XSF Discussion - 2020-08-31

  1. flow

    https://blogs.gentoo.org/mgorny/2020/08/25/is-an-umbrella-organization-a-good-choice-for-gentoo/ reminds me a little bit about us (i.e. the XSF)

  2. flow

    not sure how the XSF's standing with the IRS is ;)

  3. Guus

    I am not aware of any issues with the IRS.

  4. Ge0rG

    So that recent IETF thread make me think... are we able and willing to offer our help to the IETF to run a modern XMPP server for them?

  5. jonas’

    willing yes

  6. jonas’

    able: I’d be happy to support with that if it isn’t open-registration

  7. pep.

    Can somebody tldr the thread? "hey we're in 2020, we want features. Here have a Slack instance"?

  8. Ge0rG

    pep.: kinda sorta

  9. Ge0rG

    pep.: also XMPP prejudice based on experiences from 2005

  10. jonas’

    so I can’t in the next month, but I’d be happy to help supporting in the long term

  11. jonas’

    I have some experience in monitoring things

  12. pep.

    Do they have clients for the server we'd provide?

  13. Ge0rG

    jonas’: that's something that needs to be done right now, though

  14. pep.

    With features(tm)

  15. Ge0rG

    pep.: educating them about modern clients would be part of the task

  16. jonas’

    Ge0rG, I can’t do right now, I have literally no time for any additional tasks until the move is over

  17. jonas’

    the remaining buffer time is reserved for emergencies

  18. jonas’

    (and user requests like "add me to o.j.n")

  19. jonas’

    Ge0rG, but we can surely send out the offer and then find someone to do it

  20. Ge0rG

    jonas’: maybe my comment was not directed to you as much as a general call to action for the XSF / Board / iteam

  21. pep.

    Ge0rG: you mean the great web clients we have? :p

  22. jonas’

    maybe we can get Holger to provide them with a free account and C download?

  23. jonas’

    Holger + Daniel that is

  24. jonas’

    Ge0rG, or you could spin up a vhost on yax.im?

  25. jonas’

    as a demo server

  26. Ge0rG

    pep.: Converse, Conversations and Siskin should cover most users, right?

  27. Ge0rG

    jonas’: sure thing.

  28. Ge0rG

    jonas’: just that yax.im isn't exactly a highly available machine

  29. jonas’

    true again

  30. jonas’

    I could also add a vhost to the s.j.n host (where I can comfortably give others shell)

  31. Ge0rG

    and it's got a bus factor of 1

  32. pep.

    xmpp.org doesn't have any users btw? (unrelated)

  33. Zash

    apart from memberbot, no

  34. pep.

    maybe apart from memberbot

  35. Zash


  36. Ge0rG

    jonas’: don't get me wrong, I'd immediately set up whatever is required to host user accounts on my box, it's just not the perfect place to convince IETF people of the awesomeness of xmpp

  37. Zash

    IETF does have an ejabberd running, so it someone with ejabberd know-how might be able to help them hook that up some SSO with their document management thing

  38. jonas’

    I’ll also be happy to provide a monitored empty debian-buster box.

  39. Ge0rG

    jabber.ietf.org is just a MUC server

  40. Ge0rG

    Info> ietf.org is running ejabberd version 19.08 Oh.

  41. Ge0rG

    wow, the list is so long I can't scroll it

  42. jonas’


  43. Ge0rG

    but 19.08 should be sane enough for all of their requirements, right?

  44. jonas’

    what are their requirements? ;)

  45. jonas’

    but probably yes

  46. Ge0rG

    jonas’: "modern IM"

  47. jonas’

    19.08 should do then I guess

  48. Ge0rG


  49. Ge0rG

    I suppose we'd need Holger to volunteer re ejabberd maintenance

  50. jonas’


  51. Ge0rG

    Error> No Contact Addresses for ietf.org

  52. Ge0rG


  53. jonas’

    and maybe also jcbrand to support a converse setup there

  54. pep.

    Are we volunteering Holger already

  55. jonas’

    pep., yeah of course, mdosch was volunteered to be member of editor@ just a few hours after his election ;)

  56. jonas’

    can’t wait that much longer with holger!

  57. pep.


  58. Ge0rG

    heh, one of the users in the IETF hallways is running on yax.im

  59. Holger

    Yes yes. I'll ask the XSF to double my member's salary and we're good to go.

