/irc-logs / freenode / #whatwg / 2010-01-29 / end

Options:

1. # Session Start: Fri Jan 29 00:00:01 2010
2. # Session Ident: #whatwg
3. # [00:02] * Joins: dimich (n=dimich@74.125.59.65)
4. # [00:04] * Joins: mpilgrim (n=mark@rrcs-98-101-146-174.midsouth.biz.rr.com)
5. # [00:05] <mpilgrim> zcorpan: i have gone through my "Dive Into HTML5" inbox and addressed everyone's feedback except yours
6. # [00:05] <mpilgrim> and now it's 6pm and i have to go play video games with my son
7. # [00:06] <mpilgrim> maybe next week
8. # [00:06] * Quits: mpilgrim (n=mark@rrcs-98-101-146-174.midsouth.biz.rr.com) (Client Quit)
9. # [00:09] * Joins: Rik`_ (n=Rik`@pha75-2-81-57-187-57.fbx.proxad.net)
10. # [00:10] * Quits: Rik` (n=Rik`@pha75-2-81-57-187-57.fbx.proxad.net) (Read error: 60 (Operation timed out))
11. # [00:10] * Rik`_ is now known as Rik`
12. # [00:11] * Quits: mpt (n=mpt@canonical/mpt) ("Ex-Chat")
13. # [00:15] <AryehGregor> Is there a way in CSS to force table cells to have the heights you want, like table-layout: fixed does for widths?
14. # [00:16] <AryehGregor> Also, any way to force something to be exactly one page tall, in print?
15. # [00:30] * Joins: Utkarsh (n=admin@117.201.81.52)
16. # [00:32] * Quits: ttepass- (n=ttepas--@dslb-084-060-080-019.pools.arcor-ip.net) ("404")
17. # [00:39] * Quits: Rik` (n=Rik`@pha75-2-81-57-187-57.fbx.proxad.net) (Read error: 54 (Connection reset by peer))
18. # [00:40] * Joins: Rik` (n=Rik`@pha75-2-81-57-187-57.fbx.proxad.net)
19. # [00:41] * Quits: jgornick (n=joe@199.199.210.66)
20. # [00:42] * Joins: jorlow (n=jorlow@nat/google/x-kevonnsvohxcsofh)
21. # [00:46] * Quits: _Utkarsh (n=admin@117.201.80.36) (Read error: 110 (Connection timed out))
22. # [00:50] * Quits: svl (n=me@ip565744a7.direct-adsl.nl) ("And back he spurred like a madman, shrieking a curse to the sky.")
23. # [00:52] * Quits: giroro_ (i=Ruetobas@114-32-107-228.HINET-IP.hinet.net)
24. # [00:54] * Joins: slightlyoff (n=slightly@72.14.229.81)
25. # [00:56] * Joins: Ruetobas (i=Ruetobas@114-32-107-228.HINET-IP.hinet.net)
26. # [00:58] * Joins: erikvold (n=erikvold@S01060024012860e9.gv.shawcable.net)
27. # [00:59] * Quits: starjive (i=beos@81-233-16-19-no30.tbcn.telia.com) (leguin.freenode.net irc.freenode.net)
28. # [00:59] * Quits: pablof (n=palbo@pat-tdc.opera.com) (leguin.freenode.net irc.freenode.net)
29. # [00:59] * Quits: salavas (i=salavas@h4n1fls31o279.telia.com) (leguin.freenode.net irc.freenode.net)
30. # [00:59] * Joins: salavas (i=salavas@h4n1fls31o279.telia.com)
31. # [00:59] * Joins: starjive (i=beos@81-233-16-19-no30.tbcn.telia.com)
32. # [00:59] * Joins: pablof (n=palbo@pat-tdc.opera.com)
33. # [01:01] * Joins: dglazkov_ (n=dglazkov@nat/google/x-qnnwcfaqcbaijnmu)
34. # [01:04] * Quits: aroben (n=aroben@unaffiliated/aroben) (Read error: 60 (Operation timed out))
35. # [01:16] * Quits: dglazkov (n=dglazkov@nat/google/x-udlljqgsjfkijgmt) (Read error: 110 (Connection timed out))
36. # [01:16] * dglazkov_ is now known as dglazkov
37. # [01:18] * Quits: cying (n=cying@70.90.171.153) (Read error: 110 (Connection timed out))
38. # [01:31] * Joins: aroben (n=aroben@unaffiliated/aroben)
39. # [01:31] * Quits: Sidnicious (n=Sidney@pdpc/supporter/professional/sidney)
40. # [01:35] * Joins: seventh (i=seventh@189.59.202.245)
41. # [01:38] * Quits: dglazkov (n=dglazkov@nat/google/x-qnnwcfaqcbaijnmu) (Remote closed the connection)
42. # [01:38] * Joins: dglazkov (n=dglazkov@nat/google/x-ybeqfspnygwdgshz)
43. # [01:41] * Joins: boblet (n=boblet@p1072-ipbf36osakakita.osaka.ocn.ne.jp)
44. # [01:43] * Joins: wakaba_ (n=wakaba_@122x221x184x68.ap122.ftth.ucom.ne.jp)
45. # [01:54] * Joins: _Utkarsh (n=admin@117.201.80.93)
46. # [02:00] * Joins: cedric_ (n=cedric@124.197.108.103)
47. # [02:00] * Joins: taf2_ (n=taf2@pool-98-117-216-229.bltmmd.fios.verizon.net)
48. # [02:02] * Joins: jdandrea (n=jdandrea@pool-98-109-116-42.nwrknj.fios.verizon.net)
49. # [02:02] * Parts: jdandrea (n=jdandrea@pool-98-109-116-42.nwrknj.fios.verizon.net)
50. # [02:06] * Quits: cedricv (n=cedric@112.199.140.226) (Read error: 110 (Connection timed out))
51. # [02:09] * Quits: Utkarsh (n=admin@117.201.81.52) (Read error: 110 (Connection timed out))
52. # [02:11] * Joins: takoratta (n=takoratt@203.39.247.241)
53. # [02:17] * Quits: dglazkov (n=dglazkov@nat/google/x-ybeqfspnygwdgshz)
54. # [02:24] * Quits: JonathanNeal (n=Jonathan@rrcs-76-79-114-213.west.biz.rr.com) (Read error: 113 (No route to host))
55. # [02:26] * Joins: paul_irish (n=paul_iri@c-71-192-163-128.hsd1.nh.comcast.net)
56. # [02:27] * Joins: enigmus (n=e@S0106001d7e52d1d9.vc.shawcable.net)
57. # [02:27] <enigmus> Any work on multitouch gesture support?
58. # [02:28] <AryehGregor> What, like some kind of multitouch event?
59. # [02:28] <AryehGregor> What would the API for that look like?
60. # [02:28] <enigmus> AryehGregor: Yeah. Mozilla has a few, the iPhone browser has also some support.
61. # [02:29] <enigmus> https://developer.mozilla.org/En/DOM/Mouse_gesture_events
62. # [02:29] <AryehGregor> Interesting.
63. # [02:29] <AryehGregor> Dunno.
64. # [02:29] * Joins: Utkarsh (n=admin@117.201.81.213)
65. # [02:30] <enigmus> http://developer.apple.com/safari/library/documentation/AppleApplications/Reference/SafariJSRef/TouchEvent/TouchEvent.html#//apple_ref/doc/uid/TP40001482-CH4-SW1
66. # [02:30] <enigmus> AryehGregor: well, yeah. That's going to be pretty fundamental very soon, with devices like the iPad popping up everywhere, I'm sure.
67. # [02:31] <AryehGregor> Is HTML really supposed to add new input events for every new type of input device that comes along?
68. # [02:31] <AryehGregor> I guess so.
69. # [02:31] <enigmus> I'm afraid there might be some problem with patents, though. Apple seems to claim ownership of a few of these, more or less.
70. # [02:31] * Joins: JonathanNeal (n=Jonathan@99-59-124-67.lightspeed.irvnca.sbcglobal.net)
71. # [02:32] <enigmus> AryehGregor: I would think so. Otherwise, the goal of webapps taking over the world doesn't make a lot of sense if the only way of using newer devices is to write a native app.
72. # [02:32] <AryehGregor> Well, we have the W3C Patent Policy to handle the patents.
73. # [02:33] <AryehGregor> If they have any, they have to declare them when the feature gets to Working Draft, and say whether they'll license them royalty-free.
74. # [02:34] <enigmus> Great. I'm not sure what is the situation: Android and Nokia n900 do not support things like "pinch" to zoom (even though the hardware is capable), but Windows 7 does.
75. # [02:35] <enigmus> I've heard claims Apple says they feel they have a patent on that one. Would they license it to MS? Maybe.
76. # [02:36] <AryehGregor> If it's in a Working Draft, they have to say so, yes or no.
77. # [02:36] <AryehGregor> Royalty-free for all implementations of the standard, or not.
78. # [02:36] <AryehGregor> And if not, they could be kicked out of the Working Group, as I understand it, if the patent covers "essential claims" (although I might misremember).
79. # [02:36] <AryehGregor> (this is all W3C stuff, not WHATWG)
80. # [02:37] <enigmus> I see. I really hope there is going to be some work toward multitouch, then.
81. # [02:39] * Quits: tndH (n=Rob@cpc2-leed18-0-0-cust427.leed.cable.ntl.com) ("ChatZilla 0.9.86-rdmsoft [XULRunner 1.9.0.1/2008072406]")
82. # [02:39] <AryehGregor> You could propose it on the whatwg list.
83. # [02:39] <Hixie> multitouch would be better addressed to the www-dom list
84. # [02:39] <enigmus> AryehGregor: OK, I'll check the archives and suggest the topic.
85. # [02:39] <AryehGregor> Okay, listen to Hixie.
86. # [02:39] <AryehGregor> Not me. :)
87. # [02:39] <Hixie> but first we need a spec for single-touch, and, like, mice. and keyboards, even.
88. # [02:39] <AryehGregor> Heh.
89. # [02:45] * Quits: _Utkarsh (n=admin@117.201.80.93) (Read error: 110 (Connection timed out))
90. # [02:46] * Joins: nattokirai (n=nattokir@118.243.226.86)
91. # [02:54] <enigmus> There was some discussion on the list: http://lists.w3.org/Archives/Public/www-dom/2009AprJun/0134.html and Nokia, more recently http://lists.w3.org/Archives/Public/www-dom/2009OctDec/0041.html
92. # [02:57] * Quits: Utkarsh (n=admin@117.201.81.213) (Connection timed out)
93. # [02:58] <enigmus> But not much since October.
94. # [03:00] * Joins: dglazkov (n=dglazkov@c-67-188-0-62.hsd1.ca.comcast.net)
95. # [03:04] * Joins: enigmus_ (n=e@S0106001d7e52d1d9.vc.shawcable.net)
96. # [03:07] * Quits: borismus (n=borismus@bl4-167-3.dsl.telepac.pt)
97. # [03:11] * Quits: aroben (n=aroben@unaffiliated/aroben) (Read error: 104 (Connection reset by peer))
98. # [03:15] * Joins: MikeSmithX (n=MikeSmit@EM114-48-109-50.pool.e-mobile.ne.jp)
99. # [03:16] * Quits: ap (n=ap@17.246.19.5)
100. # [03:17] <TabAtkins> Hasn't it already been established that text/html-sandboxed is *not* mime-sniffed as text/html in legacy browsers?
101. # [03:17] <AryehGregor> I don't think it's been established as true under all circumstances beyond any doubt.
102. # [03:17] <AryehGregor> I saw someone say that IE might pose a problem if the URL ends in .html or such.
103. # [03:18] <TabAtkins> Hmm, k.
104. # [03:19] * Quits: KevinMarks (n=KevinMar@157.22.22.46) (Client Quit)
105. # [03:33] * Quits: MikeSmith (n=MikeSmit@EM114-48-156-170.pool.e-mobile.ne.jp) (Read error: 110 (Connection timed out))
106. # [03:35] * MikeSmithX is now known as MikeSmith
107. # [03:42] <othermaciej> IE will potentially sniff even if the query part ends in .html
108. # [03:42] <othermaciej> so tricky to defend against
109. # [03:47] <TabAtkins> Just don't have the file end in .html?
110. # [03:47] <TabAtkins> I mean, for the common case of serving comments, it'll be a .php or .aspx commonly.
111. # [03:48] <boblet> Re: “A header element is intended to usually contain the section's heading (an h1–h6 element or an hgroup element), but this is not required”
112. # [03:49] <boblet> “this is not required” seems to mean the header doesn’t have to contain a heading, but a section generally should have a heading, right?
113. # [03:50] <boblet> is “this is not required” for multiple headings in one section & web app use where the section lacks a heading?
114. # [03:50] <boblet> or are untitled sections not as ‘to be avoided’ as I thought?
