XMPP Summit - 2018-02-01


  1. vanitasvitae

    Off I go, on my way to brussels :)

  2. Guus

    As am I! See you in a bit!

  3. Guus

    Awesome, one of my trains got cancelled.

  4. vanitasvitae

    Up to this point I was lucky

  5. vanitasvitae

    Toi toi toi

  6. Guus

    New ETA is 10:00, but then I still need to travel from Brussel North station to Cisco

  7. vanitasvitae

    Guus: same here. My train from cologne to brussels is 30 mins late

  8. Kev

    Looking at the timings for the Brussels trains this morning, I think we'd be best off leaving the Thon at about 08:30, grabbing the metro to Centraal, and going from there to Diegem on the 09:04, which'd get us to the office at about 09:30. Other options get us to the office very close to 10:00, and don't give us scope for the train delays and cancellations we've seen in previous years.

  9. Tobias

    Sounds reasonable

  10. Kev

    The 09:31 from Schuman sounds lovely, but if that's cancelled we're stuck for another hour and don't get to Cisco until 11:00.

  11. Kev

    Right. Heading down for breakfast :)

  12. Guus

    Ooh, two of my trains got cancelled. This is going to be interesting.

  13. vanitasvitae

    I'm on the road again (or on the rails to be more precise)

  14. michalwski

    Kev i'm going with you. Where do we meet at the hotel?

  15. michalwski

    Did my message was really copied 4 times or is my app only?

  16. Guus

    I got two distinct messages from you in this MUC, michalwski

  17. Vaulor

    Same here, michalwski

  18. Guus

    I'm going to try and get off the train in Mechelen, and grab a cab to Cisco. Might just make it in time, if traffic isn't to bad.

  19. Guus

    Assuming that traffic is not bad in Brussels is not unlike assuming snowfall in hell, but hey, I'm an optimist.

  20. jonasw

    good morning dear summiteers

  21. Guus

    Hello dear jonasw

  22. jonasw

    so, I don’t know how these things usually work, but it’d be great to have agendums announced here, if there is such a thing like an agenda

  23. jonasw

    so, I don’t know how these things usually work, but it’d be great to have agenda announced here, if there is such a thing like an agenda

  24. jonasw

    even if A/V is available, I won’t be able to pay attention the whole day, but XMPP 2.0 would be very interesting for me and I’d like to "tune in"

  25. Guus

    jonasw: well create the agenda during the first few minutes of the meeting

  26. jonasw

    swete :)

  27. jonasw

    cool :)

  28. Guus

    I'll share the a/v details on the wiki now

  29. Guus

    If I can, from mobile... Give me a sec

  30. jonasw

    share them here and I’ll put them on the wiki?

  31. Guus

    Join the meeting from https://acecloud.webex.com Thursday, February 1, 2018 8:45 am | 10 hrs Meeting number (access code): 202 512 989 Meeting password: jsf2S8P9 Friday, February 2, 2018 8:45 am | 10 hrs Meeting number (access code): 208 941 793 Meeting password: M8nR97ws

  32. jonasw

    shall I put that on the wiki?

  33. Guus

    Yes please

  34. Guus

    I will send the host key to Kev in private. Ralph should also have it.

  35. jonasw

    are those timestamps UTC?

  36. Guus

    I assume they are local time

  37. jonasw

    hm, early

  38. jonasw

    Guus, https://wiki.xmpp.org/web/Summit_22#A.2FV

  39. jonasw

    oh, I’ll put the URL to the webex thing too

  40. Guus

    Thanks. I'd make it part of the "remote participation" section, just to be clear.

  41. ralphm

    We have arrived at Cisco.

  42. jonasw

    Guus, it is, it’s a subsection

  43. Guus

    Ah, that didn't render well on mobile. Thanks jonasw. I've also posted it on the summit mailinglist.

  44. Guus

    ralphm: everything in order there?

  45. ralphm

    We're still in the parking lot

  46. Ge0rG

    Guus: do you have an ETA for the Summit start?

  47. ralphm

    Going in soonish

  48. Guus

    I'll get off this last train in a few moments. Hope to be there around 10.

  49. Guus

    Ge0rG: 10.

  50. Ge0rG

    I was kindly asked to present XMPP2 in an earlyish slot, and I have some real-work appointments to route around it

  51. Guus

    CET

  52. Ge0rG

    so -30mins

  53. Guus

    Yup

  54. jonasw

    hm

  55. jonasw

    that’s uncool timing for me. sounds like XMPP2 will run into my lunch break :)

  56. jonasw

    but whatever

  57. jonasw

    should’ve been there :)

  58. mathieui

    MattJ: ping?

  59. MattJ

    On our way sorry... my brother being slow 😭

  60. Guus

    ralphm: mind having a stab at setting up the remote participation?

  61. jonasw

    this sounds like a fun get-together already, I think I’ll have to join next year.

  62. Guus

    jonasw: you should!

  63. SaltyBones

    So far the jokes are pretty bad ;)

  64. jonasw

    SaltyBones, sounds great

  65. Ge0rG

    "The meeting you are trying to join is not currently in progress."

  66. Intosi

    That's entirely correct.

