XMPP Council - 2011-03-02

  1. MattJ has left
  2. Tobias has joined
  3. Tobias has left
  4. Tobias has left
  5. Tobias has joined
  6. Tobias has left
  7. Tobias has joined
  8. Tobias has left
  9. Tobias has joined
  10. Tobias has left
  11. Tobias has joined
  12. Tobias has left
  13. Tobias has joined
  14. Tobias has left
  15. bear has left
  16. bear has joined
  17. Tobias has joined
  18. stpeter has joined
  19. Kev So, we have nothing to discuss, but we still have a meeting scheduled.
  20. Kev I think a quick discussion of RTT would be sensible, though.
  21. Kev waits for 7minutes.
  22. linuxwolf has joined
  23. linuxwolf I take it we only have an "unofficial" topic today? (-:
  24. Kev Oh, two, actually.
  25. linuxwolf !agenda
  26. Kev Oh, maybe I should write a bot to do that.
  27. Kev That'd be good.
  28. Kev makes a note.
  29. linuxwolf at one point, chatbot could do that
  30. linuxwolf or some variation of it
  31. Kev Well, I'm working (very slowly) on a new bot, so I'll add that to it when it's ready.
  32. linuxwolf heh
  33. linuxwolf I'm working very slowly on a new voting bot
  34. Kev Oh, wonderful.
  35. linuxwolf but it's built with EVIL CLOSED SOURCE EMPIRE SOFTWARE right now
  36. Kev Heh.
  37. Kev Language?
  38. linuxwolf er, was…I'll have to see if that repo survived some migrations
  39. Kev You're welcome to throw it up on top of Swiftob if it's something C++ish.
  40. linuxwolf it started Java, then moved to JavaScript…now, we'll see
  41. linuxwolf well, I need to get it through Legal first (-:
  42. Kev Well, javascript is almost Lua, and I'll be putting Lua support in shortly.
  43. Kev Ah, I hadn't understood that the code itself was evil and closed-source as well as what it was using :)
  44. linuxwolf the "joys" of working for a multi-national
  45. linuxwolf yeah, unfortunately
  46. linuxwolf the closed-source part might change…you know, eventually…maybe
  47. linuxwolf /sigh
  48. Kev Everyone has been poked.
  49. Kev 6 billion people took a while, but it's been done.
  50. Fritzy has joined
  51. MattJ has joined
  52. linuxwolf raflmao
  53. Kev There we go, right room this week :)
  54. Kev No response from Ralph to the ping, but we have quorum.
  55. Kev Autobots transform and roll out!
  56. MattJ What's on the agenda?
  57. Kev MattJ: I'm making it up as we go along, sorry.
  58. linuxwolf RTT
  59. MattJ k
  60. Kev 1) Roll call
  61. linuxwolf presente
  62. Fritzy here
  63. Kev I like the idea of people replying to Roll calls, so I at least have an idea they're listening at the start :)
  64. Kev I used to always just list who was here, and then spend the meeting wondering if they were at their PCs :)
  65. MattJ I'm here, I think
  66. Kev Marverful.
  67. Kev 2) Agenda bashing.
  68. Kev So, yeah, until 30mins ago I didn't think we had much agenda, but now I think we do.
  69. Fritzy oh, ok
  70. Kev So I'm making it up as we go along, it basically goes "RTT" and then "XEP-Correct"
  71. linuxwolf hehehe
  72. Kev Anyone else have things we need to talk about?
  73. Fritzy I just finished catching up on that thread.
  74. Tobias has left
  75. Kev I'll take that as a no :)
  76. Fritzy Kev: nothing here
  77. Kev So...
  78. Kev 3) RTT
  79. Kev That thread seems to be making progress to address my concerns.
  80. Fritzy Sounds like he's going to do a revised spec.
  81. MattJ Indeed
  82. Kev They've persuaded me that all the gumph around delay transmission and stuff is useful to a subset of users.
  83. linuxwolf yes
  84. Fritzy and you pretty much summed up all of the concerns that I had (plus some). I still need to poke the thread a bit now that I've caught up.