  60. jonas’

    Holger, GRANTED! There you go.

  61. MattJ

    You know you need Board approval for that

  62. MattJ

    Come on

  63. jonas’

    oh sorry

  64. jonas’

    MattJ, can I? ;)

  65. MattJ

    You'll have to wait a few weeks

  66. Ge0rG

    MattJ: well, technically the only "salary" you get is the Summit Dinner, and you can bring another person, so...

  67. MattJ

    One week to approve the discussion, one week for people to be busy, another week to decide we need a work team to approve salaries

  68. jonas’

    I find your lack of faith disturbing

  69. MattJ

    Oh I have faith, it will be approved ;)

  70. Zash

    Before or after the next Board election?

  71. MattJ


  72. Ge0rG

    BTW, https://www.ietf.org/how/meetings/jabber/

  73. Guus

    > Yes yes. I'll ask the XSF to double my member's salary and we're good to go. You've been in almost a week now, right? 🙂

  74. eta

    fwiw I'd also be willing to help out for the IETF if spare hands are needed

  75. Guus

    I think this is more about being able to guarantee long-term support than to find volunteers that can do things right now.

  76. pep.

    Ge0rG, it'd be great to have them update that page yeah :/

  77. Guus

    also see: jabber.org

  78. Ge0rG

    looks like Glen is the server admin, see https://mailarchive.ietf.org/arch/msg/ietf-announce/dGQaXplloQp9J8CwKFhWWt2IDJY/

  79. Guus

    interesting that he lists ignite as the first server

  80. Guus

    at that time, I don't think it had registrations enabled.

  81. Ge0rG

    So we should: - contact Glen directly, maybe offering help to set up ietf.org as the xmpp user domain for the members? - send a PoC suggestion to the ML, offering to run a webchat fronted for them? - offer a PR to that page?

  82. Guus


  83. jonas’

    Ge0rG, yes & yes & yes

  84. Ge0rG

    in what order?

  85. jonas’

    we should also get glen into operators@

  86. Zash

    By the power of "if you can get two others to agree with your proposal, go ahead!", .... go for it

  87. jonas’

    Ge0rG, go with Glen first

  88. Ge0rG

    CC to anybody?

  89. jonas’

    feel free to put any of the email addresses you have of me in CC and tell them that I’ll also be happy to help with trying to establish links to other operators and debugging remote issues

  90. jonas’

    since I have some experience with that as I run s.j.n

  91. Ge0rG

    and o.j.n.

  92. jonas’

    also point them at o.j.n because we by now ... exactly.

  93. jonas’

    also point them at o.j.n because we by now have a bunch of prober accounts where we do have contact info to trace down issues

  94. Ge0rG

    anybody else wants to be in CC?

  95. jonas’

    point them at operators@ (both the ML and the chat) and invite them to also add me directly to their contacts if that helps

  96. jonas’

    (though in that case, I prefer the @zombofant.net JID)

  97. jonas’

    I can’t help with ejabberd admin unfortunately, but I’ll be happy to help with anything I can do around that.

  98. Guus

    Ge0rG if needed, I'd be happy to help with Openfire-specific issues that they might have (I read that they push that impl) - I'd actually love to get their feedback on that. I'll be less involved with the more generic solution though

  99. Ge0rG

    jcbrand: do you happen to be awake and available?

  100. Zash

    I can probably (try to) help with Prosody connectivity issues, if there's a need for one dev per server impl 🙂

  101. jonas’

    it certainly doesn’t hurt

  102. jonas’

    pity that stpeter isn’t around

  103. Ge0rG


  104. MattJ

    stpeter has previously tried to encourage them to improve things here, not sure who he was talking to

  105. jonas’

    FWIW, I’ll also be happy to provider longer-term support for this issue, I can’t do the short-term stuff right now though.

  106. jonas’

    (cc @ Guus ^)

  107. Ge0rG

    > anybody else wants to be in CC?

  108. jonas’

    (and maybe don’t mention that in the email, Ge0rG :))

  109. pep.

    Ge0rG, does it have to be a private email? Can it happen on operators@?

  110. Ge0rG

    pep.: it might be impolite to invite somebody to an ML by crossposting there

  111. pep.


  112. Guus

    also, this is an offer specific to IETF - not something suitable for the more generic operators mailinglist.

  113. pep.

