/irc-logs / w3c / #html-wg / 2007-04-14 / end

Options:

  1. # Session Start: Sat Apr 14 00:00:00 2007
  2. # Session Ident: #html-wg
  3. # [01:30] * Disconnected
  4. # [01:30] * Attempting to rejoin channel #html-wg
  5. # [01:30] * Rejoined channel #html-wg
  6. # [01:30] * Topic is 'W3C HTML WG http://www.w3.org/html/wg/ - http://krijnhoetmer.nl/irc-logs/ (logged) - http://esw.w3.org/topic/HTML/ProposedDesignPrinciples'
  7. # [01:30] * Set by anne on Tue Mar 27 12:28:46
  8. # [01:34] * Quits: foca (foca@190.64.4.27) (Quit: foca)
  9. # [01:54] * Quits: mw22 (chatzilla@63.245.220.228) (Ping timeout)
  10. # [02:01] * Parts: hasather (hasather@81.235.209.174)
  11. # [02:01] * Parts: hasather_ (hasather@81.235.209.174)
  12. # [02:37] * Joins: Shunsuke (kuruma@219.110.80.235)
  13. # [02:51] * Quits: marcos (chatzilla@203.206.31.102) (Ping timeout)
  14. # [02:55] * Quits: loic (loic@90.27.88.46) (Ping timeout)
  15. # [03:15] * Joins: gavin_ (gavin@74.103.208.221)
  16. # [03:27] * Quits: adele (adele@17.255.100.139) (Quit: adele)
  17. # [03:33] * Quits: mjs (mjs@17.255.98.24) (Quit: mjs)
  18. # [03:35] * Joins: mjs (mjs@17.255.98.24)
  19. # [03:38] * Quits: mjs (mjs@17.255.98.24) (Ping timeout)
  20. # [03:49] * Joins: marcos___ (chatzilla@203.206.31.102)
  21. # [03:49] * marcos___ is now known as marcos
  22. # [03:54] * Joins: adele (adele@67.170.236.225)
  23. # [03:54] * Quits: adele (adele@67.170.236.225) (Quit: adele)
  24. # [03:55] * Joins: Shunsuke_ (kuruma@219.110.80.235)
  25. # [03:57] * Quits: Shunsuke (kuruma@219.110.80.235) (Ping timeout)
  26. # [04:27] * Joins: marcos__ (chatzilla@203.206.31.102)
  27. # [04:27] * marcos__ is now known as marcos_
  28. # [04:31] * Joins: mw22_ (chatzilla@209.79.152.188)
  29. # [04:31] * mw22_ is now known as mw22
  30. # [04:43] * Quits: marcos_ (chatzilla@203.206.31.102) (Ping timeout)
  31. # [04:49] * Joins: marcos__ (chatzilla@203.206.31.102)
  32. # [04:49] * marcos__ is now known as marcos_
  33. # [04:53] * Joins: myakura (myakura@60.239.122.32)
  34. # [04:59] * Lachy writes another rant
  35. # [05:17] * Quits: gavin_ (gavin@74.103.208.221) (Ping timeout)
  36. # [05:18] * Quits: marcos_ (chatzilla@203.206.31.102) (Ping timeout)
  37. # [05:22] * Joins: gavin_ (gavin@74.103.208.221)
  38. # [05:23] * Joins: Zeros (Zeros-Elip@69.140.48.129)
  39. # [05:28] * Quits: dbaron (dbaron@63.245.220.242) (Quit: 8403864 bytes have been tenured, next gc will be global.)
  40. # [05:39] * Joins: marcos__ (chatzilla@203.206.31.102)
  41. # [05:40] * marcos__ is now known as marcos_
  42. # [06:05] * Quits: marcos_ (chatzilla@203.206.31.102) (Ping timeout)
  43. # [06:07] * Quits: myakura (myakura@60.239.122.32) (Quit: Leaving...)
  44. # [07:21] * Quits: Zeros (Zeros-Elip@69.140.48.129) (Quit: Leaving)
  45. # [07:23] * Quits: gavin_ (gavin@74.103.208.221) (Ping timeout)
  46. # [07:24] * Joins: Zeros (Zeros-Elip@67.154.87.254)
  47. # [07:29] * Joins: gavin_ (gavin@74.103.208.221)
  48. # [08:11] * Joins: anne (annevk@81.68.67.12)
  49. # [08:11] <anne> morning
  50. # [08:13] <Lachy> hi anne
  51. # [08:14] <anne> flight leaves in four hours or so
  52. # [08:14] <Lachy> how long is the flight?
  53. # [08:14] <anne> after which I'll be on nice 26 hour trip
  54. # [08:14] <anne> on a*
  55. # [08:15] <anne> 12 hours to Kuala Lumpur and another 10 to Brisbane with a stop at KL
  56. # [08:17] <Hixie> christ, that meeting better be productive to be worth that much travelling :-P
  57. # [08:18] <anne> i hope so
  58. # [08:18] <Lachy> regardless of whether it's productive or not, it should be fun
  59. # [08:20] <Lachy> aargh! Mike Schinkel responded to me off-list about my latest rant :-/
  60. # [08:29] <anne> he also e-mailed another twenty e-mails or something which I deleted
  61. # [08:29] <anne> oh, more!
  62. # [08:29] <anne> +1 ++1+!+!!+1
  63. # [08:29] <Hixie> +pi!