  85. Kev And I think they've agreed to tidy up the spec, split it into Stuff You Must Have, and The Complicated Stuff.
  86. linuxwolf I do find it somewhat funny we're introducing a spec that benefits a protected class of individuals but could be a detriment to everyone else (-:
  87. Fritzy yeah, but there still isn't a super-easy base.
  88. Kev Fritzy: There is, I think.
  89. MattJ Personally I hate the idea of transmitting text as I type
  90. Kev If they agree to do replace-only as the base.
  91. linuxwolf we'll have to see what the next rev is like
  92. Kev Right, I dislike using this feature, but I can see the merit in clients supporting it in case you ever talk to a deaf person or whatever.
  93. MattJ but it seems plenty of people want it, and I'd rather we spec it through the XSF than leave them to cook up something themselves
  94. Kev I'd turn it on, despite disliking it, if I was talking to someone who'd greatly benefit.
  95. Fritzy I'd use it for Frontdesk.im
  96. Fritzy if it were simple enough to implement
  97. linuxwolf my typing abilities are atrocious enough
  98. linuxwolf (-:
  99. Kev I think there's merit in what he says, too.
  100. Fritzy because I have customers that want it
  101. Kev The basic set of operations *are* just basic deterministic string manipulations.
  102. linuxwolf I don't know that our customers would allow it in their networks…but we'll see
  103. Kev The pain is in writing the parsers and serialisers for it all.
  104. Fritzy He's not afraid to send a wall of text back at every point.
  105. Kev So, basically, I'd like people to reply on-list please, even if it's just to say that you feel your concerns are/aren't being addressed by the way the conversation's going.
  106. Fritzy ok
  107. Kev (and if they aren't, why, of course)
  108. Kev I'd like it if when they came back with a v2 spec, they'd had all the feedback they need to make it something we could accept.
  109. linuxwolf /nod
  110. Kev So, onwards
  111. Kev 4) Xep-correct.
  112. Kev I wrote up http://doomsong.co.uk/extensions/render/xep-correct.html (could you have a quick read please, it's very short)...
  113. Fritzy reading..
  114. Fritzy (saw you mentioned it in the thread)
  115. linuxwolf I think I've seen this before...
  116. Kev some time ago, and didn't get around to filling in the blanks. Various people keep prodding me because they want it published. Is the approach acceptable to Council, and if so what's the minimum I need to do to the spec to get it past a pseudo-vote to experimental?
  117. ralphm has joined
  118. Kev Hi Ralph.
  119. ralphm hi
  120. MattJ "Standard stuff. To come later."
  121. MattJ I mean, the approach is clearly ok
  122. linuxwolf Kev: namespace to urn:xmpp:correct:0, and but a schema in place; maybe the boiler plate about using Disco
  123. MattJ Just editorial stuff
  124. Fritzy you don't address multiple <body /> here.
  125. Fritzy which seems like it could pose a problem for multi-language messages
  126. Kev Fritzy: This is true. This is easily worded around, though.
  127. MattJ Good catch
  128. linuxwolf I wouldn't have held it up for the l10n problems
  129. Kev "Only do the replacement on the body with the same xml:lang as the one you've received"
  130. MattJ Just choose the body with the same xml:lang as the <replace>
  131. linuxwolf not to get to experimental
  132. Fritzy yeah, I think filling in the blanks is enough for experimental
  133. linuxwolf for features, just pick some not-insane values for now, and let's push it!
  134. Kev Ok, so, summary of what I need to do is 1) Add the Disco/caps examples 2) change namespace (although I think this act technically happens at the moment it become Experimental) 3) add Schema.
  135. linuxwolf +1
  136. MattJ +1
  137. Fritzy +1
  138. MattJ Though the multiple body issue is probably just a couple of sentences, so if you can... :)
  139. linuxwolf I have a BCP for resource locking I need to get cleaned up and submitted
  140. Kev I'll see how my 'free time' looks, and see if I can get it to Peter and through to the inbox in time for next week.
  141. ralphm +1
  142. Fritzy beyond experimental, you might separate the client-uses out of the protocol language in a separate section
  143. stpeter um I thought we weren't having a meeting today
  144. linuxwolf yeah, use cases versus implementation notes…but we can work through that on the lists
  145. Fritzy stpeter: I thought so too, but apparently minds were changed
  146. Kev stpeter: I don't think anyone proposed skipping it.
  147. stpeter aha
  148. Fritzy yeah, not directly
  149. linuxwolf we never said we *weren't*
  150. linuxwolf (-:
  151. Kev linuxwolf: We didn't say we weren't, and last Thursday we said we were, so ... ;)
  152. linuxwolf exacetally
  153. Kev Sure, though, I can see me sending a mail asking for agenda items could be interpreted as a meeting cancellation.
