Julian> If the server responds with 200, the client is supposed to assume that it got a representation of the resouce...
  252. # [20:40] <Julian> ...not a login form.
  253. # [20:40] <Julian> Sending authentication forms with status 200 essentially breaks all programmatic clients (those that do not have a human to assist understanding the content)
  254. # [20:41] * Quits: rking3 (rking3@24.5.77.167) (Quit: rking3)
  255. # [20:41] * Joins: rking3 (rking3@24.5.77.167)
  256. # [20:47] <Julian> I've seen a web spider overwriting his whole index with login forms, because it was sent with 200 and it didn't notice...
  257. # [21:00] <DanC> ok, that spider story is more concrete than anything I'd come up with
  258. # [21:01] <Julian> I've also seen WebDAV clients download pages containing login forms (PROPFIND did HTTP auth, but GET didn't)
  259. # [21:08] <Julian> I've also seen HTTP authentication hardwired for certain UAs (such as MS Webfolder), causing "new" Webdav clients to break.
  260. # [21:08] <Julian> It's a mess, and it's not HTTP's fault.
  261. # [21:12] * Quits: dbaron (dbaron@63.245.220.241) (Quit: 8403864 bytes have been tenured, next gc will be global.)
  262. # [21:12] * Joins: dbaron (dbaron@63.245.220.241)
  263. # [21:15] * Joins: aaronlev_ (chatzilla@85.180.226.59)
  264. # [21:15] * aaronlev_ is now known as aaronlev
  265. # [21:17] * Quits: Julian (chatzilla@80.143.192.242) (Ping timeout)
  266. # [21:24] * Quits: hsivonen (hsivonen@130.233.41.50) (Ping timeout)
  267. # [21:27] * Joins: hsivonen (hsivonen@130.233.41.50)
  268. # [21:32] <gsnedders> MikeSmith: You're language doc breaks ISO 2145
  269. # [21:32] <gsnedders> *Your
  270. # [21:33] * Quits: aaronlev (chatzilla@85.180.226.59) (Quit: ChatZilla 0.9.83-rdmsoft [XULRunner 1.9.0.1/2008072406])
  271. # [21:46] <Hixie> julian: if you go to a page that contains useful information as well as a login form, why isn't 200 OK correct? the login form isn't the point of the page.
  272. # [21:48] <Hixie> julian: furthermore, a spider wouldn't be able to login anyway, unless it knew that it had credentials, in which case it could just login regardless (and wouldn't try not doing so)
  273. # [22:17] * Quits: ROBOd (robod@89.122.216.38) (Quit: http://www.robodesign.ro )
  274. # [22:24] * Quits: ChrisWilson (cwilso@131.107.0.80) (Ping timeout)
  275. # [22:37] * Quits: heycam (cam@210.84.56.87) (Quit: bye)
  276. # [22:47] * Quits: MichaelC (Michael@128.30.52.30) (Ping timeout)
  277. # [23:17] * Joins: Schenker80 (Schenker80@86.165.88.15)
  278. # [23:21] * Quits: Schenker80 (Schenker80@86.165.88.15) (Quit: Leaving)
  279. # Session Close: Wed Nov 19 00:00:00 2008

The end :)