115. # [03:58] * Joins: KevinMarks (n=KevinMar@157.22.22.46)
116. # [04:01] <othermaciej> TabAtkins: php is often set up to respond even if you tack on arbitrary stuff to the URL
117. # [04:01] <othermaciej> TabAtkins: and if your URL takes a query parameter, and IE looks at that, you can tack on .html via a query
118. # [04:01] <othermaciej> http://example.com/script.php?foo=bar.html
119. # [04:02] <TabAtkins> That's the most ridiculous bullshit I've ever heard of, and I hate whoever did that with a fiery passion.
120. # [04:04] <othermaciej> the IE team? or the PHP team?
121. # [04:05] <othermaciej> I don't see how either could be any more of a hate target than they are already...
122. # [04:05] <TabAtkins> IE team.
123. # [04:06] <othermaciej> so up til now, you've been totally cool with the 90s / early 2000s IE team
124. # [04:06] <othermaciej> but now that you have learned this one additional fact...
125. # [04:07] <TabAtkins> No. But I hate the specific person in charge of the decision to activate sniffing if the query params happen to end in ".html".
126. # [04:07] <TabAtkins> It was a generalized hate before. But now it's focused on that one person.
127. # [04:08] * Quits: breakmau5 (n=breakz@erft-5d80d2dd.pool.mediaWays.net) (Read error: 113 (No route to host))
128. # [04:10] * Quits: dglazkov (n=dglazkov@c-67-188-0-62.hsd1.ca.comcast.net)
129. # [04:17] * Quits: othermaciej (n=mjs@17.246.17.36)
130. # [04:18] <MikeSmith> Hixie: what does a datalist element with no option children represent?
131. # [04:18] <MikeSmith> or to put it another way, what's the use case for a datalist element with no option children?
132. # [04:20] * Quits: drunknbass_work (n=aaron@71.107.253.243) (Client Quit)
133. # [04:31] * Quits: Heimidal (n=heimidal@unaffiliated/heimidal) (Read error: 110 (Connection timed out))
134. # [04:42] * Joins: dglazkov (n=dglazkov@c-67-188-0-62.hsd1.ca.comcast.net)
135. # [04:43] <MikeSmith> ?????
136. # [04:47] * Quits: dbaron (n=dbaron@nat/mozilla/x-inxbocvsqxfkvegu) ("8403864 bytes have been tenured, next gc will be global.")
137. # [04:51] <karlcow> followup on the previous one http://googleblog.blogspot.com/2010/01/unicode-nearing-50-of-web.html
138. # [04:52] * Parts: quuxman (i=foobar@174-21-110-160.tukw.qwest.net)
139. # [04:53] * Parts: MikeSmith (n=MikeSmit@EM114-48-109-50.pool.e-mobile.ne.jp) ("Tomorrow to fresh woods, and pastures new.")
140. # [04:54] * Joins: MikeSmith (n=MikeSmit@EM114-48-109-50.pool.e-mobile.ne.jp)
141. # [04:54] <MikeSmith> 豪華な邸宅
142. # [04:55] <karlcow> MikeSmith: you are moving :p
143. # [04:55] <MikeSmith> heh
144. # [04:56] <MikeSmith> I just switched my IRC client to XChat Aqua and was having some trouble figuring out how to set it to use UTF-8
145. # [04:56] <MikeSmith> Its UI is suboptimal
146. # [04:57] <MikeSmith> I tried Colloquy and really liked it but it has some weird problems with tircd
147. # [04:58] <MikeSmith> karlcow: btw, there is this fat tomcat at the building next door to t45 at Keio
148. # [04:58] <karlcow> オンボロ家屋
149. # [04:58] <MikeSmith> the building where the Murai-sensei faculty have their labs
150. # [04:58] <MikeSmith> WIDE project
151. # [04:58] <MikeSmith> they call this tomcat 社長
152. # [04:59] <karlcow> hehe
153. # [04:59] <MikeSmith> until yesterday, all he had was a dinky cardboard box to sleep in
154. # [04:59] <MikeSmith> but apparantly today he has this big wooden doghouse
155. # [05:00] <MikeSmith> I'm not sure he likes it much
156. # [05:00] <karlcow> a zero-yen house after the collapse of the economy ;)
157. # [05:00] <MikeSmith> heh
158. # [05:00] <MikeSmith> so far I think he has not gone into it
159. # [05:00] * Joins: pmuellr (n=pmuellr@user-0c99bjg.cable.mindspring.com)
160. # [05:00] <MikeSmith> dude pretty much sits around all day and barely even opens his eyes when you pet him
161. # [05:01] <karlcow> maybe a dudette could help him move in a better place instead of his shabby shelter. who knows :)
162. # [05:01] <MikeSmith> they had a sign attached to the cardboard box: Please consider Shacho's health
163. # [05:02] <MikeSmith> implication of the sign is, don't feed him, because he's too fat already
164. # [05:02] <karlcow> :D
165. # [05:02] <MikeSmith> but being that it's Japan, even signs instructing people not to fee fat tomcats have to be polite
166. # [05:03] <MikeSmith> anyway, Naoko has some pictures
167. # [05:04] <karlcow> I'll have to come by myself to see that
168. # [05:05] <karlcow> I'm planning to come to Japan. No date defined yet. Probably in April. not sure
169. # [05:09] * Quits: cpearce (n=cpearce@203-97-204-82.dsl.clear.net.nz) ("ChatZilla 0.9.86-rdmsoft [XULRunner 1.9.0.15/2009101909]")
170. # [05:23] * Joins: dbaron (n=dbaron@c-98-234-51-190.hsd1.ca.comcast.net)
171. # [05:25] <MikeSmith> karlcow: would be great to see you
172. # [05:25] <karlcow> yep yep
173. # [05:25] <karlcow> impatient
174. # [05:26] <karlcow> ラメンとなまビウ
175. # [05:27] * Joins: othermaciej (n=mjs@c-69-181-42-237.hsd1.ca.comcast.net)
176. # [05:34] * Quits: seventh (i=seventh@189.59.202.245) (Read error: 104 (Connection reset by peer))
177. # [05:34] * Joins: seventh (i=seventh@189.59.202.245)
178. # [05:36] * Quits: paul_irish (n=paul_iri@c-71-192-163-128.hsd1.nh.comcast.net) (Remote closed the connection)
179. # [05:36] * Joins: cpearce (n=cpearce@ip-118-90-40-252.xdsl.xnet.co.nz)
180. # [05:36] * Joins: paul_irish (n=paul_iri@c-71-192-163-128.hsd1.nh.comcast.net)
181. # [05:48] * Quits: enigmus_ (n=e@S0106001d7e52d1d9.vc.shawcable.net) (Read error: 104 (Connection reset by peer))
182. # [05:49] * Joins: cying (n=cying@adsl-75-18-231-95.dsl.pltn13.sbcglobal.net)
183. # [05:58] * Quits: Rik` (n=Rik`@pha75-2-81-57-187-57.fbx.proxad.net) (Read error: 113 (No route to host))
184. # [05:58] * Joins: Rik`_ (n=Rik`@pha75-2-81-57-187-57.fbx.proxad.net)
185. # [06:04] * Joins: Rik` (n=Rik`@pha75-2-81-57-187-57.fbx.proxad.net)
186. # [06:05] * Quits: Rik`_ (n=Rik`@pha75-2-81-57-187-57.fbx.proxad.net) (Read error: 104 (Connection reset by peer))
187. # [06:06] * Joins: drunknbass (n=drunknba@cpe-76-173-195-145.socal.res.rr.com)
188. # [06:13] * Quits: slightlyoff (n=slightly@72.14.229.81)
189. # [06:17] * Quits: Amorphous (i=jan@unaffiliated/amorphous) (Read error: 113 (No route to host))
190. # [06:20] * Joins: Amorphous (i=jan@unaffiliated/amorphous)
191. # [06:26] * Hixie ponders how to solve the microdata-to-rdf problem
192. # [06:34] <othermaciej> it isn't solved already?
193. # [06:35] * Quits: MikeSmith (n=MikeSmit@EM114-48-109-50.pool.e-mobile.ne.jp) ("Tomorrow to fresh woods, and pastures new.")
194. # [06:36] <Hixie> othermaciej: issues were brought up
195. # [06:36] * Quits: KevinMarks (n=KevinMar@157.22.22.46) (Client Quit)
196. # [06:36] <othermaciej> sadness
197. # [06:37] <Hixie> specifically, how to convert the following:
198. # [06:37] <Hixie> <div itemscope itemtype="http://example.com/a" itemref="x"></div>
199. # [06:37] <Hixie> <div itemscope itemtype="http://example.com/b" itemref="x"></div>
200. # [06:37] <Hixie> <div id="x"> <p itemprop="q" itemscope> <span itemprop="r">s</span> </p> </div>
201. # [06:38] <othermaciej> I don't think I am clear enough on the semantics of Microdata to help
202. # [06:39] <othermaciej> I imagine the problem is that the contents of div#x are being interpreted in two different ways?
203. # [06:41] <Hixie> yeah
204. # [06:41] <Hixie> not a problem for microdata, but when converting to rdf it causes an issue
205. # [06:41] <Hixie> because they need a URL for the properties
206. # [06:50] * Joins: takoratt_ (n=takoratt@203.39.247.241)
207. # [07:01] * Joins: zalan (n=zalan@213.28.64.66)
208. # [07:06] * Quits: takoratta (n=takoratt@203.39.247.241) (Read error: 110 (Connection timed out))
209. # [07:23] * Quits: dbaron (n=dbaron@c-98-234-51-190.hsd1.ca.comcast.net) ("8403864 bytes have been tenured, next gc will be global.")
210. # [07:25] * Quits: dglazkov (n=dglazkov@c-67-188-0-62.hsd1.ca.comcast.net)
211. # [07:48] * Joins: cardona507 (n=cardona5@c-67-180-160-250.hsd1.ca.comcast.net)
212. # [07:50] * Quits: jwalden (n=waldo@nat/mozilla/x-zanihpyatrwncefe) ("ChatZilla 0.9.85-rdmsoft [XULRunner 1.9.2/20100122095031]")
213. # [07:51] <Hixie> any rdf people here? Is there any meaning to the concept of a triple being stated twice?
214. # [07:52] <Hixie> e.g. if I write (in Turtle): <about:1> <about:2> <about:3> . <about:1> <about:2> <about:3> .
215. # [07:52] <Hixie> is that any different than just saying it once?
216. # [07:54] * Joins: breakmau5 (n=breakz@erft-5d80bd4e.pool.mediaWays.net)
217. # [07:58] * Quits: zalan (n=zalan@213.28.64.66) (Read error: 110 (Connection timed out))
218. # [07:59] <Lerc> The spec mentions the canvas dropshadow is Gaussian blur with defined SD. Does that mean compliant browsers shouldl be pixel identical? (not counting precision errors).
219. # [08:00] <Lerc> images made of dropshadows http://screamingduck.com/Lerc/jspic/
220. # [08:00] * Joins: maikmerten (n=merten@ls5dhcp196.cs.uni-dortmund.de)
221. # [08:00] <Hixie> not counting precisions errors, yes, more or less
222. # [08:02] <Lerc> I'm not sure who's wrong then but those pics render differently in FF and Safari.
223. # [08:02] <Lerc> Hard to tell because I'm generating the images in flash and its blur might be different again.
224. # [08:03] <Lerc> They make a funky testcase though :-)
225. # [08:04] <Hixie> heh
226. # [08:06] <Lerc> Next up is attempting JS+Canvas only Video.
227. # [08:10] * Quits: nessy (n=Adium@124-171-43-189.dyn.iinet.net.au) ("Leaving.")
228. # [08:21] * Joins: FireFly (n=firefly@unaffiliated/firefly)
229. # [08:21] * Joins: zalan (n=zalan@192.100.124.156)
230. # [08:23] * Joins: ment (i=thement@ibawizard.net)
231. # [08:28] <Hixie> foolip: if you're around, I'd be interested in your input on the edits to the microdata section
232. # [08:28] <Hixie> that i just generated
233. # [08:28] <Hixie> i can produce a diff before i check in if that would help
234. # [08:36] * Quits: virtuelv (n=virtuelv@162.179.251.212.customer.cdi.no) ("Ex-Chat")
235. # [08:41] <hsivonen> Where do people get this idea that HTML5 needs to make conforming everything that has ever been conforming?