  64. # [08:29] <anne> <indent>+!
  65. # [08:29] <Lachy> yeah, I didn't read most of those either
  66. # [08:32] <anne> "Is there any reason note to standardize the switch syntax so it's not some IE specific thing, e.g."
  67. # [08:32] <anne> some bad crack that guy is on :)
  68. # [08:32] * Quits: sbuluf (xrzy@200.49.140.20) (Ping timeout)
  69. # [08:33] <Lachy> yay! Let's standardise browser sniffing and require authors to write specific switches for every browser
  70. # [08:33] <Hixie> IE is going to have an opt-in
  71. # [08:33] <anne> despite what you think, this will make life easier
  72. # [08:34] <anne> i'm interested in their whole story
  73. # [08:34] <anne> for instance, an answer to my e-mail would be great
  74. # [08:34] <Hixie> the best we can hope for is to guide them to designing it in a safe way that screws other browsers the least and isn't condoned by the spec
  75. # [08:35] <anne> he has a point about browser sniffing though
  76. # [08:35] <anne> we had some cases where we support the same set of features as Firefox a given page uses, but we still break it
  77. # [08:36] <Lachy> there's no way the web developers in the web standards community are going to accept a UA-specific switch
  78. # [08:36] <Hixie> good
  79. # [08:36] <anne> this is mostly because authors are not always informed enough to make the right descision
  80. # [08:36] <Hixie> that would kill it then
  81. # [08:36] <anne> for IE this problem is likely far worse
  82. # [08:37] <Lachy> I'd like to see the response if he published the nonsense he's currently writing on public-html, on the IE blog
  83. # [08:37] <anne> as all "web standards pages" are authored with IE specific switches
  84. # [08:37] <anne> either simple UA sniffing or other stuff
  85. # [08:37] <anne> dropping it all at once and arriving at the same featureset as Firefox seems like a very hard and impossible task to do
  86. # [08:37] <Hixie> the response from most web devs would be uninformed
  87. # [08:38] <Lachy> the thing Chris fails to realise is that CSS Hacks (the good ones, at least) were designed in such a way that when the browser because more standards compliant, they would no longer be applied
  88. # [08:38] <anne> much like it is on digg.com /. etc.
  89. # [08:38] <Hixie> it's the same people who simlultaneously wanted xhtml2, backwards compatibility, text/html, no browser sniffing, and the browsers not to break their pages when they upgrades
  90. # [08:38] <Lachy> they broke that when they fixed things poorly in IE7
  91. # [08:38] <Hixie> anyway
  92. # [08:38] <Hixie> i don't know where to go from here
  93. # [08:39] <anne> yeah, IE7 was just guarenteed to break stuff
  94. # [08:39] <anne> the CSS compat issue bite them and now they thing it applies to everything
  95. # [08:39] <anne> virtually every web dev i have spoken too wants their DOM to be a tree like Firefox for instance
  96. # [08:40] <anne> they even write their own fricking HTML tokenizers / parsers in javascript purely because IE generates a incest DOM
  97. # [08:40] <Lachy> Chris needs to understand that it is unrealistic to aim for 0 incompatibilities with every single browser upgrade
  98. # [08:43] <anne> it's realistic
  99. # [08:43] <anne> they can do it
  100. # [08:43] <anne> apparently market leaders can quite easily destroy stuff
  101. # [08:43] <anne> this was never really clear to me before
  102. # [08:44] <Lachy> they claim they can do it by not making any changes and requiring an opt-in. But that isn't the way to make progress
  103. # [08:46] <anne> Hixie, you took back Window?
  104. # [08:46] <Lachy> oh, gosh, Mike says just: "hehe!" and "ROTFLMFAO!!! :-) " - that's really pointless
  105. # [08:50] <anne> my 20 above was joking but he's already set 16
  106. # [08:50] <anne> today
  107. # [08:57] <Lachy> I responded to Mike's off-list mail and cc'd www-archive
  108. # [08:58] <Lachy> http://lists.w3.org/Archives/Public/www-archive/2007Apr/0026.html
  109. # [09:00] <anne> from markp: "I look at XHTML 2 and all I see is politics
  110. # [09:00] <anne> I look at HTML 5 and all I see is passion"
  111. # [09:00] <anne> he just published http://diveintomark.org/public/2005/12/whitelights/
  112. # [09:01] <anne> anyway, I should get moving
  113. # [09:01] <anne> ttyal
  114. # [09:04] * Quits: anne (annevk@81.68.67.12) (Ping timeout)
  115. # [09:31] * Joins: marcos__ (chatzilla@203.206.31.102)
  116. # [09:31] * marcos__ is now known as marcos_
  117. # [09:31] * Quits: gavin_ (gavin@74.103.208.221) (Ping timeout)
  118. # [09:36] * Joins: gavin_ (gavin@74.103.208.221)
  119. # [09:50] * Quits: beowulf (carisenda@91.84.50.132) (Ping timeout)
  120. # [09:53] * Joins: myakura (myakura@60.239.122.32)
  121. # [09:53] * Joins: beowulf (carisenda@91.84.50.132)
  122. # [10:03] * Quits: Zeros (Zeros-Elip@67.154.87.254) (Ping timeout)
  123. # [10:19] * Joins: loic (loic@90.27.88.46)
  124. # [10:27] * Joins: edas (edaspet@88.191.34.123)
  125. # [10:32] <Lachy> I'm going to try and write an article about the versioning debate and IE standards-mode++ switches
  126. # [10:32] <Lachy> anyone like to help?