  67. ralphm

    Yeah, we have some bootstrapping issues

  68. Guus

    Bah. Apologies. I'm inbound, 10 minutes out.

  69. vanitasvitae

    I just drove past cisco with my train :D

  70. vanitasvitae

    But I'll arrive at midi soon

  71. mathieui

    We'll arrive around 10:15

  72. Guus

    I'm there.

  73. vanitasvitae

    Hm i missed the train because the ticket machine didnt work :( will be there in 40 minutes

  74. dwd

    vanitasvitae: Does your train have a flight number?

  75. vanitasvitae

    S2 3659

  76. Ge0rG

    So, are issues sorting out, slowly?

  77. Holger is at Brussels airport.

  78. alameyo

    I am at the airport too

  79. Tobias

    Ge0rG, slowly

  80. vanitasvitae

    Finally on the last train :)

  81. ralphm

    Ok. New webex stuff: https://acecloud.webex.com/ Meeting number: 155 347 963 Meeting password: XMPP$$123

  82. jonasw

    ralphm, will you update the wiki page?

  83. jonasw

    "Invalid number or room ID. Try again." ralphm

  84. jonasw

    (while the meeting numbers from the wiki get me to "not in progress")

  85. jonasw

    I’m heading out now though

  86. ralphm

    oh, let me check the url then

  87. ralphm

    https://go.webex.com/go/e.php?MTID=mfc24e0ece7aead2f2d38ecb8f88fd94f

  88. ralphm

    Try that

  89. alameyo

    Holger: improvisation

  90. Tobias

    Ge0rG, ^

  91. alameyo

    I might go back to piano at he airport unless you are closer to the exit

  92. Holger

    I'm right at the piano.

  93. alameyo

    so wait 2-3 minutes for me

  94. Holger

    Will do! 👍

  95. vanitasvitae

    You have to negotiate a secret symbol

  96. alameyo

    Holger: security doesn't allow me to go back to piano

  97. alameyo

    go up the stairs and I am behind security gate

  98. alameyo

    in way to exit

  99. Holger

    Ah ok, I'm on the way.

  100. vanitasvitae

    Was it the first building or the second?

  101. Ge0rG

    So I'm in the WebEx now. What's the topic? :>

  102. ralphm

    disco items caching

  103. Ge0rG

    Awesome. We need that.

  104. Ge0rG

    The video just degraded into a surrealistic aquarel painting.

  105. ralphm

    The initial idea is to put the hash of the item list into a XEP-0004 form that becomes part of the disco info response that can be cached like normal with CAPS

  106. dwd

    Merkle-Caps.

  107. dwd

    Which is much of the discussion at the moment - how deep is the Merkle-Caps-tree.

  108. Tobias

    Ge0rG, are you ready?

  109. Ge0rG

    can you hear me

  110. Tobias

    no

  111. Tobias

    now

  112. Ge0rG

    ralphm: https://op-co.de/tmp/whats-wrong-with-xmpp-2017.pdf

  113. vanitasvitae

    did you receive a wifi password?

  114. vanitasvitae

    I fear I havent

  115. dwd

    Nobody has. You'll have to pay attention instead. It's a disaster.

  116. vanitasvitae

    D:

  117. Holger

    Public transport is great fun.

  118. dwd

    Since you're using XMPP, you can use multiple transports, though.

  119. goffi

    hi

  120. Holger

    dwd: 😛

  121. Holger

    We're walking!

  122. goffi

    Is there anything happening at the moment ? The WebEx link tells me "The meeting is not in progress"

  123. mathieui

    do you have the latest link?

  124. goffi

    mathieui: I've used the one on the wiki

  125. mathieui

    webex is up and Ge0rG is talking

  126. goffi

    hum so I guess I don't have the right link

  127. mathieui

    10:41:41 ralphm> Ok. New webex stuff: https://acecloud.webex.com/ Meeting number: 155 347 963 Meeting password: XMPP$$123

  128. goffi

    OK thanks

  129. goffi

    updating the wiki would be a good idea

  130. goffi

    Invalid number or room ID. Try again.

  131. goffi

    ralphm: I'm getting Invalid number with those credentials (and there is no password requested)

  132. Tobias

    goffi, have you tried https://go.webex.com/go/e.php?MTID=mfc24e0ece7aead2f2d38ecb8f88fd94f ?

  133. ralphm set the topic to

    XMPP Summit 22 Brussels | https://wiki.xmpp.org/web/Summit_22 | WebEx: https://go.webex.com/go/e.php?MTID=mfc24e0ece7aead2f2d38ecb8f88fd94f

  134. goffi

    Tobias: no, but nobody gave me this link I think

  135. Tobias

    please try it then :)

  136. mathieui

    right, I didn’t read enough

  137. goffi

    Tobias: working, thanks

  138. Tobias

    you're welcome

  139. goffi

    I have video but not sound, checking if it's coming from my side

  140. goffi

    OK it's good now

  141. mathieui

    goffi, we hear you

  142. Tobias

    could you silence you're mic

  143. mathieui

    thanks

  144. goffi

    sorry :)

  145. mathieui

    Ge0rG, we lost you

  146. jonasw waves

  147. vanitasvitae waves back

  148. intosi

    "The list up there" is:

  149. intosi

    * Presence Sorting / per resource

  150. intosi

    * Message routing

  151. intosi

    * Stable IDs

  152. intosi

    * Read state

  153. intosi

    * History / Inbox

  154. intosi

    * Metadata

  155. intosi

    * Rooms

  156. intosi

    * Impromptu

  157. intosi

    * Notifications

  158. intosi

    * Fast Connections

  159. intosi

    * BIND2

  160. intosi

    * SASL2

  161. intosi

    * Disco caching

  162. intosi

    * Non IM

  163. intosi

    * Low prio

  164. Ge0rG

    intosi: you urgently need to update to a multi-line capable client :P

  165. intosi

    It can do multi-line

  166. intosi

    I just couldn't be bothered

  167. Ge0rG

    re "Carbons needs to go away": we still need it as a transport to deliver sent-copies of outgoing messages to our other client

  168. jonasw

    who’s speaking by the way? I guess it’s Kev.

  169. intosi

    It's Kev.

  170. Zash

    Sounds like

  171. intosi

    Guus speaking now.

  172. jonasw

    does that video-zooms-in-on-some-speaker thing happen automatically :-O

  173. Zash

    Is there video?

  174. intosi

    It does.

  175. Ge0rG

    jonasw: it's supposed to.

  176. intosi

    There are a bunch of cameras in this room, and a lot of Dark Magic™ happens.

  177. jonasw

    I bet

  178. Ge0rG

    and this is why I'm seeing the same position for an hour now.

  179. Zash

    Hm, works in Chromium but not Firefox :|

  180. jonasw

    yeah, I guessed that firefox will be icky so I went with chromium right away

  181. Ge0rG

    The password is "XMPP Summit"

  182. jonasw

    and then there was silence

  183. Zash

    *crickets*

  184. Zash

    /nick tumbleweed

  185. jonasw

    someone is stealing a chair!

  186. Ge0rG

    Hopefully not the meeting chair.

  187. jonasw

    hah

  188. Ge0rG

    I wondered about 100% CPU load, but then realized it's Chromium for the WebEx

  189. goffi

    same here, while on Firefox it was fine

  190. goffi

    not sure if video worth it

  191. goffi

    well maybe not, Firefox is high too when switching back to it, weird it was fine before

  192. goffi

    can somebody ping this room when it's starting again ? I've stopped webex for now

  193. jonasw

    +1 ^

  194. winfried

    +1

  195. Zash

    People seem to be falling back in

  196. jonasw

    do we maybe wanna make a soundcheck for the remote people? I have no idea if my audio is okay-ish configured

  197. Zash

    Ge0rG was LOUD compared to the people in the room

  198. jonasw

    yeah

  199. goffi

    thanks Zash

  200. jc

    Vaulor: hey, I told him ☺️

  201. goffi

    is this webex thing using some kind of XMPP ?

  202. Ge0rG

    Zash: sorry, I started out silent and tuned my microphone to 11

  203. jonasw

    what’s the topic?

  204. jonasw

    I lack context

  205. MattJ

    Nothing formal - half of everyone is still at lunch

  206. jonasw

    sounds like something technical DNSy

  207. MattJ

    jabber.org has had DNS issues for a while

  208. jonasw

    while we’re at it, can we somehow increase volume of the microphone in that room?

  209. MattJ

    When ralphm gets back, perhaps

  210. jonasw

    ahk. I don’t know faces. I saw sams badge and I recognize guus, I guess who’s holger, but otherwise I’ve got no idea :)

  211. goffi

    I have a video in small on the right, and the passed presentation in big in the center. Is there any way to move video to main video ? I don't seen anything in UI to do that.

  212. jonasw

    goffi, the separator between them can be dragged

  213. goffi

    oh right thanks, could hardly see it

  214. Zash

    oooooooh, I wondered about that as well

  215. jonasw

    yah

  216. jonasw

    I also only found that by accident

  217. MattJ

    Is someone note-taking?

  218. Tobias

    yes, jc

  219. jonasw

    go jc!

  220. MattJ

    Thanks jc!

  221. dwd

    No, don't go! Keep taking notes!

  222. jonasw

    well

  223. Holger

    > I guess who’s holger Was still at lunch when you wrote that though 🙂

  224. jc

    You're welcome, I'll upload once I have internet in the laptop

  225. Zash

    who was phone^W Holger

  226. jonasw

    Holger, true, but I think I recognized you when you came in late

  227. jonasw

    I didn’t mean to say that you were there when I wrote that

  228. dwd

    Pretty exciting. Everyone's getting WiFi.

  229. goffi

    is is a surprise test ?

  230. goffi

    is it*

  231. goffi

    you have 2 hours

  232. vanitasvitae

    They block XMPP!!!!

  233. Zash

    Typical corporate firewall?

  234. jonasw

    what?

  235. jonasw

    seriously?

  236. jonasw

    at Cisco?

  237. vanitasvitae

    I cant connect to any of my servers

  238. Holger

    jonasw: Right, "came in late" sounds like me.