236. # [08:41] <hsivonen> HTML 2.0 doctypes aren't conforming per 3.2
237. # [08:41] <hsivonen> and 2.0 and 3.2 doctypes aren't conforming per 4.0
238. # [08:49] * Joins: pesla (n=retep@procurios.xs4all.nl)
239. # [08:59] * Quits: breakmau5 (n=breakz@erft-5d80bd4e.pool.mediaWays.net) (Read error: 113 (No route to host))
240. # [09:00] <Hixie> hsivonen: HTML4 Transitional documents aren't conforming HTML4 Strict, even
241. # [09:04] <othermaciej> I think people are starting with an assumption that IETF mime type registration updates can never make previous content of that MIME type noncomforming, combined with the fact that HTML5 obsoletes the old registration and does not explicitly mention the old versions
242. # [09:04] <othermaciej> to conclude that
243. # [09:04] <othermaciej> (I do not know if there is really such an IETF rule or if it applies to the W3C's inline registrations)
244. # [09:04] <Hixie> there's a whole section on older versions
245. # [09:04] <Hixie> "History", in the introduction
246. # [09:05] <othermaciej> then it seems like the logic being applied is fallacious if it is along the lines I said
247. # [09:06] <Hixie> http://www.ietf.org/rfc/rfc2854 explicitly says it's not compatible with HTML2, even
248. # [09:06] <Hixie> and HTML3:
249. # [09:06] <hsivonen> Wouldn't it be sufficient to say something along the lines of: "The text/html content type may be used to label documents that conform to previous levels of HTML."
250. # [09:06] <Hixie> Note that [HTML20] included an optional "level" parameter; in
251. # [09:06] <Hixie> practice, this parameter was never used and has been removed from
252. # [09:06] <Hixie> this specification. [HTML30] also suggested a "version"
253. # [09:06] <Hixie> parameter; in practice, this parameter also was never used and has
254. # [09:06] <Hixie> been removed from this specification.
255. # [09:07] <hsivonen> s/previous levels/a previous level/
256. # [09:07] <Hixie> hsivonen: maybe. would that mean you can't write a validator unless you implement every version?
257. # [09:07] <Hixie> i think the spec already allows you to bail if the doctype is some older version's, no?
258. # [09:07] <Hixie> or did i remove that
259. # [09:08] <hsivonen> Hixie: at least the spec allows validators to go into a legacy checking mode
260. # [09:09] <hsivonen> Hixie: even if the MIME registration permitted HTML 2.0, I have no intention to support HTML 2.0 validation
261. # [09:09] <Hixie> holy crap, this is a must in 2854:
262. # [09:09] <Hixie> Implementors of text/html interpreters must be prepared to be
263. # [09:09] <Hixie> "bug-compatible" with popular browsers in order to work with many
264. # [09:09] <Hixie> HTML documents available the Internet.
265. # [09:10] <Hixie> you know, i really can't see anything in 2854 that is any different than what html5 does, really
266. # [09:10] <hsivonen> Hixie: how did an anti-competitive statement like that get into an RFC edited by L. Masinter?
267. # [09:10] <Hixie> so i don't understand the problem at all
268. # [09:10] <Hixie> hopefully bugs have been filed explaining what parts of 2854 i need to include to make things better
269. # [09:11] <Hixie> assuming 2854 is ok
270. # [09:11] <Hixie> which it presumably is
271. # [09:11] <othermaciej> Hixie: it would mean you couldn't write a validator for conformance to the text/html content type without implementing every version, but you could still write an HTML5 validator
272. # [09:18] <hsivonen> hmm. the full screen controls in http://jilion.com/sublime/video are different from the full window controls
273. # [09:18] <hsivonen> are the full screen controls Safari-native?
274. # [09:21] * Quits: Omnipotent (n=youshoul@unaffiliated/dor) (Connection timed out)
275. # [09:21] * Quits: cardona507 (n=cardona5@c-67-180-160-250.hsd1.ca.comcast.net) ("zzzzz")
276. # [09:22] * Joins: mhausenblas (n=mhausenb@wg1-nat.fwgal01.deri.ie)
277. # [09:22] <boblet> any webkit peeps here? tried on #webkit but no informative replies :)
278. # [09:22] <othermaciej> I think Larry has a point about overspecification being anti-competitive - it's true if you are more worried about protecting the 0.01% implementation from the 2% implementation than the 2% implementation from the 60% implementation
279. # [09:25] <hsivonen> I can see the point, but it seems to me that in practice, having undocumented but de facto necessary behaviors is more anti-competitive
280. # [09:25] <hsivonen> overspecifying the unnecessary could be anti-competitive
281. # [09:26] <hsivonen> but the thing is that whenever top 4 or even top 3 UAs have converged on a behavior, someone somewhere starts relying on it
282. # [09:26] <Hixie> othermaciej: actually, roc made a good point about that
283. # [09:26] * Joins: jacobolus (n=jacobolu@71.104.157.62)
284. # [09:26] <Hixie> othermaciej: which is that over-specification is harmless in practice
285. # [09:26] <Hixie> othermaciej: since if it's not required to compete, you can just ignore it
286. # [09:26] <othermaciej> of course, US antitrust law (and probably European too) would be more worried about a 60% vs 2% situation than a 2% vs 0.01% situation
287. # [09:27] <Hixie> othermaciej: and if it is, it helps the underdog
288. # [09:27] <Hixie> othermaciej: so in fact, it's better to overspecify
289. # [09:27] <hsivonen> Hixie: sometimes, it's really hard to tell which parts of HTML5 are known to be required to compete and which parts aren't
290. # [09:27] <othermaciej> Hixie: my underlying model is that the far underdog (as opposed to the still-significant underdog) does not have the resources to independently determine what parts of the spec are required
291. # [09:28] <hsivonen> Hixie: so one could argue you waste resources by implementing parts that aren't competitive requirements
292. # [09:28] <othermaciej> nor the resources to just implement the whole spec
293. # [09:28] <hsivonen> Hixie: and developing the knowledge of which parts are hard and which soft requirements for competitiveness partially comes back to the reverse engineering problem
294. # [09:28] <Hixie> i view the spec as being like a high resolution image
295. # [09:29] <Hixie> and an implementation as being like an approximation of that image
296. # [09:29] <hsivonen> although it's still better than just reverse engineering, because the spec gives you a hypothesis
297. # [09:29] <Hixie> now there are two ways to do this:
298. # [09:29] <Hixie> fill in each pixel one by one, or fill in the broad brush strokes and slowly improve each bit
299. # [09:29] <Hixie> if you do the former, then (a) you'll never be done and (b) over-specifying is a bad idea
300. # [09:29] <Hixie> if you do the other one, then you'll ship soon, and you'll immediately get feedback about which parts to try to paint in more detail
301. # [09:30] <othermaciej> I think Larry's incorrect premise is that the level of benefit to #300 vs #4 is as important as the level of benefit to #4 vs #1
302. # [09:30] <Hixie> and over time you approximate the spec in more and more detail
303. # [09:30] <Hixie> so i stand by roc's premise
304. # [09:30] <othermaciej> that's not how we tend to evaluate whether a market is competitive, at least in law, economics or common-sense understanding
305. # [09:31] <hsivonen> othermaciej: indeed, we'd be lucky (in terms of preserving freedom and choice) to have even a 5th major impl.
306. # [09:31] <othermaciej> well, we have a 4-and-a-halfth at least
307. # [09:31] <hsivonen> the and a half part fools people into thinking there are actually 5
308. # [09:32] <hsivonen> so that stuff that is WebKit only, gets promoted as Safari, Chrome and Chrome Frame and counted as 3
309. # [09:32] <hsivonen> s/,//
310. # [09:32] <othermaciej> -and-a-half probably overestimates the degree of difference actually
311. # [09:36] * Joins: borismus (n=borismus@bl4-167-3.dsl.telepac.pt)
312. # [09:36] <othermaciej> I think citing Safari, Chrome and Chrome Frame is only relevant in terms of telling content authors the size of their potential deployment target; definitely is not 3 separate implementations for standards purposes
313. # [09:36] <othermaciej> (nor do Safari on iPhone, Safari on iPad, Arora, Epiphany, OmniWeb, iCab, Sunrise Browser, etc all count as additional implementations)
314. # [09:38] * Joins: mat_t (n=mattomas@62-64-162-34.dynamic.dial.as9105.com)
315. # [09:39] * Joins: KevinMarks (n=KevinMar@c-67-161-44-219.hsd1.ca.comcast.net)
316. # [09:39] * Joins: nessy (n=Adium@203.39.247.242)
317. # [09:40] <Hixie> man, even the w3c wasn't this arrogant when we discussed the w3c and the whatwg working together on html5
318. # [09:40] <Hixie> the ietf people are making me have renewed respect for the w3c
319. # [09:40] <Hixie> how dare they!
320. # [09:42] <hsivonen> If IETF doesn't offer patent protection, what's the upside of taking the Web Socket protocol to the IETF?
321. # [09:42] <Hixie> none as far as i can tell
322. # [09:42] <hsivonen> does the W3C require the protocol to be an IETF spec in order to get the PP on the API?
323. # [09:43] <Hixie> (in theory more review)
324. # [09:43] <Hixie> the W3C does not require the protocol to be an IETF spec at all
325. # [09:43] <Hixie> it's only not in the W3C spec because the IETF whined to the W3C than the W3C were overstepping their agreement on who would work on what
326. # [09:43] <othermaciej> IETF requires patent disclosure at least (IIRC)
327. # [09:43] <Hixie> othermaciej: as far as i can tell it requires patent disclosure of patents that the contributors know about
328. # [09:44] <othermaciej> and minimum RAND licensing, though a group can ask for RF commitments
329. # [09:44] <Hixie> and i certainly don't know jack about google's patents
330. # [09:44] <othermaciej> I don't remember the details that clearly
331. # [09:44] * Joins: Utkarsh (n=admin@117.201.84.49)
332. # [09:44] <hsivonen> what's the IETF definition of RAND?
333. # [09:45] <othermaciej> it's somewhere in the process RFC
334. # [09:45] <hsivonen> interesting. seeing green on dev.w3.org: http://dev.w3.org/2006/webapi/WebTiming/
335. # [09:45] * Quits: nessy (n=Adium@203.39.247.242) ("Leaving.")
336. # [09:46] <Hixie> hsivonen: "RAND" is not free-software compatible, so the definition is moot
337. # [09:47] <hsivonen> Hixie: I know RAND is neither reasonable nor non-discriminatory. I was curious, though.
338. # [09:47] <Hixie> the relevant RFCs are linked to from every RFC's boilerplate
339. # [09:47] <Hixie> well. I say "linked to". But of course the IETF is stuck using text/plain, so I really mean "mentioned".
340. # [09:49] <Lachy> The problem with RAND is that what is considered reasonable is left entirely up to the patent holder, and the definition of non-discriminitory is that the fee applies equally to everyone, despite discriminating against those who can't be expected to afford the "reasonable" fee
341. # [09:50] <Hixie> the fee isn't the problem, when it comes to free software
342. # [09:50] <Hixie> but yeah
343. # [09:50] <othermaciej> I would make fun of them for using plaintext, but I do all my editing in emacs
344. # [09:50] <othermaciej> which I think folks on my team find equally laughable
345. # [09:50] <Hixie> othermaciej: me too, but i edit _html_ :-P
346. # [09:51] * hsivonen longs for the day when C++ can be edited the way Java is edited in Eclipse JDT
347. # [09:52] * Joins: tndH (n=Rob@cpc2-leed18-0-0-cust427.leed.cable.ntl.com)
348. # [09:52] <Hixie> does java not have macros?
349. # [09:52] <hsivonen> Hixie: it doesn't
350. # [09:52] <Hixie> i'd have thought that was the biggest problem with making "real" c++ IDEs
351. # [09:52] <hsivonen> Hixie: it is
352. # [09:52] <Hixie> k
353. # [09:52] <hsivonen> Hixie: templates, too
354. # [09:52] <hsivonen> Hixie: and typedefs
355. # [09:52] <Hixie> i thought java had both of those now
356. # [09:53] <hsivonen> Hixie: with Java, the semantic role of any piece of syntax is decidable from a sane grammar
357. # [09:53] <hsivonen> without having to process any include files
358. # [09:53] <hsivonen> Java has generics, which offer the benefits of casual templates
359. # [09:53] <Hixie> right
360. # [09:54] <hsivonen> but generics don't support all the crazy computation in the compile phase stuff
361. # [09:55] <hsivonen> yay. my C++ compile phase completed. now I can edit.