  127. # [10:56] * Joins: ROBOd (robod@86.34.246.154)
  128. # [11:27] * Quits: ROBOd (robod@86.34.246.154) (Quit: http://www.robodesign.ro )
  129. # [11:38] * Quits: gavin_ (gavin@74.103.208.221) (Ping timeout)
  130. # [11:41] * Joins: zcorpan (zcorpan@84.216.42.156)
  131. # [11:43] * Joins: gavin_ (gavin@74.103.208.221)
  132. # [11:48] * Joins: claudio (claudioc@89.97.35.74)
  133. # [12:14] <krijnh> Lachy: In what way?
  134. # [12:15] <Lachy> krijnh, not sure, any suggestions?
  135. # [12:15] <krijnh> Have to read up on the mailing list first I think
  136. # [12:16] <krijnh> Only 50 mails
  137. # [12:16] <Lachy> I have to review it before I start, find some that mention the key points
  138. # [12:19] * krijnh skips the <indent> ones
  139. # [12:19] * Lachy too
  140. # [12:20] * zcorpan 3
  141. # [12:20] * myakura already thrown them out :p
  142. # [12:24] <zcorpan> Lachy: could you reply to my <button type> email on list-whatwg with the suggestion you made on public-html
  143. # [12:24] <zcorpan> ?
  144. # [12:25] <Lachy> I already sent that to Hixie off-list last night (in response to his off-list mail to me), but I could if you like
  145. # [12:25] <zcorpan> i just thought so that it wouldn't be lost when hixie looks at my email in a few months or something
  146. # [12:26] <Lachy> he's quite good at keeping track of all mail on a particular topic, he keeps folders for each one
  147. # [12:26] <zcorpan> ok
  148. # [12:26] <Lachy> though it would be good to have the suggestion publicly archived anyway
  149. # [12:28] <krijnh> "Is there any reason note to standardize the switch syntax so it's not some IE specific thing, e.g."
  150. # [12:28] <krijnh> :|
  151. # [12:28] <krijnh> <!--[opera 10]-->
  152. # [12:28] <krijnh> Cool stuff!
  153. # [12:28] <Lachy> if IE introduces a version switch of some sort, then I'm just going to do this...
  154. # [12:29] <claudio> "version switch"?
  155. # [12:29] <Lachy> e.g. if versioning is accepted like <!DOCTYPE html "5"> (or whatever syntax)
  156. # [12:29] <claudio> ah, that way
  157. # [12:29] <Lachy> I'm just going to use <!DOCTYPE html "[any large, random number > 5]">
  158. # [12:30] <Lachy> that way, my documents will always trigger the latest standards mode
  159. # [12:30] <claudio> this is in the topic of the LONG Chris email, IIRC
  160. # [12:30] <claudio> *was
  161. # [12:30] <Lachy> yes
  162. # [12:30] <krijnh> <!DOCTYPE html "42"> !
  163. # [12:30] * Quits: Ashe (Ashe@213.47.199.86) (Ping timeout)
  164. # [12:30] <claudio> ;)
  165. # [12:30] <edas> anyone with some time to help me understand why http://eric.daspet.name/htmlwg/opt-in.txt may not be a good idea ?
  166. # [12:30] <Lachy> I was thinking mroe like "9999999"
  167. # [12:31] <Lachy> but 42 should work for the next 100 years or so
  168. # [12:32] <Lachy> but it would have to be random, so MS couldn't just treat a commonly used value as equivalent to "5"
  169. # [12:33] * Joins: Ashe (Ashe@213.47.199.86)
  170. # [12:33] <zcorpan> Lachy: this would indeed be a problem for ms versioning idea. if lots of content hacked around the versioning mechanism they would have to invent a *new versioning mechanism* each time
  171. # [12:33] <beowulf> generally i feel way out of my depth on this wg, but the indent thread is helping a lot
  172. # [12:33] <Lachy> ha!
  173. # [12:33] <myakura> HTML 4.2 would solve 1/10 of all problems :p
  174. # [12:33] <zcorpan> this time it's a doctype. next time an attribute. then a comment. what next?
  175. # [12:33] <krijnh> zcorpan: A mime type
  176. # [12:34] <zcorpan> krijnh: yes, why not
  177. # [12:34] <Lachy> web developers are going to fight back if they go ahead with this versioning, I'd love to see them try to completely change it every time
  178. # [12:34] <krijnh> text/html+improved+compliance
  179. # [12:34] <zcorpan> should we point it out on the list?
  180. # [12:35] <Lachy> I think MS should accept text/vnd.ms-html+illformed-xml
  181. # [12:35] <krijnh> zcorpan: Yeah, do, Mike Schinkel might like it :)
  182. # [12:36] <zcorpan> -_-
  183. # [12:36] <Lachy> no, don't give him somthing else to respond to!!!! :-/
  184. # [12:36] <krijnh> Okay, that's nasty, if I replied to the list I'd probably make the same mistakes :p
  185. # [12:36] <claudio> the problem is, I suppose, HTML5 try to get rid of DTD/DOCTYPE, but then proposely use the DTD/DOCTYPE for versioning...