    Guus, sure, that doesn't prevent us from doing that in public

  114. jonas’

    Ge0rG, for the PoC suggestion: I unfortunately 100% lack experience with running converse.js, so I can’t really support there.

  115. Ge0rG

    jonas’: https://op-co.de/tmp/glen-xmpp-mail.txt

  116. jonas’

    Ge0rG, expand the first reference to XSF

  117. Ge0rG

    > > anybody else wants to be in CC?

  118. jonas’

    Ge0rG, link to https://xmpp.org/chat#converse/room?jid=operators@muc.xmpp.org for the operators@ room

  119. Ge0rG

    jonas’: added

  120. Holger

    Ge0rG: Feel free to add me.

  121. Ge0rG

    Holger: also as a cosignatory?

  122. jonas’

    Ge0rG, I’d like to re-read this with the fixes, can you let me know when you updated it?

  123. Holger

    Ge0rG: Sure (as you see fit).

  124. Ge0rG

    jonas’: updated

  125. jonas’

    Ge0rG, maybe add affiliations to the signatures? e.g. Georg Lukas (yax.im admin, JabberSPAM founder) Jonas Schäfer (observe.jabber.network and search.jabber.network operator) Holger Weiß (conversations.im operator)

  126. Ge0rG

    jonas’: thought about that

  127. Ge0rG

    Holger is also an ejabberd developer, can't leave that out

  128. jonas’

    oh, very true

  129. jonas’

    I thought about how to slip an ejabberd affiliation in there, but I didn’t come up with the obvious

  130. pep.


  131. Ge0rG

    another update

  132. jonas’

    Ge0rG, if we want to further polish this, s/webchat/anonymous webchat/

  133. Ge0rG


  134. jonas’

    s/better service monitoring/additional service monitoring/

  135. jonas’

    since we don’t know what they currently have

  136. Ge0rG


  137. jonas’

    and claiming that o.j.n will be "better" is presumptuous or however you type thath word

  138. jonas’

    and claiming that o.j.n will be "better" is presumptuous or however you type that word

  139. Ge0rG

    Holger: any feedback from you?

  140. Guus

    Don't overpolish things - this is great.

  141. Ge0rG

    mail sent

  142. Holger

    Yes, looked good to me.

  143. jonas’

    Ge0rG, thank you very much for typing the words

  144. jonas’

    someone explain to me why fairemail includes the trailing / in the linkification of o.j.n, but not for s.j.n

  145. Holger

    They should add an option for whether or not to include trailing /.

  146. Ge0rG

    Okay, now does it make sense to respond to https://mailarchive.ietf.org/arch/msg/ietf/DHy4rk5qxBjIb4hOTRH0E3pFJqg/ ?

  147. jonas’


  148. pep.

    I think I would have responded there in the first place rather

  149. Ge0rG

    pep.: but you haven't

  150. pep.

    I haven't :)

  151. jonas’

    First, rebut the individual non-facts. Second, hint that we contacted Glen off-list to offer support.

  152. Ge0rG

    jonas’: I don't want to go to their place and start a fight about perceived realities

  153. jonas’

    Ge0rG, then skip the reply for now and see what Glen says

  154. Ge0rG

    are there any of the suggested steps that we could do right now, proactively?

  155. Ge0rG

    like... rm -rf pidgin from the internet

  156. jonas’

    publish a "Pidgin considered harmful" blogpost? ;D

  157. Ge0rG

    also why is adium listed on https://xmpp.org/software/clients.html

  158. jonas’

    because neutrality?

  159. Zash

    Is it maintained? Who renewed it?

  160. Ge0rG

    Zash: Robbie <robertvehse@fastmail.fm> did

  161. mdosch

    No Dino or profanity there. This em thing doesn't look like an xmpp client.

  162. mdosch

    No Dino, poezio or profanity there. This em thing doesn't look like an xmpp client.

  163. pep.

    Nor Conversations

  164. jonas’


  165. jonas’

    the latter is really a problem

  166. Zash

    Looks like all those were renewed last 14 months ago

  167. Ge0rG

    I'd really like the maintainers to do that job

  168. Zash

    Maintainers of popular modern clients are likely busy with maintainging/developing the client

  169. Ge0rG

    Otherwise the periodic update to get rid of unmaintained clients is pointless

  170. Zash

    And unmaintained clients have gathered a bazillion fans over the years, who will never let them get kicked off the list

  171. pep.