362. # [09:56] <othermaciej> to make a C++ IDE you have to parse all the C++ you are working with, but I don't think macros are a significantly larger part of the challenge than anything else
363. # [09:56] <Hixie> othermaciej: i would have thought that for an IDE's purposes, it'd be an issue because you'd have to keep track of what came from what
364. # [09:56] <hsivonen> incremental Java compiles as you edit are also awesome compared to waiting for a Firefox build
365. # [09:57] <Hixie> othermaciej: i.e. it's not like you can just quickly reparse the current bit and say "oh yeah, that's an identifier"
366. # [09:57] <othermaciej> the hard parts are aspects of the grammar that are ambiguous unless you know which identifiers are types, or the like
367. # [09:57] <Hixie> othermaciej: you have to keep track of everything and how it corresponded to macros and so on
368. # [09:57] <hsivonen> othermaciej: that part of C++ language design is insane
369. # [09:58] <othermaciej> Hixie: the preprocessor processing model is way simpler than parsing actual C++ itself
370. # [09:58] <Hixie> othermaciej: fair enough
371. # [09:59] <othermaciej> I have been using C++ professionally for something like 12 years now and I'm still occasionally learning things that totally surprise me
372. # [10:02] * Quits: syp_ (n=syp@lasigpc9.epfl.ch) (Read error: 54 (Connection reset by peer))
373. # [10:02] <hsivonen> Hixie: you were right that making streamless about:blank fire onload is a pain
374. # [10:02] <Hixie> heh
375. # [10:03] <hsivonen> all this onload firing machinery wants to see a stream-bound request
376. # [10:03] * Joins: syp (n=syp@lasigpc9.epfl.ch)
377. # [10:03] <Hixie> the problem in the spec would be making sure it doesn't fire twice
378. # [10:03] <Hixie> not really sure how to do that
379. # [10:03] <Hixie> since when you create the about:blank, you don't know that you're _not_ going to load a page
380. # [10:04] <Hixie> and when you don't load a page, well, you didn't load a page. There's nothing to hook onto to fire the load.
381. # [10:04] <hsivonen> Hixie: I put the about:blank creation on the code path that with any other URL kicks off the first actual load
382. # [10:05] <hsivonen> Hixie: and if the browser context hasn't had anything loaded to it and is about to load about:blank, I generate a document and return early
383. # [10:05] <Hixie> yeah i wish i could do that too, but the spec ended up too far from that to really do it sanely without a lot of tweaking
384. # [10:05] <hsivonen> invoking the machinery for DOMContentLoaded is done
385. # [10:05] <hsivonen> but when I try to fake onload, I keep hitting assertions that want to see a network request
386. # [10:06] <hsivonen> so I guess I'll end up faking one
387. # [10:08] <hsivonen> or maybe I'll make the loading machinery have a flag that tells it it's actually loading a fake about:blank
388. # [10:08] <hsivonen> special cases...
389. # [10:10] <hsivonen> Hixie: can XBL2 do evil things and inject scripts into about:blank synchronously with element creation?
390. # [10:14] <Hixie> assuming you inject enough stuff into the document that it has bindings at all, i guess so
391. # [10:14] <Hixie> about:blank isn't special in that sense
392. # [10:14] <hsivonen> Hixie: can you inject bindings into about:blank somehow?
393. # [10:14] <Hixie> sure, just inject a <link> to a style sheet or something
394. # [10:15] <Hixie> or loadBindingsDocument()
395. # [10:15] <Hixie> er, loadBindingDocument() even
396. # [10:15] <hsivonen> Hixie: How do you inject the <link> before the sync about:blank has already been created?
397. # [10:15] <Hixie> actually i guess a style sheet won't do it, that's always async
398. # [10:15] <Hixie> wait, _before_ about:blank is created?
399. # [10:16] <Hixie> about:blank is created as soon as you create the browsing context
400. # [10:16] <Hixie> nothing can premept that
401. # [10:16] <Hixie> preempt
402. # [10:16] <hsivonen> Hixie: in any case, having scripts run synchronously with element insertion to the DOM seems like huge badness
403. # [10:16] * Hixie points to mutation events
404. # [10:17] <hsivonen> Hixie: what's the sync/async story for bindings that haven't loaded by the time a bound element goes live?
405. # [10:17] <Hixie> (btw, you can bind elements that aren't even in the document)
406. # [10:17] <hsivonen> Hixie: mutation events don't fire for parser-inserted stuff
407. # [10:17] <Hixie> so, just for the record, i'm literally answering by reading teh spec here
408. # [10:17] <Hixie> wait, parser-created?
409. # [10:17] <Hixie> i thought we were talking about about:blank
410. # [10:17] <Hixie> i'm confused
411. # [10:18] <hsivonen> Hixie: isn't the fake about:blank conceptually parser-created as far as mutition event craziness goes?
412. # [10:18] <hsivonen> I guess I should read the whole XBL2 spec one of these days
413. # [10:18] <hsivonen> I saw some scary code pertaining to XBL1
414. # [10:18] * Joins: MikeSmith (n=MikeSmit@EM114-48-70-161.pool.e-mobile.ne.jp)
415. # [10:19] <Hixie> hsivonen: it's just created, there's no parser
416. # [10:19] <Hixie> hsivonen: i don't see how mutation events would be relevant here
417. # [10:19] <Hixie> hsivonen: there's no way to see the about:blank doc before it's completely ready
418. # [10:20] <Hixie> you can't load binding documents before about:Blank is done
419. # [10:20] <MikeSmith> what's the CSS selector syntax for saying, "element of class foo NOT followed by adjacent sibling of class bar"?
420. # [10:20] <Hixie> MikeSmith: .foo:not(:matches(#~.bar)) if my proposal goes through, but it's not possible currently
421. # [10:21] <MikeSmith> Hixie: so there is no way to express it currently?
422. # [10:21] <MikeSmith> oh
423. # [10:21] <MikeSmith> you just said that
424. # [10:21] <MikeSmith> sorry
425. # [10:21] <MikeSmith> well shit
426. # [10:21] <MikeSmith> that sucks
427. # [10:21] <Hixie> hsivonen: http://www.mozilla.org/projects/xbl/xbl2.html#binding is the key part
428. # [10:22] <Hixie> hsivonen: the few paragraphs up to the first example in that section
429. # [10:22] <Hixie> MikeSmith: it'd be a perf nightmare
430. # [10:22] <GarethAdams|Home> MikeSmith: you'd need one rule to match all .foo and then another overriding .foo + .bar
431. # [10:22] <GarethAdams|Home> at the moment
432. # [10:22] * Joins: cpearce_ (n=cpearce@ip-118-90-91-135.xdsl.xnet.co.nz)
433. # [10:22] <Hixie> GarethAdams|Home: nah he wants to style the .foo, not the .bar
434. # [10:23] <GarethAdams|Home> of course, and my idea was broken in any case
435. # [10:23] <hsivonen> Hixie: "If the binding document has yet to be (fully) loaded when it becomes known that the binding applies, then the user agent must wait until all running scripts have completed before attaching the binding." looks like a huge can of worms to me
436. # [10:23] <GarethAdams|Home> it's too early
437. # [10:24] <hsivonen> Hixie: oops. never mind
438. # [10:24] * Quits: cpearce (n=cpearce@ip-118-90-40-252.xdsl.xnet.co.nz) (Read error: 60 (Operation timed out))
439. # [10:24] * cpearce_ is now known as cpearce
440. # [10:25] <Hixie> hsivonen: that's just saying (in terms that predate the html5's event loop spec) to queue a task instead of preempting a script
441. # [10:25] <Hixie> hsivonen: to keep things predictable
442. # [10:25] <hsivonen> Hixie: am I reading this correctly that binding loads race with script loads?
443. # [10:26] <Hixie> yes
444. # [10:26] <hsivonen> fun
445. # [10:26] <Hixie> i was young, what can i say
446. # [10:26] <hsivonen> but not making it so would be painful, I guess
447. # [10:26] <Hixie> not sure how else to do it, yeah
448. # [10:26] <othermaciej> xbl2 is still young too
449. # [10:26] <Hixie> short of doing what we do with css
450. # [10:26] <othermaciej> at least in implementation years
451. # [10:26] <Hixie> which is painful too
452. # [10:27] <Hixie> othermaciej: yeah sometime this year i'll take the feedback collected and do another CR
453. # [10:27] <hsivonen> I guess blocking the parser on inline scripts if there are pending bindings would be doable
454. # [10:27] <Hixie> othermaciej: (sometime after sicking sends some more feedback)
455. # [10:28] <hsivonen> also, it looks like a sane XBL2 impl. is going to continue to need the Gecko scriptrunner facility
456. # [10:28] <othermaciej> I hope we get a chance to start on it soon but we only have bandwidth for so many architecture-level changes at a time
457. # [10:28] <Hixie> (i have 52 e-mails in the XBL folder)
458. # [10:28] <hsivonen> so I guess we aren't getting rid of that
459. # [10:28] * Joins: virtuelv (n=virtuelv@pat-tdc.opera.com)
460. # [10:28] * Quits: GarethAdams|Home (n=GarethAd@pdpc/supporter/active/GarethAdams)
461. # [10:28] <Hixie> othermaciej: yeah, it's just not as high priority as a lot of the other new stuff
462. # [10:28] <othermaciej> after our current semi-stealth one the next big change will probably be one of HTML5 parsing, XBL2, or refactoring the render tree
463. # [10:28] * Joins: svl (n=me@ip565744a7.direct-adsl.nl)
464. # [10:28] <Hixie> i vote for parsing
465. # [10:29] * hsivonen wonders if other engines have a facility to queue tasks that go into a local task queue that is executed at a safe point between now and the next event loops spin
466. # [10:29] <othermaciej> but more than any of those things I want to see implementation of more HTML5 elements and form controls
467. # [10:30] <hsivonen> I'd love to see new form controls in Gecko *and* a story for styling them
468. # [10:30] <Hixie> hsivonen: you mean like the spec's "synchronous section"?
469. # [10:30] <hsivonen> Hixie: XBL2 spec's?
470. # [10:30] <hsivonen> oh HTML5's
471. # [10:30] <Hixie> yeah, html5
472. # [10:31] <hsivonen> Hixie: it says: "A synchronous section never mutates the DOM, runs any script, or has any other side-effects."
473. # [10:31] <hsivonen> so no
474. # [10:31] <Hixie> ah
475. # [10:32] <hsivonen> the whole point of scriptrunners is that they can do that kind of scary stuff
476. # [10:32] <Hixie> man, i fixed both session history's async hairball and the microdata sub-vocab-in-rdf issue today. i'm on a roll.
477. # [10:32] <Hixie> i hope this is indicative of a renewed level of productivity
478. # [10:33] <Hixie> hsivonen: right now those are used to do async stuff that needs to inspect the DOM without being racy against running scripts
479. # [10:33] <Hixie> hsivonen: basically it's a synchronisation section for other threads
480. # [10:34] <Hixie> at some point i might have to add (e.g. for xbl) a new thing about the same place, like a high-priority queue or something
481. # [10:34] <Hixie> so far i've avoided it
482. # [10:35] * Joins: zcorpan (n=zcorpan@90-224-190-71-no135.tbcn.telia.com)
483. # [10:35] <Hixie> ok, w3c microdata editorial stuff, let's see
484. # [10:35] * Quits: Rik` (n=Rik`@pha75-2-81-57-187-57.fbx.proxad.net) (Read error: 113 (No route to host))
485. # [10:35] * Joins: Rik`_ (n=Rik`@pha75-2-81-57-187-57.fbx.proxad.net)
486. # [10:35] <hsivonen> the idea of scriptrunners is that they look synchronous from the POV of scripts but are deferred from the POV of C++
487. # [10:35] * Rik`_ is now known as Rik`
488. # [10:38] * Joins: ROBOd (n=robod@89.122.216.38)
489. # [10:38] <Hixie> othermaciej: any suggestions for how i should describe microdata and 2dcontext in terms of "context and rationale"?
490. # [10:38] <Hixie> (it's one of the things i have to put in the status-of-this-document section)
491. # [10:39] <othermaciej> Hixie: can you give an example of text along those lines in another document's sotd for a point of reference?
492. # [10:39] <othermaciej> I am not very familiar with the sotd requirements
493. # [10:39] <Hixie> it's the paragraph starting "This specification is a module that forms part of the HTML5 series" in the microdata spec currently
494. # [10:41] <othermaciej> I'm checking what Web Storage says in the equivalent place
495. # [10:42] <othermaciej> I don't see any text analogous to that in Web Storage, it goes straight from the responsible Working Group paragraph to the patent policy paragraph
496. # [10:43] <Hixie> uh, yes indeedy, i appear to have omitted the context section in that spec
497. # [10:43] <othermaciej> Workers says "This specification is intended to specify a part of the Web platform closely related to HTML5. It is defined in a separate document primarily to ease the cognitive load on reviewers."