  186. # [12:37] <edas> zcorpan, the opt-in mecanism can be cumulative, is to say "just a number". There is no need for a new mecanism each time. There is just a need to increment a number
  187. # [12:37] * zcorpan actually wrote an email to mike on the <indent> thread. thinking about it again i probably shouldn't send it
  188. # [12:37] <Lachy> no, don't send it, just ignore the thread
  189. # [12:37] <zcorpan> edas: the point was that people would use "99999" as the number to hack around the versioning mechanism
  190. # [12:37] <Lachy> or send it off-list, but cc www-archive
  191. # [12:37] <zcorpan> edas: so they would break content anyway
  192. # [12:38] <zcorpan> Lachy: he will just reply on-list
  193. # [12:38] <krijnh> What if MS just says <!--[if IE]> is the same as if lte IE7
  194. # [12:38] <zcorpan> krijnh: cwilso didn't reply to that
  195. # [12:38] <Lachy> that was suggested on the klist already
  196. # [12:38] * zcorpan suggested it
  197. # [12:38] <krijnh> Yeah it was, I'm just discussing it here :)
  198. # [12:39] <zcorpan> ask him again?
  199. # [12:39] * zcorpan thinks it should be workable
  200. # [12:39] <edas> zcorpan, may be, but people will always be able to shoot themself in the foot. We will never be able to forbid that to them
  201. # [12:39] <edas> whatever we use
  202. # [12:39] <Lachy> that would be good if IE did that, it would solve the problem they created when they told everyone to use conditional comments without a version numberr
  203. # [12:39] <zcorpan> edas: as i said, ms could invent a new versioning mechanism each time
  204. # [12:39] <zcorpan> edas: i'm not saying it's a good idea to have versioning
  205. # [12:39] <Lachy> conditional comments should either be unsupported in IE8 standards mode, or [if IE] without a number ignored
  206. # [12:40] <zcorpan> Lachy: yeah i was left pondering when they suggested people stop using * html and instead use [if IE]
  207. # [12:40] <zcorpan> * html doesn't break anything. [if IE] does.
  208. # [12:41] <Lachy> yes
  209. # [12:41] <krijnh> IE should be named IE.next
  210. # [12:41] <krijnh> if IE would be false
  211. # [12:41] <Lachy> that's what I kept trying to argue when everyone kept arguing for cond comments
  212. # [12:41] <krijnh> if IE.next would be true
  213. # [12:41] <edas> zcorpan, a new mecanism each time would be so painful that i doubt even ms will take this path only for some people who shoot themself in the foot explicitly and conscienciously. They only care about big companies compatibility, and those won't lie about a date or a version they "support"
  214. # [12:41] <krijnh> Lachy: Or unsupported indeed
  215. # [12:42] <Lachy> personally, * html is one of the few hacks I use regularly because I know it targets IE < 6
  216. # [12:42] <Lachy> IE6 only
  217. # [12:42] <zcorpan> edas: what's the difference from css hacks? surely if 3% use it, they don't have a choise
  218. # [12:42] <zcorpan> people using css hacks *are* shooting themselves in the foot
  219. # [12:43] <zcorpan> that's why they want another version now
  220. # [12:44] <edas> zcorpan, people who use hack don't ask "I want the new feature I don't even know about"
  221. # [12:44] <edas> css hack is like using a mecanism/version they already know about
  222. # [12:44] <zcorpan> what's your point?
  223. # [12:45] <edas> that *if* we want versionning, a simple date in an attribute or in doctype would do the trick
  224. # [12:45] <hsivonen> "P.S. As an aside, I happen to believe that CSS is one of the worst poxes ever to be hoisted on the web" -- Mike Schinkel
  225. # [12:45] <hsivonen> wow
  226. # [12:46] <beowulf> *sigh*
  227. # [12:46] <krijnh> What is that guy talking about?
  228. # [12:46] <edas> the only drawback is what you told : people who will use 2037 as date even in 2007. But I don't think there will be many of them, and probably none of them in the "big companies" or people who really need compatibility
  229. # [12:46] <Lachy> http://www.w3.org/mid/46207535.1030909@mikeschinkel.com
  230. # [12:46] <beowulf> that would summarise most of the response i read on /. too
  231. # [12:46] <Lachy> below his sig
  232. # [12:47] <krijnh> I think he's a bot
  233. # [12:47] <beowulf> :)
  234. # [12:47] <krijnh> Same as http://krijnhoetmer.nl/zooi/irc/qnet/sillybot.txt
  235. # [12:47] <zcorpan> edas: say we have <html tested="date"> as the mechanism. 3% of the web use <html tested="9000-01-01"> and would break in the next ie should they use the same versioning mechanism. what will ms do?
  236. # [12:48] <edas> who are thos 3% and why do they use 9000-01-01 ?