  239. vanitasvitae

    Internet works fine though

  240. debacle

    Holger, like in the Berlin XMPP meetup? :~)

  241. mathieui

    Ge0rG, jonasw, goffi, wegre starting

  242. debacle

    But I'm in the same group then

  243. dwd

    l'll stick with 4G. We're starting again, by the way.

  244. mathieui

    vanitasvitae, also they don’t block XMPP

  245. jonasw

    mathieui, thanks, I’m already in

  246. vanitasvitae

    I remember last year I worked around it by connecting through a vpn

  247. goffi

    mathieui: thx

  248. mathieui

    it’s fine on 443 :p

  249. jonasw

    winfried

  250. Tobias

    we're starting again

  251. vanitasvitae

    Ah that might be

  252. winfried

    yes?

  253. jonasw

    winfried, meeting’s resuming

  254. winfried

    (y)

  255. Holger

    debacle: Same procedure as everywhere.

  256. goffi

    jonasw: we can see you, no sure if this is intended

  257. jonasw

    goffi, it is

  258. goffi

    OK :)

  259. jonasw

    a fancy green light is telling me

  260. goffi

    it's hard to get everything, what's the topic ?

  261. jonasw

    goffi, we’re still at XMPP 2.0, but I’m not sure what the PEP thing is about

  262. jonasw

    can someone from the room please clarify?

  263. MattJ

    Before lunch we discussed per-user presence, instead of what we currently have: per-device

  264. Zash

    Status stuff in PEP instead of in presence

  265. dwd

    <status/> and <show/> moving from presence to PEP.

  266. jonasw

    ah

  267. MattJ

    This would be an additional status, in PEP

  268. jonasw

    now things make sense

  269. MattJ

    SamWhited offered to write a XEP

  270. dwd

    MattJ, Replacement, not additional.

  271. MattJ

    Indeed

  272. jcbrand

    I'll continue with the notes here: https://beta.etherpad.org/p/XMPP_Summit_2018_Day_1

  273. goffi

    jcbrand: neat

  274. Zash

    VERY LOUD YES

  275. jonasw

    sorry

  276. jonasw

    was it better then?

  277. jonasw

    otherwise I’ll tune it down even more. in the preview it was like 2 out of 10 bars, so I assumed it was too silent, Zash

  278. Zash

    I've vol'd up to hear the room

  279. jonasw

    ah right

  280. jonasw

    I forgot that you aren’t there

  281. jonasw

    yah, the room is really quiet

  282. Zash

    I think it's the room that is low, you and Ge0rG are probably normal

  283. jonasw

    Ge0rG, ping

  284. mathieui

    (no ipv6 at cisco :x)

  285. jonasw

    what’s the etiquette for enqueing a remark to the discussion? I’m not sure if people in the room are giving hand signs.

  286. Ge0rG

    jonasw: back

  287. jonasw

    Ge0rG, meeting resumed

  288. jonasw

    topic is impromptu mucs

  289. Ge0rG

    I see backs

  290. jonasw

    cheers, Zash

  291. dwd

    jonasw, W're waving hands about. I suggest you shout, and/or say here you've a comment.

  292. dwd

    If you say here, we'll try and notice and pass on your hand waving.

  293. jonasw

    heh

  294. winfried

    privacy by default, sharing optional

  295. Ge0rG

    Having an "impromptu MUC group creation" was on my 45 agenda for a while, but first I wanted to implement it

  296. jonasw waves hand

  297. jonasw

    (cc @ dwd maybe)

  298. Ge0rG

    jonasw: I can't see you waving :P

  299. dwd

    Much discussion about hand waving. You're covered.

  300. Ge0rG

    Zash: now I understand what you meant by *LOUD*

  301. jonasw

    Ge0rG, you were as loud :)

  302. Ge0rG

    jonasw: sorry

  303. jonasw

    wee message routing

  304. dwd

    "Temporarily". Kev demonstrating optimism there.

  305. jonasw

    haha

  306. jonasw

    my thought exactly

  307. jonasw

    but then he said "on a single jid" and it seemed more realistic

  308. Zash

    Until the heat death of the universe then

  309. MattJ

    Obligatory mention of vcard-temp

  310. Ge0rG

    Did the conf video collapse for everyone or just for me?

  311. MattJ

    Ralph is "fixing" it

  312. jonasw

    while you’re messing with things

  313. jonasw

    could you make it louder?

  314. jonasw

    it’s really quiet for us remotelers

  315. jonasw

    which is awful when somebody non-remote says somtehing

  316. dwd

    ralphm: ^^

  317. jcbrand

    I'm sorry but the etherpad is doing weird stuff, I don't trust it with taking notes...

  318. jcbrand

    I'll make the notes locally for now and then upload them later

  319. jonasw

    seems good

  320. jonasw

    I didn’t even konw that there was a pad :)

  321. Zash

    jonasw: that is the right thing to do

  322. Zash

    don't trust etherpads

  323. Zash

    they eat your data

  324. dwd

    It's how they survive in the wild.

  325. jonasw

    hand-wavy note from the floor: why would IoT even worry about the Archive? it could just disable archive and not worry about getting pubsub archived?

  326. Zash

    SIMS!