498. # [10:44] <othermaciej> WebSocket says "This specification is being developed in conjunction with an Internet Draft for a wire protocol, the Web Socket Protocol, available from the IETF at the following location"
499. # [10:44] <Hixie> i can use the web worker text if you like
500. # [10:44] <othermaciej> I would say something like "This specification was developed in tandem with HTML5 and is intended to be used in conjunction with it"
501. # [10:45] <othermaciej> or something like the Workers spec, though maybe people would get irked at "closely related"
502. # [10:45] <Hixie> that works
503. # [10:45] <othermaciej> the sentence I suggested seems more fact-based and so hopefully less controversial
504. # [10:45] <Hixie> tis in
505. # [10:45] <Hixie> though if you can predict what's going to be controversial, you're a better man than i
506. # [10:46] <othermaciej> sometimes I am able to guess
507. # [10:46] <othermaciej> certainly not always
508. # [10:46] <othermaciej> signing off for a bit to debug something, will be back
509. # [10:47] * Joins: breakmau5 (n=breakz@erft-5d80bd4e.pool.mediaWays.net)
510. # [10:47] * Quits: othermaciej (n=mjs@c-69-181-42-237.hsd1.ca.comcast.net) (Read error: 104 (Connection reset by peer))
511. # [10:48] * Joins: othermaciej (n=mjs@c-69-181-42-237.hsd1.ca.comcast.net)
512. # [10:51] <hsivonen> figcaption, etc., pale in comparison when you consider that the token for denoting a person's name is called fn
513. # [10:53] * Quits: Utkarsh (n=admin@117.201.84.49) (Read error: 110 (Connection timed out))
514. # [10:54] * jgraham always assumes fn means "file name"
515. # [10:54] <othermaciej> I assume it means "function"
516. # [10:55] <gsnedders> fullname
517. # [10:55] <othermaciej> or that it's a symbol for the n-th fibonacci number
518. # [10:56] <Philip`> Or an abbreviation of fnord
519. # [10:56] <othermaciej> an abbreviation of what?
520. # [10:56] <othermaciej> your sentence got cut off it looks like
521. # [10:59] * Quits: tkent (n=tkent@220.109.219.244) ("Leaving...")
522. # [11:05] <Hixie> othermaciej: so... is it ok to refer to the HTML5 spec at all? Or...?
523. # [11:05] <Hixie> othermaciej: i'm confused about how to handle all the cross-references
524. # [11:05] <Hixie> like the references to "boolean attributes", or "HTML element", or "applicable specification"
525. # [11:05] <othermaciej> Hixie: it is ok to normatively reference the HTML5 spec
526. # [11:05] <Hixie> k
527. # [11:06] <othermaciej> not ok to claim to be part of it
528. # [11:06] <othermaciej> you can look at HTML+RDFa as an example of ways in which it is apparently uncontroversial to reference HTML5
529. # [11:07] <othermaciej> its context paragraph is "This specification is an extension to the HTML5 language. All normative content in the HTML5 specification, unless specifically overridden by this specification, is intended to be the basis for this specification."
530. # [11:07] <Hixie> k
531. # [11:07] <Hixie> should i just use that?
532. # [11:07] <othermaciej> that is actually a well-worded paragraph IMO
533. # [11:14] * Quits: broquaint (i=09ea70f4@spc2-brig11-0-0-cust103.asfd.cable.virginmedia.com) (Remote closed the connection)
534. # [11:14] <Hixie> k
535. # [11:18] * Joins: JonathanNeal_oww (n=Jonathan@76-219-69-134.lightspeed.breaca.sbcglobal.net)
536. # [11:19] * Quits: MikeSmith (n=MikeSmit@EM114-48-70-161.pool.e-mobile.ne.jp) ("Tomorrow to fresh woods, and pastures new.")
537. # [11:19] * Quits: JonathanNeal_oww (n=Jonathan@76-219-69-134.lightspeed.breaca.sbcglobal.net) (Client Quit)
538. # [11:20] * Joins: JonathanNeal_oww (n=Jonathan@76-219-69-134.lightspeed.breaca.sbcglobal.net)
539. # [11:21] * Quits: JonathanNeal (n=Jonathan@99-59-124-67.lightspeed.irvnca.sbcglobal.net) (Nick collision from services.)
540. # [11:24] * Joins: Maurice (i=copyman@5ED548D4.cable.ziggo.nl)
541. # [11:25] * Quits: erikvold (n=erikvold@S01060024012860e9.gv.shawcable.net) (Excess Flood)
542. # [11:28] <Hixie> othermaciej: how much of http://www.whatwg.org/specs/web-apps/current-work/multipage/infrastructure.html#conformance-requirements (section 2.2 and its two subsections) should i have in the microdata spec? or should i just reference that too?
543. # [11:29] <Hixie> i'm referencing most of the rest of the 2.x sections
544. # [11:30] <othermaciej> Hixie: my own judgment would be that Microdata should probably have most or all the bits besides the definitions of conformance classes, but could just cite the conformance class definitions
545. # [11:30] <othermaciej> at the very least I think it would be good to cite RFC2119 directly instead of via an indirect reference
546. # [11:30] <Hixie> k
547. # [11:31] <Hixie> let's see if i can do that without duplicating the text
548. # [11:31] <Hixie> this ought to be fun
549. # [11:32] * hsivonen wonders if the document loading machinery really needs to be this complex
550. # [11:36] <hsivonen> this stuff dates back all the way to 1998
551. # [11:37] * Quits: borismus (n=borismus@bl4-167-3.dsl.telepac.pt) (Read error: 110 (Connection timed out))
552. # [11:37] <Hixie> hsivonen: then the answer is almost certainly "no"
553. # [11:42] * Quits: pablof (n=palbo@pat-tdc.opera.com) (K-lined)
554. # [11:42] * Quits: virtuelv (n=virtuelv@pat-tdc.opera.com) (K-lined)
555. # [11:49] <Hixie> um
556. # [11:49] <Hixie> opera got k-lined?
557. # [11:49] * Joins: sbublava (n=stephan@77.116.213.114.wireless.dyn.drei.com)
558. # [11:50] <othermaciej> that's... mysterious
559. # [11:51] * jgraham is still here
560. # [11:51] <Hixie> you're not going through the office nat
561. # [11:52] <jgraham> True :)
562. # [11:54] <Hixie> i have a new feature request for gsnedders
563. # [11:54] <Hixie> sanitising the <hx> to fit the outline
564. # [11:54] * jgraham wonders what Hixie means by "sanitizing"
565. # [11:55] <Hixie> turning <h2>...</h2> <h5>...</h5> into <h2> and <h3> respectively
566. # [11:55] <Hixie> or just out-and-out supporting <section>
567. # [11:55] * Joins: nessy (n=Adium@124-171-43-189.dyn.iinet.net.au)
568. # [11:55] <Hixie> and outputting appropriately numbered <hx>s
569. # [11:58] * Quits: Lachy (n=Lachlan@85.196.122.246) ("This computer has gone to sleep")
570. # [11:58] * Joins: myakura (n=myakura@p3213-ipbf4202marunouchi.tokyo.ocn.ne.jp)
571. # [12:01] * Quits: gavin_ (n=gavin@firefox/developer/gavin) (Read error: 110 (Connection timed out))
572. # [12:02] * Joins: gavin_ (n=gavin@firefox/developer/gavin)
573. # [12:06] * Joins: erikvold (n=erikvold@S01060024012860e9.gv.shawcable.net)
574. # [12:16] * Quits: tndH (n=Rob@cpc2-leed18-0-0-cust427.leed.cable.ntl.com) ("ChatZilla 0.9.86-rdmsoft [XULRunner 1.9.0.1/2008072406]")
575. # [12:18] <gsnedders> Hixie: There already is such a thing, written for lac
576. # [12:19] <Hixie> ah
577. # [12:19] <gsnedders> (with section, that is)
578. # [12:19] <gsnedders> Dunno how brilliant it is
579. # [12:19] <Hixie> then i guess my suggestion is to jgraham to enable it?
580. # [12:20] <gsnedders> Hixie: You just need to send whatever option to PMS
581. # [12:20] <Hixie> ah ok
582. # [12:20] <Hixie> guess i should look into that
583. # [12:20] <Hixie> :-)
584. # [12:22] <gsnedders> (where lac is a failed attempt to autocomplete Lachy)
585. # [12:23] <gsnedders> http://googleblog.blogspot.com/2010/01/unicode-nearing-50-of-web.html
586. # [12:23] * Quits: Rik` (n=Rik`@pha75-2-81-57-187-57.fbx.proxad.net) (No route to host)
587. # [12:30] <foolip> Hixie: I'll have a look at the recent microdata edits over the weekend
588. # [12:30] <Hixie> ;cool
589. # [12:33] * Quits: syp (n=syp@lasigpc9.epfl.ch) (Read error: 104 (Connection reset by peer))
590. # [12:33] * Joins: syp (n=syp@lasigpc9.epfl.ch)
591. # [12:35] * Quits: zcorpan (n=zcorpan@90-224-190-71-no135.tbcn.telia.com) (Remote closed the connection)
592. # [12:35] * Joins: zcorpan (n=zcorpan@90-224-190-71-no135.tbcn.telia.com)
593. # [12:36] * Joins: ttepasse (n=ttepas--@ip-95-222-120-117.unitymediagroup.de)
594. # [12:53] * Joins: Utkarsh (n=admin@117.201.88.246)
595. # [12:53] * Joins: Lachy (n=Lachlan@london.perfect-privacy.com)
596. # [12:55] * Quits: cedric_ (n=cedric@124.197.108.103) (Client Quit)
597. # [13:05] * Quits: pmuellr (n=pmuellr@user-0c99bjg.cable.mindspring.com)
598. # [13:06] * Quits: ttepasse (n=ttepas--@ip-95-222-120-117.unitymediagroup.de) ("404")
599. # [13:07] * Joins: ttepasse (n=ttepas--@ip-95-222-120-117.unitymediagroup.de)
600. # [13:07] <hsivonen> I signed up to give a lecture about HTML5 again
601. # [13:08] <hsivonen> I wonder what adjustments I should make to last year's lecture except mentioning Microdata
602. # [13:08] <hsivonen> And mentioning that instead of spec vs. buzzword there's now a class of things that used to be in the spec but have been split out
603. # [13:09] * Quits: takoratt_ (n=takoratt@203.39.247.241) (Read error: 110 (Connection timed out))
604. # [13:09] * Joins: takoratta (n=takoratt@202.171.161.131)
605. # [13:20] <Dashiva> hsivonen: Examples of progress in implementation?
606. # [13:21] <hsivonen> Dashiva: yeah, I should update my screenshots to show Firefox instead of Minefield, etc.
607. # [13:22] <Dashiva> You could mention the @autobuffer case as an example of how people can get involved and effect change and improvement :)
608. # [13:23] <hsivonen> Dashiva: well, it's not a great example, because gruber didn't bother emailing the WG.
609. # [13:25] <foolip> also, nothing has actually happened spec-wise with it yet (unfortunately)
610. # [13:26] <Hixie> gruber said he would send in his feedback
611. # [13:26] * Hixie mails him to ask him when
612. # [13:26] * Joins: Omnipotent (n=youshoul@unaffiliated/dor)
613. # [13:26] <Hixie> i guess i should do some video feedback sooner rather than later
614. # [13:27] <Philip`> I guess it's a good example of how the people writing the specs take account of feedback for authors, even if it's a bad example of how authors should give feedback to the people writing the specs
615. # [13:27] <hsivonen> also, it's not clear yet if the change effected (if effected) is an improvement :-)
616. # [13:27] <Philip`> s/for/from/
617. # [13:27] <danbri> you could mention changes in the browser landscape (faster .js etc...)?
618. # [13:27] <foolip> yes please :) we are holding off supporting autobuffer (or anything like it) until that issue is resolved
619. # [13:28] <Hixie> ok it's on my list to do after the websocket feedback
620. # [13:28] <Hixie> which is my plan for next week
621. # [13:28] <foolip> ok, no panic on this end, sounds good
622. # [13:29] <Hixie> right, i should go sleep
623. # [13:30] * boog|afk is now known as boogyman
624. # [13:32] <Hixie> nn
625. # [13:32] * Parts: zcorpan (n=zcorpan@90-224-190-71-no135.tbcn.telia.com)
626. # [13:44] * Quits: Utkarsh (n=admin@117.201.88.246) (Read error: 104 (Connection reset by peer))
627. # [13:51] * Joins: Utkarsh (n=admin@117.201.89.88)
628. # [14:07] * Joins: pmuellr (n=pmuellr@nat/ibm/x-fxmgctbtvhaqmwub)
629. # [14:09] * Quits: nattokirai (n=nattokir@118.243.226.86)
630. # [14:16] * Quits: gavin_ (n=gavin@firefox/developer/gavin) (Read error: 60 (Operation timed out))
631. # [14:16] * Joins: gavin_ (n=gavin@firefox/developer/gavin)
632. # [14:18] * Quits: nessy (n=Adium@124-171-43-189.dyn.iinet.net.au) ("Leaving.")