  237. # [12:48] <edas> this is an important question
  238. # [12:48] <zcorpan> same as the sites that today break because of [insert whatever here]
  239. # [12:48] <Lachy> edas, so that we get full standards mode in all future versions of IE
  240. # [12:48] <edas> MS care about big companies, I don't even see why those want to risk putting a date=9000
  241. # [12:48] <krijnh> Perhaps it's mostly about IE only intranet applications
  242. # [12:48] <krijnh> They just shouldn't upgrade to IE.next then :]
  243. # [12:48] <Lachy> instead of being locked into the state of bugs when IE decides to add a new version
  244. # [12:49] <zcorpan> edas: you seeing why they would do so is not the point, just pretend that they will
  245. # [12:49] <edas> Lachy, in doing so they *know* they are cheating and they will break their pages. Do you think many of them will do that ? (geek excepted)
  246. # [12:49] <Lachy> yes, we, the web standards community, will encourage authors to do it
  247. # [12:49] <edas> zcorpan, I do not think they will, that is the point
  248. # [12:49] <zcorpan> what if they do?
  249. # [12:49] <zcorpan> what will ms do?
  250. # [12:49] <edas> Lachy, we ? are we stupid ? ;)
  251. # [12:50] <Lachy> edas, what?
  252. # [12:50] <edas> Lachy, what would be the point in using a date in future ? Your document won't be "more correct", and won't display more gracefully
  253. # [12:51] <edas> so why do you think standard-people will recommande putting a date=9000 ?
  254. # [12:51] <zcorpan> edas: is what authors do ever logical?
  255. # [12:51] <edas> zcorpan, well point taken ;)
  256. # [12:51] <Lachy> because when I write a web page, I don't make it 100% bug free in IE. I expect that future versions of IE will make up for those limitatison
  257. # [12:52] <edas> Lachy, ok, point taken too
  258. # [12:52] <edas> I shut up
  259. # [12:52] <edas> well, so I do not see any good solution
  260. # [12:52] <Lachy> edas, see lachy.id.au in IE for instance
  261. # [12:52] <Lachy> I don't want IE7 rendering to apply to that forever
  262. # [12:52] <zcorpan> edas: my proposed solution is to not have versioning at all
  263. # [12:53] <edas> zcorpan, MS will do versionning, despite of what we say
  264. # [12:53] <zcorpan> tell me something i don't know :)
  265. # [12:53] <Lachy> edas, yes, and we will make fight back and make it harder for them to work with it
  266. # [12:53] <Lachy> we will actively work against it to ruin it for them, in whatever way we can
  267. # [12:53] <zcorpan> lol
  268. # [12:54] <Lachy> zcorpan, I'm serious! This is no lauging matter!
  269. # [12:54] <beowulf> i have a dumb question, we discuss the various ways in which the web is broken, but we have a design principal of "don't break the web", is this a flaw?
  270. # [12:54] <zcorpan> Lachy: oh i didn't think you were kidding, it was just funny :)
  271. # [12:54] <Lachy> ok
  272. # [12:55] <Lachy> beowulf, don't break the web means don't try to fix the broken parts of the web in incompatible ways, rather we should standardise them
  273. # [12:55] <krijnh> Don't break the broken web
  274. # [12:55] <Lachy> yes
  275. # [12:55] <edas> and my point is that if mozilla, opera and apple keep enough market share, one of them will do a bug somewhere in the future. This bug will be exploited by some. These browser will have the same problems that MS. Either they correct the faulty one, either they change the others browers. Both ways will break the web so they will have to do versionning and/or opt-ins.
  276. # [12:55] <Lachy> "if it aint fixed, don't fix it"
  277. # [12:56] <zcorpan> "if it aint fixed, don't break it (more)"
  278. # [12:56] <edas> this may be today a ie-specific question but I do not think it will stay ie-specific
  279. # [12:56] <beowulf> so it's more "maintain the broken state"
  280. # [12:56] <Lachy> edas, yes. And that's exactly what IE think they can do because of their monopoly
  281. # [12:57] <zcorpan> edas: mozilla, opera and apple don't want to introduce versioning
  282. # [12:57] <krijnh> Even php doesn't maintain the broken state :p
  283. # [12:57] <edas> zcorpan, are they accepting future breaks in the web if they need to correct something ? (I hope a yes but I heard a "no" somewhere)
  284. # [12:58] <edas> (if one of them have to correct something, not all of them)
  285. # [12:58] <beowulf> i asked because I was wondering about any mileage in convincing cwilso to accept a certain amount of change in the broken state
  286. # [12:59] <zcorpan> edas: if they don't want to correct it, the spec has to change. if two vendors don't want to correct something and they disagree, we have to find out which would break least and the other vendor would have to give
  287. # [12:59] <beowulf> and I have every sympathy for the guy in what that means
  288. # [12:59] <zcorpan> edas: that way, in the end, we still achieve interop
  289. # [13:00] <Lachy> we have to standardise the path of least resistance
  290. # [13:00] * Joins: Charl (charlvn@196.209.106.139)
  291. # [13:00] <Lachy> hey Charl
  292. # [13:00] <Charl> hi Lachy :)
  293. # [13:00] <edas> zcorpan, ok, so you're saying that if we have a need for that, we will break some pages. I'm ok with that, but that's not the consensus I've seen on the list
  294. # [13:01] <Lachy> edas, we will try to break as few pages as possible
  295. # [13:01] <zcorpan> edas: there is no consensus on the list afaict
  296. # [13:01] <Lachy> see my latest rant on the list, I talked about that issue
  297. # [13:02] <beowulf> that was a good rant
  298. # [13:02] <Lachy> http://www.w3.org/mid/462051BA.4090709@lachy.id.au
  299. # [13:02] <Lachy> beowulf, thanks :-)
  300. # [13:02] <zcorpan> beowulf: indeed
  301. # [13:04] * Joins: ROBOd (robod@86.34.246.154)
  302. # [13:05] <krijnh> What makes this URI easier/better ^ ?