  327. Ge0rG

    "Resource Locking"

  328. jcbrand

    The notes are here now, I'll update them from time to time: https://xmppsummit2018.opkode.com/2018-Summit%20Day%20one.html

  329. jonasw

    404

  330. Ge0rG

    jonasw: NET::ERR_CERT_COMMON_NAME_INVALID

  331. jonasw

    it redirects to opkode.com

  332. Ge0rG

    then 404s.

  333. jonasw

    yeah

  334. jonasw

    jcbrand, I got redirected here: https://opkode.com/2018-Summit%20Day%20one.html

  335. jonasw

    dwd, handwave

  336. Ge0rG

    And then somebody will attempt resource locking to bare-JID, breaking even more

  337. MattJ

    Ge0rG, hmm?

  338. Ge0rG

    MattJ: when rewriting `from` from full to bare

  339. goffi

    jcbrand: s/Moving/Movim/

  340. MattJ

    I don't understand "resource locking to bare-JID"

  341. Tobias

    yeah...these words in that sequence don't make much sense imo

  342. Tobias

    Ge0rG, can you rephrase that?

  343. Ge0rG

    maybe I misunderstood Kev's proposal as rewriting the *from* JID

  344. Ge0rG

    so a client would resource-lock to bare JID after receiving a message from a bare JID

  345. dwd

    jonasw: I was doing something else when you did the handwave, and instead it came up on my watch, which was really quite impressive.

  346. Ge0rG

    Gajim used so send normal-to-full, a looong time ago

  347. Ge0rG

    Gajim used to send normal-to-full, a looong time ago

  348. Ge0rG

    yaxim never implemented resource locking

  349. jonasw

    same for aioxmpp. when I was about to do it, people were already ranting against it and I was like "eh, so why bother"

  350. jonasw

    I’m gonna get me some tea, my throat hurts, see you later

  351. mathieui

    jonasw, you’re lucky, we got it in poezio before the complaints

  352. Ge0rG

    mathieui: if only you had asked me :P

  353. Link Mauve

    I removed it just recently.

  354. Ge0rG

    Link Mauve: I was positively surprised

  355. jonasw

    ok, there’s something about crypto now, I missed a context switch, where are we at?

  356. mathieui

    jonasw, entropy in message IDs

  357. jonasw

    ah

  358. Link Mauve

    jonasw, generating random ids.

  359. jonasw

    yah

  360. Zash

    > With entropy in message ids, you don't need encryption

  361. Ge0rG

    the automatic camera is awesome.

  362. jonasw

    jcbrand, if you make your screen a bit brighter and font larger, we don’t need an etherpad :-)

  363. jonasw

    it’s weird too, Ge0rG

  364. Zash

    Is that Edvins or Daves back of the head we got earlier?

  365. jcbrand

    stanza-ids

  366. jcbrand

    jonasw: better?

  367. Ge0rG

    I just realized that we *can* see Kev - mirrored in the glass door

  368. jonasw

    jcbrand, I was mostly kidding, but your link still 404s :(

  369. dwd

    Onscreen is Edwin's head.

  370. Ge0rG

    A client could be OOM killed at any time, so it needs to persist the non-persistent message IDs.

  371. jcbrand

    jonasw: https://xmppsummit2018.opkode.com/

  372. jonasw

    redirects to opkode.com :(

  373. Ge0rG

    How long is "this long"?

  374. jcbrand

    hmm

  375. jcbrand

    https://xmppsummit2018.opkode.com/2018-Summit%20Day%20one.html

  376. jcbrand

    jonasw ?? ^

  377. jonasw

    same thing, but my browser might’ve cached the redirect from your first attempt

  378. jonasw

    doesn’t work even with fresh cache

  379. mathieui

    id='cb1ef50e-36cf-4b1f-90d0-0a1808e10676-1bdfd5b7-dc4a-4444-b13e-4d045f17c12a'

  380. Zash

    Hrm, how do you get chromium to pick a different camera?

  381. Ge0rG

    we need the message ID to consist of two UUIDs, separated by a "/"

  382. jonasw

    Zash, for sending?

  383. Ge0rG

    one generated by the client, the other by the server

  384. Zash

    jonasw: yeah, can't seem to switch to an external webcam instead of the built in laptop one

  385. Ge0rG

    or maybe a chain of UUIDs.

  386. Ge0rG

    every hop adds one.

  387. mathieui

    Ge0rG, do I hear blockchain?

  388. Ge0rG

    mathieui: Yes!

  389. jonasw

    Zash, I’m not sure if it let me pick one when I first connected

  390. Zash

    jonasw: I only had one camera at first, plugged in a usb webcam now.

  391. Ge0rG

    The UUID chain idea is actually half-serious.

  392. jonasw

    Zash, maybe re-join the thing?

  393. Zash

    Camera selection dropdown field is disabled tho

  394. Zash

    And I can't select one when joining

  395. jonasw

    hm

  396. jonasw

    weird then

  397. Ge0rG

    client sends id="foobar", server forwards as id="foobar/bazboom", MUC stores as id="foobar/bazboom/dadada"

  398. jcbrand

    jonasw: I think it caches the redirect yes

  399. Ge0rG

    jcbrand: curl doesn't cache. curl says: < Location: https://opkode.com/

  400. Ge0rG

    (301)

  401. mathieui

    (jcbrand’s link works for me fyi)

  402. jonasw

    huh

  403. dwd

    Ge0rG, I think we run the risk that this might involve about 6 or 7 segments by the time it gets to a client again

  404. jcbrand

    Seems to be an ipv6 issue

  405. dwd

    Otherwise, it's horribly close to being possible.