633. # [14:23] * Quits: taf2_ (n=taf2@pool-98-117-216-229.bltmmd.fios.verizon.net)
634. # [14:45] * Quits: takoratta (n=takoratt@202.171.161.131) (Read error: 110 (Connection timed out))
635. # [14:50] * Quits: Utkarsh (n=admin@117.201.89.88) (Read error: 60 (Operation timed out))
636. # [14:50] * Joins: Utkarsh (n=admin@117.201.82.208)
637. # [14:55] * Quits: jacobolus (n=jacobolu@71.104.157.62) (Read error: 110 (Connection timed out))
638. # [14:57] * Quits: riven (n=riven@53518387.cable.casema.nl) (Excess Flood)
639. # [14:57] * Quits: erikvold (n=erikvold@S01060024012860e9.gv.shawcable.net) (Excess Flood)
640. # [14:57] * Joins: riven (n=riven@53518387.cable.casema.nl)
641. # [15:06] * Quits: beilabs (n=beilabs@ppp121-44-34-85.lns20.syd6.internode.on.net) (Remote closed the connection)
642. # [15:16] * Joins: miketaylr (n=miketayl@38.117.156.163)
643. # [15:19] * Quits: Yudai (n=Yudai@p78be59.kngwnt01.ap.so-net.ne.jp) ("SIGTERM received; exit")
644. # [15:19] * Joins: Yudai (n=Yudai@p78be59.kngwnt01.ap.so-net.ne.jp)
645. # [15:36] * Quits: starjive (i=beos@81-233-16-19-no30.tbcn.telia.com) (Client Quit)
646. # [15:37] * Joins: BlurstOfTimes (n=blurstof@168.203.117.66)
647. # [15:48] * Joins: annevk (n=annevk@5355737B.cable.casema.nl)
648. # [15:49] <annevk> ah yes, fetching >4000 emails
649. # [15:49] <annevk> joy
650. # [15:49] <jgraham> Welcome home :)
651. # [15:50] <jgraham> I recommend ignoring all the emails, most of them are boring and anything important enough to worry about will probably get resent anyway
652. # [15:52] * Quits: Lachy (n=Lachlan@london.perfect-privacy.com) ("Leaving")
653. # [15:54] <annevk> sounds like a plan
654. # [15:54] <annevk> about 350 are spam apparently
655. # [15:56] * Quits: maikmerten (n=merten@ls5dhcp196.cs.uni-dortmund.de) (Remote closed the connection)
656. # [15:56] <TabAtkins> How long were you gone, Anne?
657. # [15:56] <TabAtkins> 350 sounds like a low spam count to me.
658. # [15:56] <TabAtkins> Or high, if they actually made it to your inbox.
659. # [15:57] <annevk> about three weeks
660. # [15:59] * Joins: Lachy (n=Lachlan@london.perfect-privacy.com)
661. # [15:59] * Quits: paul_irish (n=paul_iri@c-71-192-163-128.hsd1.nh.comcast.net) ("Leaving...")
662. # [16:00] <Dashiva> "Since previously, HTML content that invoked quirks mode *was* conforming, we can't make it non-conforming."
663. # [16:00] <Dashiva> Can't we?
664. # [16:01] <Philip`> Of course not, because that's not allowed
665. # [16:02] <Dashiva> Not allowed by... the axiomatic proof?
666. # [16:03] <Philip`> If it was allowed, then we could make old content non-conforming, which we can't do because it's not allowed
667. # [16:03] <Philip`> therefore we can't allow it
668. # [16:03] <Philip`> Proof by contradiction
669. # [16:04] * Quits: wakaba_ (n=wakaba_@122x221x184x68.ap122.ftth.ucom.ne.jp) ("Leaving...")
670. # [16:05] <Dashiva> If it's not proof by axiom, it isn't valid
671. # [16:05] <annevk> I think that statement follows from the imo misguided way media types are drafted to work
672. # [16:05] * Joins: Lachy_ (n=Lachlan@london.perfect-privacy.com)
673. # [16:06] <annevk> btw, I noticed data URLs were discussed at some point here
674. # [16:06] <annevk> they do have fragment identifiers
675. # [16:06] <Dashiva> Only in Opera, apparently
676. # [16:06] <annevk> fragment identifiers depend on the media type, not URI scheme
677. # [16:06] <annevk> well, Firefox is buggy
678. # [16:06] <jgraham> And webkit
679. # [16:06] <annevk> also buggy
680. # [16:06] <Philip`> Maybe it's actually a (sensible) desire to guarantee backward compatibility within media types, but twisted by a (incorrect) notion that content conformance and interoperable operation are tied together
681. # [16:07] <annevk> see RFC 3986
682. # [16:07] <Lachy_> Dashiva, people who claim things like that need to realise that a new specification making things non-conformng with respect to that particular specification, doesn't make those things non-conforming with respect to other specifications.
683. # [16:07] <Dashiva> Lachy: I think the complaint is partially that changing the registration would make all text/html be considered HTML5
684. # [16:08] <Dashiva> So you wouldn't be able to serve HTML<5 at all
685. # [16:09] * Quits: Lachy (n=Lachlan@london.perfect-privacy.com) (Read error: 104 (Connection reset by peer))
686. # [16:09] <Lachy_> that's nonsense. You can serve it. It's just non-conforming HTML5. But it may still be conforming with the older specifications, despite those older specifications being obsolete
687. # [16:10] <Philip`> You can only serve it as text/html, and it's non-conforming text/html
688. # [16:10] <Philip`> (since the new registration information for text/html would supercede the old ones)
689. # [16:11] * jgraham wonders where all the people fretting about whether their HTML2 is valid to send as text/html are
690. # [16:11] <jgraham> Actually I know where they are, they are all on public-html
691. # [16:13] * Quits: BlurstOfTimes (n=blurstof@168.203.117.66) (Client Quit)
692. # [16:15] * Joins: zcorpan (n=zcorpan@90-224-190-71-no135.tbcn.telia.com)
693. # [16:16] * Joins: jgornick (n=joe@199.199.210.66)
694. # [16:17] <Dashiva> jgraham: I think they mostly live over at IETF
695. # [16:18] <jgraham> Hmm maybe
696. # [16:18] <jgraham> There seem to be a number on public-html as well though
697. # [16:19] <jgraham> A number that I can count on my fingers admittedly
698. # [16:19] <Dashiva> They are possibly just fighting on behalf of others, not themselves
699. # [16:21] <jgraham> Who are these others?
700. # [16:21] * Joins: BlurstOfTimes (n=blurstof@168.203.117.66)
701. # [16:21] <Lachy_> the others don't like to reveal themselves. They're a secret organisation.
702. # [16:22] <Dashiva> But they are many
703. # [16:23] <gsnedders> Lachy_: It wouldn't be conforming with the older specifications because the text/html registration would only allow HTML 5.
704. # [16:24] <Lachy_> the text/html registration is irrelevant to the question of whether or not somthing conforms with an older edition of HTML
705. # [16:25] * Lachy_ is now known as Lachy
706. # [16:25] <jgraham> Anyway the point is this is an issue so far removed from anything that real people care about that it isn't even worth the effort of discussing
707. # [16:25] <Dashiva> You said "real people"
708. # [16:25] <gsnedders> Lachy: It's a relevant question to whether the content of the resource representation of http://example.com/ which has a content-type of text/html is a valid response.
709. # [16:26] <gsnedders> Lachy: As if the text/html registration says it must be HTML 5 and it is HTML 3.2 then the response is invalid
710. # [16:26] * Quits: gavin_ (n=gavin@firefox/developer/gavin) (Read error: 60 (Operation timed out))
711. # [16:26] <Dashiva> example.com uses utf-8, it's obviously fake
712. # [16:27] <Lachy> gsnedders, yes. The response may be invalid per the text/html registration. But the file is still conforming HTML 3.2 (if the DOCTYPE was optional in that spec, which I think it was)
713. # [16:28] <gsnedders> the _file_ is, but not the response.
714. # [16:28] * Joins: gavin_ (n=gavin@firefox/developer/gavin)
715. # [16:28] <gsnedders> And this is a concern about the validity of the response, not about the file itself.
716. # [16:28] <Lachy> but regardless of that, what's the problem with the response being non-conforming now?
717. # [16:28] <jgraham> People, you are still discussing this
718. # [16:28] <Lachy> jgraham, sorry
719. # [16:28] <Lachy> do you have an alternative topic?
720. # [16:28] <Philip`> It's fun and relaxing to discuss things that don't actually matter
721. # [16:28] * Quits: zcorpan (n=zcorpan@90-224-190-71-no135.tbcn.telia.com) (Remote closed the connection)
722. # [16:29] <jgraham> There must be more fun things that don't actually matter
723. # [16:29] <Philip`> because you don't have to worry about being correct or about the consequences of your statements and decisions
724. # [16:29] * Joins: zcorpan (n=zcorpan@90-224-190-71-no135.tbcn.telia.com)
725. # [16:29] <Philip`> What could be more fun than spec pedantry?
726. # [16:29] <Dashiva> Charter pedantry, obviously
727. # [16:29] <jgraham> And also you risk giving people the impression that things do actually matter
728. # [16:30] <jgraham> Like the people who get worked up about Star Wars pedantry, or whatever
729. # [16:30] <gsnedders> jgraham: Harry Potter pedantry?
730. # [16:30] <gsnedders> I've never known you to pedant that.
731. # [16:30] <jgraham> I'm not a Harry Potter pedant, just a fanboi
732. # [16:30] <Philip`> Like how the Death Star explosion is scientifically proven to have wiped out all the ewoks?
733. # [16:31] <jgraham> Yes
734. # [16:32] <Philip`> That's not pedantry, that's genocide
735. # [16:33] <Dashiva> What about pedantry about whether something constitutes pedantry or not?
736. # [16:34] * Quits: zalan (n=zalan@192.100.124.156) (Read error: 110 (Connection timed out))
737. # [16:34] <jgraham> That's meta-pedantry
738. # [16:35] <jgraham> Which isw quite longwinded to say and should be sortened to metandry
739. # [16:35] <jgraham> erg
740. # [16:35] <jgraham> With more of the original letters
741. # [16:36] <Philip`> You can't just mix Greek and Latin like that :-(
742. # [16:37] <TabAtkins> Television says you can, and I believe television.
743. # [16:38] * Joins: mamund (i=mamund@frost.nullshells.net)
744. # [16:38] <Dashiva> Truth in Television (literally)
745. # [16:38] <Dashiva> (or should that be visually)
746. # [16:39] <TabAtkins> It'd go with literally.
747. # [16:39] <TabAtkins> s/It'd/I'd/
748. # [16:43] <Dashiva> "I've compromised some on my position, that means the rest of the compromise has to come from you"
749. # [16:43] * Joins: dglazkov (n=dglazkov@c-67-188-0-62.hsd1.ca.comcast.net)
750. # [16:45] <TabAtkins> And that sort of idea is precisely why you can win by starting from a position more extreme than what you actually want.
751. # [16:50] * Quits: m_W (n=mwj@c-69-141-106-205.hsd1.nj.comcast.net) (Read error: 113 (No route to host))
752. # [17:00] * Joins: m_W (n=mwj@c-69-141-106-205.hsd1.nj.comcast.net)
753. # [17:02] <AryehGregor> Hmph. "Fragment identifier semantics are independent of the URI scheme and thus cannot be redefined by scheme specifications."
754. # [17:03] <AryehGregor> I guess I have to memorize the percent-encoding for # now for use in data URIs.
755. # [17:03] <AryehGregor> Note to self: %23.
756. # [17:04] <zcorpan> and %25 for %
757. # [17:04] <zcorpan> iirc
758. # [17:05] <TabAtkins> Oh, hrm.
759. # [17:05] * Quits: breakmau5 (n=breakz@erft-5d80bd4e.pool.mediaWays.net) (Read error: 60 (Operation timed out))
760. # [17:05] <TabAtkins> That means that trying to use htmlspecialcharacters() will fail rarely but badly.
761. # [17:05] <AryehGregor> Why would you ever use htmlspecialchars() on a data: URL?
762. # [17:06] <AryehGregor> Well, except insofar as you have to on any URL included in an HTML document, I guess.
763. # [17:06] <AryehGregor> I don't know why it would break in that light.
764. # [17:06] * Quits: zcorpan (n=zcorpan@90-224-190-71-no135.tbcn.telia.com)
765. # [17:06] <Lachy> wow, apparently there are XBL2 implementations now, used as part of some XForms implementations
766. # [17:06] <AryehGregor> There are XForms implementations?
767. # [17:06] <Lachy> haha
768. # [17:07] <Lachy> this is one, apparently http://wiki.orbeon.com/forms/doc/developer-guide/xbl-components-guide#TOC-Why-use-XBL-and-not-simply-XSLT-
769. # [17:08] <Dashiva> I wonder if there's non-trivial content out there depending on # being valid in data URIs
770. # [17:08] <TabAtkins> AryehGregor: Because you're dumb and don't know how to escape things properly.
771. # [17:08] <AryehGregor> TabAtkins, well, obviously if you double-escape with htmlspecialchars(), it will break. Or if you only escape with that and not rawurlencode() also.
772. # [17:08] <TabAtkins> The latter is my intention.