  303. # [13:05] <beowulf> the last line should be the wh motto :)
  304. # [13:05] <beowulf> s/wh/wg
  305. # [13:06] <krijnh> Lachy: what's the difference between http://lists.w3.org/Archives/Public/public-html/2007Apr/0769.html and the link you posted? Or is that a dumb question?
  306. # [13:06] <Lachy> the one I pasted comes from the X-Archived-At header in the e-mail
  307. # [13:06] <krijnh> Ah
  308. # [13:06] <Lachy> I copied it directly from there, instead of going to the mailng list
  309. # [13:07] <krijnh> Handy!
  310. # [13:07] <Lachy> but I wish TB made it easier for me to get, without having to view source of the email
  311. # [13:07] <krijnh> Opera makes you view source as well
  312. # [13:08] <Lachy> I looked for an extension to do it once, but I never found one
  313. # [13:08] <Lachy> I should write one, when I figure out how
  314. # [13:08] <zcorpan> krijnh: how?
  315. # [13:09] <krijnh> Right mouse in the message - view headers
  316. # [13:09] <zcorpan> ah
  317. # [13:09] <zcorpan> just found it
  318. # [13:10] <krijnh> What Matt Freels proposes seems to make sense
  319. # [13:11] <Lachy> I think if MS is worried more about intranet apps, they could allow users to select the IE7 rendering for certain domains
  320. # [13:12] <krijnh> Users, or sys admins
  321. # [13:12] <Lachy> so if they have a broken intranet app, then can configure IE8 to use IE7 rendering by default on such domains
  322. # [13:12] <Lachy> sys admins
  323. # [13:12] <beowulf> ooh, interesting
  324. # [13:12] <Lachy> or even users if they don't have sys admins
  325. # [13:12] <krijnh> alert("Contact your sys admin if this page looks sucky.")
  326. # [13:13] <beowulf> do you mean like a domain whitelist?
  327. # [13:13] <Lachy> Netscape 8 did a similar thing with their IE/Firefox rendering engine switch
  328. # [13:13] <Lachy> they just did it wrongly by defaulting to IE for "trusted" sites
  329. # [13:13] * Joins: gsnedders (gsnedders@86.139.123.225)
  330. # [13:14] <Lachy> but with IE8, it wouldn't need to be as badly implemented because most sites should work with upgraded rendering engines. It's just for the few extremely bad sites that rely on edge cases
  331. # [13:15] <beowulf> i like that idea
  332. # [13:15] <beowulf> "render this page as ie5"
  333. # [13:15] * Joins: zcorpan_ (zcorpan@84.216.43.132)
  334. # [13:15] <Lachy> that feature would be awesome for web devs since they don't allow multi-iE installations
  335. # [13:16] <beowulf> and then mr sysadmin pushes that out to all the ie's in the office
  336. # [13:17] * Quits: zcorpan (zcorpan@84.216.42.156) (Ping timeout)
  337. # [13:17] <Lachy> Matt Freels has the same idea: "Wouldn't this breakage be mitigated by allowing users to manually fall back on IE7 when needed?"
  338. # [13:17] <krijnh> <krijnh> What Matt Freels proposes seems to make sense
  339. # [13:18] <krijnh> Was that coincidence?
  340. # [13:18] <Lachy> yes
  341. # [13:18] <Lachy> I hadn't read matt's post, I'd been thinking about the idea earlier today
  342. # [13:19] <beowulf> so what you're both saying is that it's not html that needs versioning, but IE
  343. # [13:19] <zcorpan_> it would mean that authors would have even harder to ignore older versions of ie... since some users would use the ie7 mode in ie8, even when no-one uses the actual ie7
  344. # [13:20] <krijnh> No, it would mean a sys admin can 'fix' their intranet application if it doesn't work in IE8
  345. # [13:20] <beowulf> it wouldn't if the setting was "render as ie n for this domain"
  346. # [13:20] * Joins: porneL (pornel@84.92.248.233)
  347. # [13:20] * Parts: porneL (pornel@84.92.248.233)
  348. # [13:21] <edas> people doesn't use settings
  349. # [13:21] <edas> (not the ones they should)
  350. # [13:21] <beowulf> people in big corporations do, they have them set by sysadmins, no?
  351. # [13:21] <beowulf> and those are the groups we seem to be most worried about
  352. # [13:21] <krijnh> We? MS
  353. # [13:22] <krijnh> And they should be
  354. # [13:22] <beowulf> true
  355. # [13:22] <krijnh> For an intranet application it makes a bit sense to not use standards
  356. # [13:22] <beowulf> if they piss them off, they might look for a new browser, who knows where that could lead!