  406. jonasw

    jcbrand, oh, that may make sense, I’ve got v6 here

  407. mathieui

    dwd, think of the j2j transports, we can recurse into infinity

  408. Ge0rG

    dwd: the only problem I see is how to find the right segment again

  409. jcbrand

    should work now (hopefully cert is fine)

  410. jcbrand

    jonasw: ^

  411. Ge0rG

    jcbrand: works for me now

  412. Ge0rG

    Please don't store MUC messages in user's archive.

  413. jonasw

    jcbrand, \o/

  414. dwd

    Ge0rG, I'm going to store everyone's messages in my archive.

  415. Zash

    *everyones*

  416. jonasw

    so that’s why everyone wants OMEMO. thanks dave.

  417. SamWhited

    I'm storing Skype messages in my archive.

  418. Ge0rG

    dwd: I'll involve the data privacy officer.

  419. dwd

    Ge0rG, We haven't reached GDPR doomsday yet. I'm safe.

  420. jonasw

    Ge0rG, take care, you with your IP logging ;-)

  421. jonasw

    re mentions I’d like to throw in the idea I had some time ago. the idea was essentially to allow clients/users to publish a list of terms (in PEP or in presence or w/e) on which they highlight

  422. jonasw

    that would probably tie in well with "detecting mentions in archive server-side"

  423. Ge0rG

    jonasw: also for push push-through

  424. Ge0rG

    jonasw: minus E2EE

  425. Ge0rG

    let's abolish E2EE

  426. Zash

    E2EE, nice things. pick one.

  427. SaltyBones

    jonasw: I think that is exactly what Kev proposed for push/CSI/battery safe/notifications

  428. jonasw

    ah maybe, I might’ve missed that

  429. Ge0rG

    jcbrand: have you written down everything from my presentation slides? :>

  430. Zash

    Did someone say something that could boil down to having mod_firewall rules for what to carbon, what to archive etc?

  431. jcbrand

    Georg: I was still feeling "fleissig"

  432. jcbrand

    that was before lunch

  433. jonasw

    Zash, "if xmpp2 && fulljid, no archive, no carbon. if xmpp2 && barejid, archive and carbon"?

  434. Ge0rG

    jcbrand: ah, IC. It'd be nice to have some colorcoding of what was said vs. what was on the slides

  435. Ge0rG

    I'm out of the WebEx, need to do some piled up work. ping me when interesting things are happening again :)

  436. jonasw

    damn, I was about to do the same thing :)

  437. SouL

    30 minutes break, just if you couldn't hear

  438. jonasw

    oh

  439. jonasw

    I assumed some kind of break already, but wasn’t sure how long

  440. Ge0rG

    SouL: I was getting coffee

  441. SouL

    I'm the guy with purple squares

  442. SouL

    Helloo :D

  443. jonasw

    I’m disconnected

  444. SouL

    I will write when we are back again

  445. Zash

    Squares?

  446. jcbrand

    Ge0rG I'm not sure how to add color coding, will check

  447. SouL

    Zash, shirt, next to SamWhited x)

  448. Ge0rG

    jcbrand: or replace the slide content with a link to the PDF (I also have ODP) and the page number

  449. jcbrand

    Ge0rG What's the URL to the PDF?

  450. Ge0rG

    jcbrand: https://op-co.de/tmp/whats-wrong-with-xmpp-2017.pdf

  451. jcbrand

    Ok, I'll update

  452. jcbrand

    BTW, the notes are in wiki syntax (because I use vimwiki), so I can later put them on the XSF wiki

  453. Zash

    which wiki syntax is that?

  454. intosi

    Good man.

  455. jonasw

    mediawiki

  456. jcbrand

    https://github.com/vimwiki/vimwiki

  457. jcbrand

    yes, mediawiki AFAIK (for vimwiki)

  458. jonasw

    it’s only similar to mediawiki

  459. jonasw

    mediawiki uses '''x''' for bold, not *x*, and I think nested lists work entirely different (** instead of indent + *)

  460. jcbrand

    ah

  461. jcbrand

    can be fixed with vim regex

  462. Ge0rG

    Good man.

  463. dwd

    We are going to start again.

  464. jonasw

    dwd, wave

  465. Ge0rG

    What's the agenda?

  466. MattJ

    Kev's scribbled list on the whiteboard, not sure if anyone can read it but him

  467. Zash

    Is there a camera pointed at the whiteboard?

  468. MattJ

    Current topic is "inbox"

  469. intosi

    MattJ: jc's transcription should have them as well.

  470. Ge0rG

    I haven't seen Kev at all yet, except briefly as a mirroring in the glass door

  471. Ge0rG

    what's "inbox"?

  472. intosi

    I also posted the list in this room this morning.