773. # [17:09] <Lachy> Dashiva, it is valid in data URIs. data:text/html,<p id="test">foo</p>#test
774. # [17:09] <Lachy> unfortunately, Firefox's implementation is broken
775. # [17:09] <workmad3> AryehGregor: people escape with the wrong things all the time... htmlspeciarchars on a uri is a bit of a biggie, but CGI escaping a url is quite common :)
776. # [17:09] <Dashiva> Lachy: As a data character, not fragment separator
777. # [17:09] <TabAtkins> Since I think people are at least somewhat familiar with htmlspecialchars, while browsers tend to be pretty lenient with urls and won't often break things if you forget to urlencode it.
778. # [17:09] <AryehGregor> Lachy, and WebKit's.
779. # [17:09] <Lachy> oh :-(
780. # [17:10] <AryehGregor> Lachy, only Opera actually interprets # in data URLs per spec AFAIK.
781. # [17:10] <AryehGregor> (haven't tested IE8)
782. # [17:10] <Dashiva> I can test if anyone has a testcase
783. # [17:10] <AryehGregor> That doesn't allow HTML documents anyway in data: URLs, right?
784. # [17:10] <gsnedders> AryehGregor: Changed in 10.50 I'm pretty sure
785. # [17:10] <AryehGregor> So I'm not sure how you'd test.
786. # [17:10] <AryehGregor> gsnedders, changed which way?
787. # [17:11] <Lachy> IE's data URL implementation is totally broken
788. # [17:12] <AryehGregor> On another note, how many months' worth of spam e-mails from "Approved VIAGRA® Store" do I need to mark as spam before Gmail takes the hint?
789. # [17:13] <Lachy> if you try to load a data URL, it gives an error. If you try to load it into an iframe, it navigates the whole page to the data URL, and fails to load it
790. # [17:13] * Quits: dglazkov (n=dglazkov@c-67-188-0-62.hsd1.ca.comcast.net)
791. # [17:13] <Dashiva> Lachy: Find an image that happens to contain data that can be represented as #?
792. # [17:13] <Lachy> how many times does GMail have to ignore your requests to treat it as spam for you to take the hint that it isn't!
793. # [17:13] <Lachy> :-)
794. # [17:17] * Quits: myakura (n=myakura@p3213-ipbf4202marunouchi.tokyo.ocn.ne.jp) ("Leaving...")
795. # [17:20] <gsnedders> AryehGregor: # is part of the content now
796. # [17:20] <gsnedders> I think
797. # [17:22] * Joins: cardona507 (n=cardona5@c-67-180-160-250.hsd1.ca.comcast.net)
798. # [17:22] * Quits: boblet (n=boblet@p1072-ipbf36osakakita.osaka.ocn.ne.jp)
799. # [17:23] <Lachy> Dashiva, using an img with a data URL, IE does treat the # as data
800. # [17:23] <Lachy> see the live dom viewer, and download what I uploaded to its clipboard
801. # [17:24] <Lachy> Hixie, the save feature still seems to be very broken.
802. # [17:24] <AryehGregor> Yay, interop, RFC 3986 can take a hike.
803. # [17:24] <Philip`> AryehGregor: Could set up a filter to delete it automatically
804. # [17:24] <AryehGregor> Philip`, hmm, clever idea.
805. # [17:24] <AryehGregor> Maybe I will.
806. # [17:25] <Lachy> wow, even Opera can handle the data URI with a hash in it, when used as an img src
807. # [17:25] <gsnedders> AryehGregor: It's not so much that, it's just URLs beginning with javascript: and data: are special-cased
808. # [17:25] <Lachy> that's unfortunate :-(
809. # [17:26] <gsnedders> It's better than fucking up every URI scheme totally
810. # [17:28] * Joins: jg (n=jg@proxy.lucent.com)
811. # [17:28] <AryehGregor> gsnedders, RFC 3986 explicitly says that URI schemes can't special-case fragment syntax, as annevk pointed out.
812. # [17:29] <gsnedders> AryehGregor: Yes, but it isn't just fragment syntax that is special cased. It the whole URI-reference that is special cased.
813. # [17:29] <AryehGregor> Really?
814. # [17:29] * AryehGregor looks more closely
815. # [17:30] <gsnedders> If the string starts with javascript: or data:, it isn't treated as a URI at all
816. # [17:31] <AryehGregor> It seems like RFC 3986 allows URIs like data: and javascript:, with the path component defined as path-rootless.
817. # [17:31] <AryehGregor> In what way is it not treated as a URI at all? It looks like a URI to me.
818. # [17:32] <gsnedders> Indeed.
819. # [17:32] <gsnedders> AryehGregor: It's not actually parsed as a URL. The leading string is stripped, and pct-encoding is decoded.
820. # [17:33] <gsnedders> AryehGregor: <iframe src="javascript://foobar%0Adocument.write('a')"></iframe>
821. # [17:34] <AryehGregor> So how is that not parsing as a URL? It's just a URL that doesn't use the hierarchy syntax.
822. # [17:34] <AryehGregor> RFC 3986 allows those as URIs.
823. # [17:34] <AryehGregor> Basically just a leading "something:" following by anything you like, and then a fragment.
824. # [17:34] <gsnedders> AryehGregor: Because we'd have to separate it up into segments, and would only apy attention to path
825. # [17:36] <AryehGregor> Well, you do. The path is just one segment, namely, the whole URL other than the initial "data:" and the trailing fragment.
826. # [17:36] <Dashiva> And query part
827. # [17:36] <AryehGregor> Yeah, that too.
828. # [17:36] <AryehGregor> Was about to say.
829. # [17:36] <gsnedders> AryehGregor: No, we don't.
830. # [17:36] <gsnedders> Look at that example above
831. # [17:36] <gsnedders> AryehGregor: The whole thing is in the host
832. # [17:36] <AryehGregor> What about it? The path is "//foobar\ndocument.write('a')".
833. # [17:36] <AryehGregor> No, there's no host.
834. # [17:37] <gsnedders> How not?
835. # [17:37] * Joins: dglazkov (n=dglazkov@nat/google/x-gyghjjfurodfcucq)
836. # [17:37] <gsnedders> A path can't start with //
837. # [17:37] <gsnedders> Because an authority starts with //
838. # [17:37] <AryehGregor> URI = scheme ":" hier-part [ "?" query ] [ "#" fragment ]
839. # [17:37] * gsnedders thinks AryehGregor has got his ABNF wrong
840. # [17:37] <AryehGregor> hier-part = "//" authority path-abempty
841. # [17:37] <AryehGregor> / path-absolute
842. # [17:37] <AryehGregor> / path-rootless
843. # [17:37] <AryehGregor> / path-empty
844. # [17:37] <AryehGregor> path-rootless = segment-nz *( "/" segment )
845. # [17:37] <gsnedders> It doesn't match path-absolute, path-rootless, or path-empty
846. # [17:37] <AryehGregor> segment-nz = 1*pchar
847. # [17:37] <AryehGregor> It matches path-rootless.
848. # [17:37] <gsnedders> pchar doesn't contain /
849. # [17:38] <AryehGregor> Hmm.
850. # [17:38] <gsnedders> Otherwise that would be ambiguous for cases like //gsnedders.com
851. # [17:38] <AryehGregor> Touche.
852. # [17:38] <Dashiva> javascript: URLs are just an abomination, we shouldn't try to treat them otherwise
853. # [17:38] <gsnedders> The only alternation there that can start with // is "//" authority path-abempty
854. # [17:38] <AryehGregor> Okay, you win.
855. # [17:39] <AryehGregor> data: and javascript: don't follow URI syntax at all, so no reason to quibble over the fragment or query parts.
856. # [17:41] * Quits: JonathanNeal_oww (n=Jonathan@76-219-69-134.lightspeed.breaca.sbcglobal.net) (Read error: 110 (Connection timed out))
857. # [17:46] * Joins: JonathanNeal (n=Jonathan@rrcs-76-79-114-213.west.biz.rr.com)
858. # [17:50] * Joins: maikmerten (n=maikmert@port-92-201-103-93.dynamic.qsc.de)
859. # [17:51] * Joins: _Utkarsh (n=admin@117.201.86.105)
860. # [17:53] * Quits: pesla (n=retep@procurios.xs4all.nl) ("( www.nnscript.com :: NoNameScript 4.21 :: www.esnation.com )")
861. # [18:02] * Quits: Utkarsh (n=admin@117.201.82.208) (Read error: 110 (Connection timed out))
862. # [18:02] * Joins: zcorpan (n=zcorpan@90-224-190-71-no135.tbcn.telia.com)
863. # [18:16] * Joins: ap_ (n=ap@17.246.19.5)
864. # [18:21] * Quits: othermaciej (n=mjs@c-69-181-42-237.hsd1.ca.comcast.net)
865. # [18:26] * Quits: zcorpan (n=zcorpan@90-224-190-71-no135.tbcn.telia.com) (Remote closed the connection)
866. # [18:31] * Quits: ROBOd (n=robod@89.122.216.38) (Read error: 60 (Operation timed out))
867. # [18:33] * Quits: cardona507 (n=cardona5@c-67-180-160-250.hsd1.ca.comcast.net) ("zzzzz")
868. # [18:33] * Quits: riven (n=riven@53518387.cable.casema.nl) (Excess Flood)
869. # [18:34] * Joins: riven (n=riven@53518387.cable.casema.nl)
870. # [18:34] * Joins: maikmerten_ (n=maikmert@port-92-201-83-61.dynamic.qsc.de)
871. # [18:35] * Quits: dave_levin (n=dave_lev@74.125.59.73)
872. # [18:35] * Quits: riven (n=riven@53518387.cable.casema.nl) (Excess Flood)
873. # [18:35] * Joins: dave_levin (n=dave_lev@74.125.59.65)
874. # [18:35] * Joins: riven (n=riven@53518387.cable.casema.nl)
875. # [18:39] * Joins: cohitre (n=cohitre@64-40-56-46-dsl.itltd.net)
876. # [18:39] * Parts: cohitre (n=cohitre@64-40-56-46-dsl.itltd.net)
877. # [18:48] * Quits: gavin_ (n=gavin@firefox/developer/gavin) (Remote closed the connection)
878. # [18:48] * Quits: Omnipotent (n=youshoul@unaffiliated/dor) (No route to host)
879. # [18:48] * Joins: gavin_ (n=gavin@firefox/developer/gavin)
880. # [18:50] * Quits: maikmerten (n=maikmert@port-92-201-103-93.dynamic.qsc.de) (Read error: 110 (Connection timed out))
881. # [18:56] * Quits: KevinMarks (n=KevinMar@c-67-161-44-219.hsd1.ca.comcast.net) (Client Quit)
882. # [18:57] * Joins: ROBOd (n=robod@89.122.216.38)
883. # [18:57] * Joins: Omnipotent (n=youshoul@unaffiliated/dor)
884. # [18:59] * Joins: breakmau5 (n=breakz@erft-5d80bd4e.pool.mediaWays.net)
885. # [19:01] * Quits: mat_t (n=mattomas@62-64-162-34.dynamic.dial.as9105.com) (Remote closed the connection)
886. # [19:03] * Joins: Utkarsh (n=admin@117.201.84.87)
887. # [19:05] <Philip`> Whoops
888. # [19:05] <Philip`> I think I broke the spec
889. # [19:06] <boogyman> nice goin!
890. # [19:08] * Quits: m_W (n=mwj@c-69-141-106-205.hsd1.nj.comcast.net) (Read error: 113 (No route to host))
891. # [19:10] <Philip`> Oh, it's okay now
892. # [19:10] * Joins: Sidnicious (n=Sidney@pdpc/supporter/professional/sidney)
893. # [19:10] * Joins: erlehmann (n=erlehman@echelon.ext.c-base.org)
894. # [19:10] * Philip` has moved the multipage thing to a new server and hopes it won't break much
895. # [19:11] <annevk> hmm, why is the notification API not under navigator?
896. # [19:13] * Quits: _Utkarsh (n=admin@117.201.86.105) (Read error: 110 (Connection timed out))
897. # [19:14] <hsivonen> does the notification API let twitter put tweets into Growl?