  357. # [13:23] <krijnh> Alternative browsers wouldn't render that app well then either
  358. # [13:24] <krijnh> So, in IE7 you had the ClearType checkbox
  359. # [13:24] <krijnh> In IE8 you have the IE7 rendering mode checkbox
  360. # [13:25] <krijnh> Although that should be a list of domains which use the IE7 rendering mode, else we would have the same trouble we have now
  361. # [13:26] * Quits: Shunsuke_ (kuruma@219.110.80.235) (Ping timeout)
  362. # [13:27] * Quits: edas (edaspet@88.191.34.123) (Quit: http://eric.daspet.name/ et l'édition 2007 de http://www.paris-web.fr/ )
  363. # [13:29] * Lachy responds to Matt Freels
  364. # [13:29] * krijnh F5's
  365. # [13:30] <Lachy> krijnh, alternative browsers are often locked out of such apps these days anyway
  366. # [13:30] <krijnh> Can't MS take an example of Opera and browser.js ?
  367. # [13:30] <Lachy> what does browser.js do?
  368. # [13:30] <krijnh> Fix sites that break in Opera
  369. # [13:30] <Lachy> how does it work?
  370. # [13:31] <krijnh> http://www.opera.com/docs/browserjs/
  371. # [13:31] <krijnh> Very clever
  372. # [13:31] <krijnh> Ugly, but clever
  373. # [13:31] <krijnh> No breakage for users
  374. # [13:31] <krijnh> They could do that for 'public' pages
  375. # [13:32] <krijnh> And use the sys admin list of IE7 rendered domains for intranet apps
  376. # [13:32] <Lachy> ah, I see. It has a list of sites and known patches to apply to them
  377. # [13:32] <krijnh> Yep
  378. # [13:32] <Lachy> I thought it was some magic script that just fixed any random site
  379. # [13:32] <krijnh> Also some common libraries are fixed
  380. # [13:32] <krijnh> Like prototype, needs some fixes in a specific Opera version
  381. # [13:33] <krijnh> If a new version is shipped, Opera modifies browser.js
  382. # [13:33] <Lachy> was that an opera or prototype bug?
  383. # [13:33] <krijnh> And afaik it's also updated automatically
  384. # [13:33] <Lachy> is it enabled by default?
  385. # [13:33] <krijnh> Does it matter whos bug it is? As long as somebody fixes it
  386. # [13:33] <krijnh> Yes
  387. # [13:34] <krijnh> "// working around incompatibility with prototype, bug 241832"
  388. # [13:34] <krijnh> From browser.js
  389. # [13:34] <Lachy> krijnh, yes it matters. If it's an Opera bug, Opera should fix it in their source code. If it's a prototype bug, they should notify the prototype devs to fix it
  390. # [13:34] <krijnh> Lachy: Once it's fixed in either prototype, or Opera, the fix in browser.js wouldn't be applied
  391. # [13:35] <Lachy> ok, so browser.js is a temporary work around
  392. # [13:35] * Joins: hasather_ (hasather@81.235.209.174)
  393. # [13:35] <krijnh> yes
  394. # [13:35] <Lachy> awesome!
  395. # [13:35] <krijnh> "Opera automatically checks for updates to Browser JavaScript once every week."
  396. # [13:36] <krijnh> I think it's pretty nasty for the developers maintaining the file, but heck
  397. # [13:37] <krijnh> The March 30 version is only 1287 lines :]
  398. # [13:38] <krijnh> I think it's a task for the WaSP or something to fix sites that break eventually
  399. # [13:38] <krijnh> Push developers to use standards, which they do already
  400. # [13:39] <krijnh> For intranet apps I don't think this is workable, since most of the time there is a lot of money involved
  401. # [13:40] <krijnh> "Broken intranet apps." from your mail - I don't think you should name 'm that way. They can work perfectly well in an intranet.
  402. # [13:41] <Lachy> well, I wanted to distinguish the apps I was referring to like autotask.net from good ones that are standards compliant
  403. # [13:41] <krijnh> I think IE was the richest development platform for a long time
  404. # [13:42] <krijnh> Perhaps it still is
  405. # [13:42] <krijnh> Applications specifically written for it make sense, and IE should always support them
  406. # [13:42] <Lachy> I don't know how developers build for it, it's so painful every time I need to
  407. # [13:42] <krijnh> You don't have to know
  408. # [13:42] <krijnh> You and I are different developers
  409. # [13:42] <Lachy> apps written specifically for it are broken by design
  410. # [13:43] <krijnh> Nah, they're not
  411. # [13:43] <Lachy> yes, they are
  412. # [13:43] <Lachy> as soon as you write an app that runs only in IE, you create vendor lock-in for yourself
  413. # [13:44] <krijnh> If you know that, it's not broken
  414. # [13:44] <krijnh> You mean every piece of software which is written only for Windows is broken by design?