  473. goffi

    I've catched late, what's this inbox thing ?

  474. goffi

    same as activityPub ?

  475. jonasw

    goffi, Ge0rG the list of currently open conversations

  476. MattJ

    A server-side shared list of open chats

  477. jonasw

    no

  478. jonasw

    yah

  479. Ge0rG

    does it make sense to have?

  480. Ge0rG

    probably yes

  481. jonasw

    I think so

  482. MattJ

    If you have lots of contacts (imagine you're part of an organisation with thousands of members) you want to keep track of open conversations

  483. jonasw

    it’s a tricky issue I’ve been pondering in JC for a while

  484. goffi

    would be neat yes

  485. jonasw

    having clients collaborate on that would be cool

  486. SouL

    Ge0rG, I'm guessing is not something that would be used in Yaxim?

  487. jonasw

    but I’m not sure if it should affect MAM queries

  488. Ge0rG

    SouL: why not?

  489. Ge0rG

    SouL: I'd like to have a "recent conversations" view, but it needs significant refactoring

  490. Zash

    Prosody?

  491. SouL

    Ge0rG, ah ok, I was talking about how is it now. Nothing to say then.

  492. MattJ

    Prosody keeps being blamed for "why not just use PEP?"

  493. MattJ

    ^ Zash

  494. goffi

    it's really hard to understand Daniel, he's far from mic

  495. Zash

    Someone wanna help with finnishing the node config persistence bits?

  496. jonasw

    I can’t into lua :(

  497. MattJ

    I can't out of Lua :(

  498. goffi

    I'm still advocating for external PEP/Pubsub component, there are still lot to do on Pubsub and servers will not catch before years.

  499. daniel

    > Someone wanna help with finnishing the node config persistence bits? Maybe

  500. daniel

    If there is a todo list. Or unfinished but pushed code

  501. jonasw

    are we still talking about the "inbox" pubsub node?

  502. daniel

    jonasw: yes

  503. jonasw

    uh

  504. jonasw

    removing read conversatinos from that sounds bad

  505. jonasw

    just because a conversation is read I wouldn’t want it to disappear from the list

  506. jonasw

    it should take user interaction to be removed

  507. MattJ

    +1

  508. jonasw

    or did I misunderstand that, daniel?

  509. Ge0rG

    jonasw: if I remove a conversation on Conversations, do I want it to vanish from poezio, too?

  510. MattJ

    Yes

  511. jonasw

    Ge0rG, I *think* there’s a point in that, but I’m also not sure if I’d always want that

  512. jonasw

    because e.g. I don’t want conversations to sync #dovecot on freenode on each connect

  513. Zash

    send that chat state that says you ended the convo?

  514. jonasw

    but I still want it on all non-mobile clients.

  515. jonasw

    Zash, smart

  516. SouL

    Could we have a "Close from here" and "Close from everywhere"?

  517. Zash

    jonasw: until we start doing chat states over presence, and there's no carbons anymore

  518. jonasw

    SouL, I don’t like that UX

  519. jonasw

    (for conversations at least)

  520. jonasw

    Zash, well, the server would obviously update the pubsub node

  521. Zash

    ah, sure, that would work

  522. Ge0rG

    SouL: that's a horrible hack.

  523. Zash

    I'd like to remind y'all that if you make MAM impossible to implement on append-only data structures, I will cry

  524. MattJ

    Make sure your webcam is turned on

  525. Ge0rG

    MattJ: now you made me want to see that too

  526. jonasw

    Zash, it is possibel, like you can implement anything append-only, but it might become very inefficient to access

  527. Zash

    Something Smart™

  528. jonasw

    some video breakage here, is it just me?

  529. Ge0rG

    jonasw: here too

  530. jonasw

    ah fixed

  531. Zash

    and here

  532. jonasw

    daniel, +1

  533. Ge0rG

    the good shit, the real shit? Audio is a little bit blocky

  534. Zash

    daniel: It's started in the sense that Link Mauve added content persistence. Configuration persistence isn't there, apart from whether to enable persistence, which is something of a hack atm.

  535. Zash

    Something like what?

  536. SamWhited

    Zash: Slack, HipChat, etc.

  537. Zash

    Ah, then me neither

  538. goffi

    notifications would be neat to keep server side too. For now we have to keep track of messages

  539. Zash

    Is this the thing were mod_firewall fits in? :)

  540. daniel

    Note that I personally care more about configurability than persistence

  541. MattJ

    The current discussion is about notification preferences being stored server-side, not actual server-side notifications

  542. daniel

    I don't restart my server 😀

  543. intosi

    In buggy CPU world, cloud hoster restarts your server for you.

  544. jonasw

    first time I’m seeing kev!

  545. mathieui

    ralphm found the camea manual controls, I think

  546. mathieui

    ralphm found the camera manual controls, I think

  547. Ge0rG

    jonasw: I was just thinking that

  548. jonasw

    not to mention that SCRAM takes some power too

  549. jonasw

    depending on the number of iterations

  550. Zash

    jonasw: cacheable tho

  551. jonasw

    SCRAM actually took most of the time for establishing a connection on my MIPS-based router back then

  552. jonasw

    Zash, only partially, isn’t it?