898. # [19:14] <annevk> i think so
899. # [19:18] * Joins: m_W (n=mwj@c-69-141-106-205.hsd1.nj.comcast.net)
900. # [19:19] * Joins: ROBOd2 (n=robod@89.122.216.38)
901. # [19:21] * Joins: takoratta (n=takoratt@202.171.161.131)
902. # [19:22] * Quits: mhausenblas (n=mhausenb@wg1-nat.fwgal01.deri.ie) ("brb")
903. # [19:23] * Quits: workmad3 (n=workmad3@cpc3-bagu10-0-0-cust651.1-3.cable.virginmedia.com) (Remote closed the connection)
904. # [19:24] * Quits: Omnipotent (n=youshoul@unaffiliated/dor) (No route to host)
905. # [19:29] * Quits: sbublava (n=stephan@77.116.213.114.wireless.dyn.drei.com) (Client Quit)
906. # [19:29] * Joins: drunknbass_work (n=aaron@71.107.253.243)
907. # [19:29] * Quits: Utkarsh (n=admin@117.201.84.87) (Read error: 60 (Operation timed out))
908. # [19:31] * Joins: Utkarsh (n=admin@117.201.80.172)
909. # [19:33] * Quits: Phae (n=phaeness@gateb.thls.bbc.co.uk)
910. # [19:35] * Quits: ROBOd (n=robod@89.122.216.38) (Read error: 110 (Connection timed out))
911. # [19:38] * Joins: onar_ (n=onar@17.216.36.168)
912. # [19:53] * Quits: onar (n=onar@nat/apple/x-kqmoiaksaszxvnlz) (Read error: 110 (Connection timed out))
913. # [19:54] * Quits: takoratta (n=takoratt@202.171.161.131) (Client Quit)
914. # [20:12] * Joins: mlpug (n=mlpug@a88-115-164-40.elisa-laajakaista.fi)
915. # [20:15] * Joins: _Utkarsh (n=admin@117.201.80.194)
916. # [20:26] * Quits: Utkarsh (n=admin@117.201.80.172) (Read error: 110 (Connection timed out))
917. # [20:33] * Quits: ment (i=thement@ibawizard.net) ("->")
918. # [20:36] * Joins: Utkarsh (n=admin@117.201.80.111)
919. # [20:45] * Quits: boogyman (n=boogyman@unaffiliated/boogyman) ("ChatZilla 0.9.86 [Firefox 3.6/20100115144158]")
920. # [20:46] * Quits: _Utkarsh (n=admin@117.201.80.194) (Read error: 110 (Connection timed out))
921. # [20:48] * Joins: cying_ (n=cying@70.90.171.153)
922. # [20:51] * Joins: gratz|home (n=gratz@cpc3-brig15-2-0-cust237.3-3.cable.virginmedia.com)
923. # [20:56] * Quits: gavin_ (n=gavin@firefox/developer/gavin) (Read error: 60 (Operation timed out))
924. # [20:58] * Joins: gavin_ (n=gavin@firefox/developer/gavin)
925. # [20:59] * Joins: _Utkarsh (n=admin@117.201.80.165)
926. # [21:07] * Quits: Lachy (n=Lachlan@london.perfect-privacy.com) ("This computer has gone to sleep")
927. # [21:09] * Quits: Utkarsh (n=admin@117.201.80.111) (Read error: 110 (Connection timed out))
928. # [21:10] * Joins: dbaron (n=dbaron@nat/mozilla/x-yxchghysxixigmid)
929. # [21:11] * Joins: Utkarsh (n=admin@117.201.81.38)
930. # [21:15] * Joins: slightlyoff (n=slightly@nat/google/x-dqtobdrivtdswnww)
931. # [21:17] * Quits: slightlyoff (n=slightly@nat/google/x-dqtobdrivtdswnww) (Client Quit)
932. # [21:18] * Joins: mat_t (n=mattomas@62-64-162-34.dynamic.dial.as9105.com)
933. # [21:21] * Joins: slightlyoff (n=slightly@nat/google/x-bgdrcybjizazianw)
934. # [21:28] * Quits: cying_ (n=cying@70.90.171.153) (Remote closed the connection)
935. # [21:28] * Joins: cying_ (n=cying@70.90.171.153)
936. # [21:32] * Quits: _Utkarsh (n=admin@117.201.80.165) (Connection timed out)
937. # [21:38] * Quits: mat_t (n=mattomas@62-64-162-34.dynamic.dial.as9105.com) (Read error: 110 (Connection timed out))
938. # [21:41] * Joins: _Utkarsh (n=admin@117.201.81.124)
939. # [21:42] * Joins: itpastorn (n=user@h235n5-th-c-a31.ias.bredband.telia.com)
940. # [21:44] * Quits: Utkarsh (n=admin@117.201.81.38) (Read error: 110 (Connection timed out))
941. # [21:44] * Quits: TabAtkins (n=chatzill@70-139-15-246.lightspeed.rsbgtx.sbcglobal.net) (Read error: 60 (Operation timed out))
942. # [21:45] * Quits: breakmau5 (n=breakz@erft-5d80bd4e.pool.mediaWays.net) (No route to host)
943. # [21:48] * Quits: maikmerten_ (n=maikmert@port-92-201-83-61.dynamic.qsc.de) (Remote closed the connection)
944. # [21:51] * Joins: KevinMarks (n=KevinMar@dsl081-053-212.sfo1.dsl.speakeasy.net)
945. # [21:51] * Quits: slightlyoff (n=slightly@nat/google/x-bgdrcybjizazianw)
946. # [21:52] * Quits: karlushi (n=karlushi@fw.vdl2.ca) ("This computer has gone to sleep")
947. # [21:53] * Joins: karlushi (n=karlushi@fw.vdl2.ca)
948. # [21:55] * Joins: aroben (n=aroben@17.203.12.32)
949. # [21:59] * Joins: Omnipotent (n=youshoul@unaffiliated/dor)
950. # [22:02] * Joins: TabAtkins (n=chatzill@70-139-15-246.lightspeed.rsbgtx.sbcglobal.net)
951. # [22:03] * Parts: itpastorn (n=user@h235n5-th-c-a31.ias.bredband.telia.com)
952. # [22:17] <TabAtkins> Good writeup, Sidnicious.
953. # [22:17] * Quits: ROBOd2 (n=robod@89.122.216.38) ("http://www.robodesign.ro")
954. # [22:17] <Sidnicious> Hey, thanks.
955. # [22:18] <AryehGregor> Where?
956. # [22:18] <Sidnicious> The list
957. # [22:19] <Sidnicious> TabAtkins was helping me work through the basics of the idea a couple of days ago.
958. # [22:19] <TabAtkins> whatwg, AryehGregor
959. # [22:22] * Parts: pmuellr (n=pmuellr@nat/ibm/x-fxmgctbtvhaqmwub)
960. # [22:22] <annevk> omg Buenos Aires was so much better
961. # [22:23] <annevk> it's fricking freezing here
962. # [22:25] <Sidnicious> annevk: One of my coworkers just came home from Argentina. He's been hanging around 85°F beaches for the last month, suddenly he's back in winter NYC.
963. # [22:25] <Sidnicious> He's perplexed, to say the least.
964. # [22:26] * Joins: aroben_ (n=aroben@unaffiliated/aroben)
965. # [22:27] * Quits: mlpug (n=mlpug@a88-115-164-40.elisa-laajakaista.fi) (Remote closed the connection)
966. # [22:27] * Joins: Utkarsh (n=admin@117.201.80.178)
967. # [22:28] <svl> annevk: be glad you missed the -10 days
968. # [22:30] * Quits: gratz|home (n=gratz@cpc3-brig15-2-0-cust237.3-3.cable.virginmedia.com) ("Leaving")
969. # [22:33] * Quits: Omnipotent (n=youshoul@unaffiliated/dor) (No route to host)
970. # [22:36] * miketaylr is now known as notmikehostetler
971. # [22:39] * notmikehostetler is now known as miketaylr
972. # [22:39] * Quits: _Utkarsh (n=admin@117.201.81.124) (Connection timed out)
973. # [22:40] <annevk> svl, I suppose :)
974. # [22:40] <annevk> about 1800 emails read/deleted so far
975. # [22:42] * Joins: _Utkarsh (n=admin@117.201.80.84)
976. # [22:43] * Joins: slightlyoff (n=slightly@nat/google/x-mtajepmkhuxhipii)
977. # [22:43] * miketaylr is now known as mijktaylr
978. # [22:43] * Quits: aroben (n=aroben@unaffiliated/aroben) (Read error: 110 (Connection timed out))
979. # [22:50] * Quits: m_W (n=mwj@c-69-141-106-205.hsd1.nj.comcast.net) (Connection reset by peer)
980. # [22:54] * Quits: Utkarsh (n=admin@117.201.80.178) (Read error: 110 (Connection timed out))
981. # [22:54] * Joins: Utkarsh (n=admin@117.201.80.76)
982. # [23:01] * Joins: Lachy (n=Lachlan@85.196.122.246)
983. # [23:02] * Quits: mijktaylr (n=miketayl@38.117.156.163) (Remote closed the connection)
984. # [23:03] * Joins: aroben (n=aroben@unaffiliated/aroben)
985. # [23:05] * Joins: m_W (n=mwj@c-69-141-106-205.hsd1.nj.comcast.net)
986. # [23:07] * Quits: _Utkarsh (n=admin@117.201.80.84) (Read error: 110 (Connection timed out))
987. # [23:09] * Quits: jg (n=jg@proxy.lucent.com) (Remote closed the connection)
988. # [23:14] * Joins: gratz|home (n=gratz@cpc3-brig15-2-0-cust237.3-3.cable.virginmedia.com)
989. # [23:17] <annevk> did anyone file a bug on s/srcdoc/srcDoc/ yet?
990. # [23:21] * Quits: aroben_ (n=aroben@unaffiliated/aroben) (Read error: 110 (Connection timed out))
991. # [23:22] * Joins: erikvold (n=erikvold@S01060024012860e9.gv.shawcable.net)
992. # [23:23] * Joins: _Utkarsh (n=admin@117.201.80.59)
993. # [23:29] <TabAtkins> what, about removing it, annevk? Yes.
994. # [23:30] <annevk> no, about properly naming the IDL attribute
995. # [23:30] <annevk> more or less every addition has a bug for removal, that's not very special :)
996. # [23:32] * Quits: gratz|home (n=gratz@cpc3-brig15-2-0-cust237.3-3.cable.virginmedia.com) ("Leaving")
997. # [23:33] * Joins: aroben_ (n=aroben@unaffiliated/aroben)
998. # [23:33] * Joins: jwalden (n=waldo@64.9.243.181)
999. # [23:35] * Quits: Utkarsh (n=admin@117.201.80.76) (Connection timed out)
1000. # [23:37] <Hixie> re the obsoleting old versions thing, see my comments on the text/html rfc from a couple of days ago on irc
1001. # [23:37] * Joins: workmad3 (n=workmad3@84.45.226.85)
1002. # [23:37] <Hixie> the gist of which was that the currrent text/html rfc does the same as far as i can tell
1003. # [23:39] * Quits: workmad3 (n=workmad3@84.45.226.85) (Remote closed the connection)
1004. # [23:40] * Joins: Utkarsh (n=admin@117.201.80.110)
1005. # [23:45] * Joins: othermaciej (n=mjs@c-69-181-42-237.hsd1.ca.comcast.net)
1006. # [23:46] * Quits: aroben (n=aroben@unaffiliated/aroben) (Nick collision from services.)
1007. # [23:46] * aroben_ is now known as aroben
1008. # [23:46] <annevk> hmm, still no File.URL? -- http://dev.w3.org/2006/webapi/FileAPI/#dfn-file
1009. # [23:46] <annevk> with everyone implementing I hope that gets fixed soon :/
1010. # [23:49] * Joins: paul_irish (n=paul_iri@c-71-192-163-128.hsd1.nh.comcast.net)
1011. # [23:51] <Hixie> send feedback again
1012. # [23:51] <Hixie> or e-mail arun directly
1013. # [23:51] <Hixie> the latter is probably best
1014. # [23:53] * Quits: _Utkarsh (n=admin@117.201.80.59) (Read error: 110 (Connection timed out))
1015. # [23:54] <Philip`> Hixie: Saying "the old version already has that problem" doesn't seem like a good reason for not fixing it, in general
1016. # [23:54] * Quits: BlurstOfTimes (n=blurstof@168.203.117.66) (Client Quit)
1017. # [23:54] <Hixie> my argument is that there is no problem
1018. # [23:55] <Philip`> If there's no problem, why does it matter what the current text/html RFC says?
1019. # [23:55] * Quits: Utkarsh (n=admin@117.201.80.110) (Read error: 54 (Connection reset by peer))
1020. # [23:55] <Philip`> I suppose it does matter if people who say the new one has a problem don't also acknowledge that the old one has the same problem
1021. # [23:56] * Joins: pmuellr (n=pmuellr@user-0c99bjg.cable.mindspring.com)
1022. # [23:56] <Hixie> Philip`: it demonstrates that there's no problem
1023. # [23:56] <Hixie> the argument is "you can't publish a registration that doesn't do X" but the existing registration doesn't do X, so it is a counterproof.
1024. # [23:57] <Dashiva> Equal rights
1025. # [23:58] <Dashiva> You can't just go changing what is considered a conforming mime-type registration because a new spec for the format has been released
1026. # [23:58] * ap_ is now known as ap
1027. # Session Close: Sat Jan 30 00:00:00 2010

The end :)