  415. # [13:44] <Lachy> it is for users who come along and want to use it in their choice of browser, not yours
  416. # [13:44] <krijnh> The same for software for Mac OS X
  417. # [13:44] <Lachy> krijnh, no
  418. # [13:44] <krijnh> I'm talking about intranet stuff here
  419. # [13:44] <Lachy> HTML was intended to be an interoperable medium
  420. # [13:45] <krijnh> Yeah, IE expanded HTML a bit, so it could be used as a development platform
  421. # [13:45] <Lachy> e.g. at my former employer, I was forced to use IE when I needed to access autotask.net for timesheets. It sucked! Our designer, who used a mac, was forced to use another PC in the office
  422. # [13:46] * Quits: gavin_ (gavin@74.103.208.221) (Ping timeout)
  423. # [13:47] <krijnh> Probably more sucky for that designer than for you
  424. # [13:47] <Lachy> no, it was sucky for me cause I hate using IE for anything
  425. # [13:50] <krijnh> Employees should complain about that with their boss
  426. # [13:50] <krijnh> Who bought that app
  427. # [13:51] <Lachy> I don't know, I only found out about it about 3 weeks ago after it was all set up and had to start using it
  428. # [13:51] * Joins: gavin_ (gavin@74.103.208.221)
  429. # [13:51] <Lachy> luckily, I don't have to use it any more
  430. # [13:51] <krijnh> :)
  431. # [13:53] <krijnh> But others do, and MS can't break that for them
  432. # [13:55] * Joins: hasather (hasather@81.235.209.174)
  433. # [14:08] * Quits: loic (loic@90.27.88.46) (Ping timeout)
  434. # [14:23] * Joins: loic (loic@90.27.89.124)
  435. # [14:24] * Joins: Shunsuke (kuruma@219.110.80.235)
  436. # [14:31] <krijnh> Wow, Google payed $3.1 billion for DoubleClick :|
  437. # [14:33] <Lachy> I can't believe it was that much. I can't belive DoubleClick is still in business after all these years
  438. # [14:34] <Lachy> I blame the stupid average users that click advertisements!
  439. # [14:34] <krijnh> I like 'm
  440. # [14:35] <hsivonen> wow indeed
  441. # [14:35] <hsivonen> will google serve graphic ads or convert the client base to AdSense with the current notion of what AdSense is?
  442. # [14:36] <Lachy> they might offer both services
  443. # [14:36] <Lachy> there's obviously still a big market for banner ads and google wants a slice of it
  444. # [14:37] * Quits: Charl (charlvn@196.209.106.139) (Quit: Leaving)
  445. # [15:53] * Quits: gavin_ (gavin@74.103.208.221) (Ping timeout)
  446. # [15:58] * Joins: gavin_ (gavin@74.103.208.221)
  447. # [18:01] * Quits: gavin_ (gavin@74.103.208.221) (Ping timeout)
  448. # [18:06] * Joins: gavin_ (gavin@74.103.208.221)
  449. # [18:16] * Joins: sbuluf (uzbywtv@200.49.140.155)
  450. # [18:22] * Quits: myakura (myakura@60.239.122.32) (Quit: Leaving...)
  451. # [18:29] <beowulf> that indent thread is... i dunno
  452. # [18:30] <beowulf> unreal
  453. # [19:05] * Quits: loic (loic@90.27.89.124) (Ping timeout)
  454. # [19:07] * Joins: Roger (roger@213.64.74.230)
  455. # [19:12] * Parts: hasather (hasather@81.235.209.174)
  456. # [19:15] * Joins: hasather (hasather@81.235.209.174)
  457. # [19:37] * Parts: hasather (hasather@81.235.209.174)
  458. # [19:40] * Joins: hasather (hasather@81.235.209.174)
  459. # [19:48] * Quits: Roger (roger@213.64.74.230) (Quit: Roger)
  460. # [19:51] * Parts: hasather (hasather@81.235.209.174)
  461. # [19:52] * Quits: zcorpan_ (zcorpan@84.216.43.132) (Ping timeout)
  462. # [19:53] * Quits: heycam (cam@203.214.79.176) (Ping timeout)
  463. # [20:02] * Joins: hasather (hasather@81.235.209.174)
  464. # [20:04] * Parts: hasather (hasather@81.235.209.174)
  465. # [20:18] * Quits: gavin_ (gavin@74.103.208.221) (Ping timeout)
  466. # [20:20] * Quits: Shunsuke (kuruma@219.110.80.235) (Quit: See you...)
  467. # [20:23] * Joins: gavin_ (gavin@74.103.208.221)
  468. # [20:59] * Joins: mjs (mjs@64.81.48.145)
  469. # [21:26] * Joins: Zeros (Zeros-Elip@69.140.48.129)
  470. # [21:45] * Joins: hasather (hasather@81.235.209.174)
  471. # [21:48] * Quits: beowulf (carisenda@91.84.50.132) (Ping timeout)
  472. # [21:57] * Joins: beowulf (carisenda@91.84.50.132)
  473. # [22:06] * Quits: ROBOd (robod@86.34.246.154) (Quit: http://www.robodesign.ro )
  474. # [22:17] * Parts: hasather_ (hasather@81.235.209.174)
  475. # [22:25] * Quits: gavin_ (gavin@74.103.208.221) (Ping timeout)
  476. # [22:27] * Parts: hasather (hasather@81.235.209.174)
  477. # [22:29] * Joins: zcorpan_ (zcorpan@84.216.41.224)
  478. # [22:30] * Joins: gavin_ (gavin@74.103.208.221)
  479. # [22:53] * Joins: hasather (hasather@81.235.209.174)
  480. # [22:55] * Quits: Zeros (Zeros-Elip@69.140.48.129) (Quit: Leaving)
  481. # [23:04] * Quits: Hixie (ianh@129.241.93.37) (Ping timeout)
  482. # [23:12] * Joins: Hixie (ianh@129.241.93.37)
  483. # [23:22] <Hixie> Lachy: as a sylistic point, i'd encourage not saying "that's a great question" when you wrote the question :-)
  484. # [23:23] * Joins: heycam (cam@203.214.79.176)
  485. # Session Close: Sun Apr 15 00:00:00 2007

The end :)