  553. Zash

    jonasw: If the salt is fixed, which it should be, then you can cache the SaltedPassword

  554. Zash

    then it's just some XOR and HMAC

  555. jonasw

    I’d like to throw in that these are possibly details which should be up to implementation experimentation?

  556. jonasw

    who’s the guy next to dave with the erlang sticker?

  557. Zash

    I doubt fast reconnect can become as cheap as just sending <{csi}active> on an existing TLS connection

  558. jonasw

    Zash, I tend to agree

  559. Zash

    Unavailable presence instead of csi-inactive?

  560. jonasw

    dwd, I’d like to throw in: you might want to discover the <identity/> of a client

  561. Ge0rG

    I really love Holger's hack of keeping an SM detached session as long as push is registered.

  562. jcbrand

    jonasw: Nicolas Vérité

  563. jonasw

    and I’d also like to throw in that servers are allowed to intercept dicso#info and answer on your behalf if you do caps

  564. Zash

    How far away is multisomething-tcp?

  565. mathieui

    jonasw, nyco

  566. Ge0rG

    Zash: very far away

  567. Zash

    in theory

  568. Ge0rG

    If I had to bet on MIX vs multi-path TCP, I'd put my money on Duke Nukem Forever.

  569. Zash

    Ge0rG: That joke doesn't work anymore

  570. Ge0rG

    Zash: I feel old now.

  571. Zash

    I for would would put my money on GNU Hurd 1.0

  572. Zash

    I for one would would put my money on GNU Hurd 1.0

  573. jonasw

    who’s the one in purple next to dave?

  574. Tobias

    https://wiki.openssl.org/index.php/Simple_TLS_Server "You should avoid 0-RTT if possible. In fact, an organization's data security policy may not allow it for some higher data sensitivity levels. "

  575. jonasw

    ah, hantu.sc

  576. intosi

    The entire section on that page reads as a "Here be dragons" regarding 0-RTT

  577. SouL

    .w cop

  578. SouL

    Ignore that message, sorry.

  579. intosi

    No space left on device

  580. Zash

    Someone print that on a t-shirt and give to SouL

  581. vanitasvitae

    Hah, i can access xmpp via tor when in the cisco wifi.

  582. goffi

    I could not follow all the discussion. About XMPP 2.0 would it be a RFCs or XEPs ?

  583. goffi

    RFCs seems more logical, but it will take years

  584. Zash

    What is XMPP 2.0?

  585. Ge0rG

    goffi: AFAIU it was decided not to put into the IETF process

  586. goffi

    Zash: the stuff which was discussed for a good part of the day, change in the heart of the protocol to fix broken stuff, cf. https://wiki.xmpp.org/web/XMPP_2.0

  587. goffi

    Ge0rG: OK, but that seems weird to have major changes without doing RFCs

  588. Ge0rG

    goffi: the changes might not be as major as anticipated

  589. goffi

    Zash: I've missed good part of the discussions, but it was about stanza ids, removing resource locking, putting stuff from presence to PEP

  590. goffi

    Ge0rG: OK, I'll guess it will be discussed on the mailing list anyway, so I'll see there.

  591. Ge0rG

    from message to presence, from presence to PEP

  592. Ge0rG

    from PEP to message.

  593. Zash

    goffi: That was actually meant as a joke.

  594. Zash

    But what does it mean or something.

  595. SaltyBones

    I have an idea on how to build stable IDs. Who could I give that to for review/shredding? :)

  596. Ge0rG

    SaltyBones: write a proto-XEP, or ask tomorrow

  597. Zash

    Is it (stream id, message counter) ?

  598. Zash

    Wait, stable ids for what?

  599. SaltyBones

    Zash: are you volunteering to discuss it? ;)

  600. jonasw

    I still don’t quite understand what’s wrong with putting 256 bits entropy into a random ID and be done with it.

  601. jonasw

    on conflict re-roll

  602. jonasw

    generate IDs on the server

  603. jonasw

    what’s the problem?

  604. Zash

    Which IDs is this about?

  605. SamWhited

    I think we were talking about the client earlier; you can't trust the client to generate IDs correctly as ralph pointed out. The server may be a different story.

  606. Zash

    But the client wants to know the ID the server picked, and be able to associate it with whatever internal ID it might have

  607. jonasw

    return the ID in <{sm}a/>

  608. jonasw

    :>

  609. jonasw

    (the IDs)

  610. jonasw

    (in a message/@id -> stanza-id mapping)

  611. Zash

    I thought the idea was to extend carbons to reflect messages back

  612. Zash

    to the sender, that is

  613. jonasw

    yeah, that’d work too

  614. Zash

    not just to your other clients

  615. Zash

    Tho the exact formatting of the message/@id → stanza-id mapping carrier probably doesn't matter that much

  616. jonasw

    yo

  617. SaltyBones

    The problem with server generated IDs is that you have to reflect them to the client. This is not a big problem but it seemed like people would have preferred to omit it.

  618. jonasw

    it’s the only way to be sure though

  619. SaltyBones

    To be sure of what?

  620. alameyo

    I think I am on the proper street but I am searching for Thon Hotel

  621. alameyo

    oh I see it