  154. Kev Probably an American vs. English thing.
  155. Kev :D
  156. stpeter reminds Fritzy that he needs to vote on XEP-0198 version 1.2
  157. linuxwolf <replace id='blah' xmlns='urn:xmpp:correct:0'>exactly</replace>
  158. ralphm I also assumed there wouln't be a meeting.
  159. Fritzy stpeter: I went to bed early/sick -- is it not too late?
  160. Kev I should have sent out an agenda etc. Until this afternoon, I thought we had no items and was going to propose cancelling, and by the time we were T-60minutes I didn't see the point in sending one out when we could discuss here.
  161. Kev Anyway.
  162. Kev 5) Any other business.
  163. Kev Anyone?
  164. MattJ NOP
  165. linuxwolf not this week
  166. Kev To keep everyone informed, Bear's halfway through doing the GSoC application.
  167. linuxwolf hopefully in time for next
  168. Fritzy till next week
  169. linuxwolf coolio
  170. stpeter Fritzy: not too late, no
  171. MattJ We're having a harder time getting GSoC ideas together for Prosody than we had last year
  172. stpeter it seems that I missed the RTT discussion :(
  173. Kev stpeter: Oh, panic ye not, it's still ongoing :)
  174. MattJ Since last year nearly all the GSoC ideas have been started by us or community members :)
  175. Kev MattJ: New archiving? :)
  176. Fritzy stpeter: to summarize -- make sure they have enough info to make their second proposal acceptable
  177. MattJ That's hardly a whole GSoC project
  178. MattJ sadly
  179. Kev If anyone can persuade people to please submit ideas for various projects, please please do. It looks lonely with just BC and Swift up.
  180. linuxwolf MattJ: vcard-temp to vcard4 converter? (-:
  181. Kev I've poked assorted projects, Bear tells me he's been poking assorted projects too.
  182. MattJ ick
  183. stpeter as to RTT, it is important for a subset of users, and we really don't want folks to use RFC 4103 because there's no XMPP solution :)
  184. linuxwolf MattJ: vcard is like broccoli; not many people like it, but you have to have it
  185. stpeter I love broccoli!
  186. Kev stpeter: I think the discussion onlist is progressing productively. The intention is to clean up the spec and resubmit.
  187. linuxwolf stpeter: fine, then jelly beans for you!
  188. Kev 6) Date of next meeting.
  189. Kev Next Wednesday?
  190. MattJ linuxwolf, I love broccoli, I don't love vcard4 :)
  191. MattJ Next Wednesday seems ok for me
  192. Fritzy +1 next wed
  193. stpeter Kev: super
  194. linuxwolf MattJ: vcard is like <insert undesirable but healthy food item here>; you don't like it, but have to have it
  195. Fritzy I'll likely miss the one after that as I'll be traveling to ITaly
  196. stpeter Kev: I've been reading Internet-Drafts, but will dive into RTT later today
  197. linuxwolf +1 for wed
  198. MattJ linuxwolf, I don't like sprouts, and I do fine without them
  199. stpeter I'm happy to see the RTT folks engaging on the list
  200. Kev Very nice, I ate lots of pizza in Italy a few weeks ago :)
  201. linuxwolf heh
  202. Kev Ok, so I think we're on for Wed then.
  203. Fritzy we'll have to talk outside of the meeting then
  204. stpeter so xep-correct is to be published?
  205. stpeter seems to be unanimous
  206. Fritzy not yet
  207. MattJ Kev is to clean up and submit it
  208. Kev stpeter: No, XEP-Correct is to have a few fixes so it's suitable for submission, then Council have agreed to publish it :)
  209. MattJ and we've pre-approved it :)
  210. Fritzy no no
  211. Kev Apart from me, I've not approved it yet :D
  212. stpeter and I saw no objections to vcard4 xep (vcard4 is another matter!) so I assume that is to be published
  213. stpeter ok
  214. stpeter Kev: :P
  215. Fritzy we've suggested that Kev should
  216. stpeter ok
  217. linuxwolf is feeling like a dirty loan servicer (-:
  218. Kev stpeter: Voting period on vcard4 is over, so that's approved, yes.
  219. Fritzy linuxwolf: dude, the inner circle has an agenda, man.
  220. Kev s/Voting/Thing that looks like voting, but isn't/
  221. MattJ vcard4 is to be published, but I'm unconvinced of it seeing adoption and going beyond that unless it changes, but we'll see
  222. Kev Ok, I think we're done, then.
  223. stpeter I'm giving Fritzy a grace period on 198 in case he finds anything bad
  224. Kev stpeter: Very generous :)
  225. Fritzy looking for bad right now
  226. stpeter but I'll push it out by end of day if we don't hear from him
  227. linuxwolf MattJ: please post your concerns to the lst
  228. stpeter Fritzy: many thanks
  229. Kev Ok, thanks all :)
  230. linuxwolf *list
  231. stpeter yep thanks
  232. Kev gangs the bavel.
  233. stpeter goes back to IESG telechat preparation
  234. MattJ linuxwolf, I discussed it a bit with waqas and he posted for the both of us :)
  235. linuxwolf adios, off to next meetage
  236. linuxwolf has left
  237. Kev adds minute-writing to his TODO
  238. stpeter heh
  239. stpeter thanks
  240. stpeter Kev: regarding message correction, I'll ping Florian because he was working on a proto-proto-xep about this once upon a time
  241. Kev Which of the assorted Florians? :)
  242. stpeter Florian Jensen
  243. Kev I did send this one out to standards@ some 6 months ago, so I imagine he's aware of it, but thanks :)
  244. stpeter ah
  245. stpeter well I poked him via IM
  246. Fritzy stpeter: section 5 "In addition, this allows entities to establish definitively which stanzas require resending and which do not, eliminating replay issues."
  247. stpeter I must've missed it 6 months ago
  248. Fritzy that seems a bit more definitive than it probably means to be
  249. Fritzy I mean, if an ack is only sent every so often, it doesn't prevent all replay
  250. stpeter well, you know which stanzas haven't been acked and therefore might need to be resent
  251. stpeter I think that text came from Dave Cridland's patch, so I'll blame him :)
  252. Fritzy haha
  253. stpeter I think this would be better: In addition, this enables entities to establish which stanzas might need to be resent.
  254. Fritzy I think that is less amiguous.
  255. Fritzy *ambiguous
  256. Fritzy I need XEP-correct
  257. stpeter :)
  258. MattJ I think the point he's trying to make is that without this, there's the chance of message duplication ("replay")
  259. stpeter right
  260. stpeter the word "replay" brings to mind replay attacks
  261. MattJ which is what convinced me in the end that XEP-0198 cannot, and should not, be split
  262. MattJ despite my original desires :)
  263. stpeter heh
  264. Fritzy stpeter: sent my vote
  265. stpeter how about this? In addition, this enables entities to establish which stanzas might need to be resent, thus reducing the likelihood that an entity will send or receive duplicate messages.
  266. stpeter s/messages/stanzas/
  267. MattJ Does it reduce the likelihood, or eliminate the possibility?
  268. stpeter well
  269. Kev Fritzy: 198 is draft, so this is in fact a real vote (i.e. your +1 is a +1, not a veto) :)
  270. MattJ I thought the latter, but I may be wrong
  271. stpeter to Fritzy's point, it might depend on how you use it
  272. stpeter eliminate is a strong word :)
  273. Kev (not a not a veto - I need XEP-correct)
  274. stpeter it's probably not good to overpromise
  275. Fritzy Kev: alright. :)
  276. Fritzy Kev: ah right,
  277. Fritzy MattJ: it depends on how it is used
  278. stpeter http://xmpp.org/about-xmpp/xsf/xmpp-council/tenth-council/ updated
  279. ralphm stpeter: cool. I also noticed that the bios on the Council page are outdated
  280. stpeter ralphm: we can give you editing privileges :)
  281. stpeter or send updates to the council@ list
  282. Fritzy has left
  283. Fritzy has joined
  284. Tobias has joined
  285. bear has left
  286. bear has joined
  287. Tobias has left
  288. Tobias has joined
  289. Tobias has left
  290. Tobias has joined
  291. stpeter has left
  292. stpeter has joined
  293. Tobias has left