    Here, poezio and slix PR

  172. pep.

    I personally think that list is harmful, but :x

  173. Zash

    Anyone wanna build a Better List?

  174. pep.

    https://github.com/xsf/xmpp.org/pull/775 somebody plz(?)

  175. moparisthebest

    iirc the thought behind the auto-expiry is that unmantained clients would fall off and that only good active ones would be left, but it looks like the opposite happened

  176. moparisthebest

    unintended consequences are the best kind

  177. pep.

    I don't mind PR for the other 3 clients that were mentioned

  178. pep.

    4. Seems like Movim isn't in there either

  179. moparisthebest

    still unclear, I thought the rule was only the developer could PR their client in

  180. pep.

    Nor SàT..

  181. Zash

    Hmmmmm/ https://github.com/xsf/xmpp.org/pull/588

  182. pep.

    moparisthebest, happy to wait for an ACK

  183. Ge0rG

    Zash: yeah.

  184. moparisthebest

    ah ok, glad I'm not the only one unclear on it

  185. pep.

    Nobody is clear with that.. :/

  186. moparisthebest

    that looks like some nice board AOB for you there pep. :)

  187. pep.

    It was already discussed at some point, and iirc the result was "case-by-case" or sth..

  188. pep.

    don't remember exactly

  189. Zash


  190. moparisthebest

    without criteria that doesn't seem helpful

  191. Ge0rG

    people should just read https://xmpp.org/2017/03/new-xmpp-software-listing-rules/ ;)

  192. Zash

    Could someone not on a corporate workstation add that link to that PR?

  193. pep.

    To what PR

  194. Zash

    Link to https://logs.xmpp.org/xsf/2019-07-25?p=h#2019-07-25-ea7cec0e0a9fc4ca in https://github.com/xsf/xmpp.org/pull/588

  195. pep.

    Ah, that was the PR that was problematic, right

  196. Ge0rG

    but that PR got merged already

  197. Zash

    For future reference etc

  198. pep.

    Seems like everybody renewed their clients on 07-16..

  199. pep.

    Probably a batch operation already

  200. Zash

    That PR?

  201. pep.


  202. Ge0rG

    all that makes me sad

  203. Zash

    Or batch merge?

  204. pep.

    Zash, probably that PR yeah

  205. pep.

    https://github.com/xsf/xmpp.org/pull/776 Daniel https://github.com/xsf/xmpp.org/pull/777 edhelas https://github.com/xsf/xmpp.org/pull/778 larma

  206. Daniel

    pep.: can I ack in here?

  207. pep.


  208. Daniel


  209. pep.

    It's just for whoever is going to merge

  210. Daniel


  211. jonas’

    I merged all of them

  212. jonas’

    they are active projects, I happen to know that

  213. pep.


  214. pep.

    Seems like the poezio one is still up there :(

  215. pep.

    I know it's not that active..

  216. jonas’

    active enough

  217. edhelas


  218. edhelas

    Happy to know that Movim is active

  219. pep.

    There's also a PR for openfire

  220. jonas’


  221. jonas’

    (I only see a merged one ;))

  222. pep.


  223. pep.

    And still no https://github.com/xsf/xmpp.org/pull/775 ?

  224. jonas’


  225. pep.


  226. jonas’


  227. jonas’


  228. jonas’

    I recently learnt that I have +w on that repo again

  229. Ge0rG

    jonas’: but you knowing about active projects is highly subjective!

  230. Ge0rG

    also I'd like to see compliance badges in there, even if the minimal non-designed ones

  231. jonas’

    Ge0rG, "PRs welcome"

  232. Ge0rG

    jonas’: ah, that interesting tooling.

  233. wurstsalat

    Ge0rG, +1 for Compliance Badges (even basic non-designed ones)

  234. Ge0rG

    wurstsalat: PRs welcome! :DDD

  235. pep.


  236. pep.

    (I'm away before somebody says PR welcome)

  237. wurstsalat

    and yes, DOAP > Compliance extaction

  238. wurstsalat

    maybe unmaintained clients won’t ever get a DOAP file? :)

  239. pep.

    So we accept every project that comes our way (is this really neutral?), but we're very eager not to accept some of them :)

  240. jonas’

    pep., PRs welcome!

  241. pep.

    Ah fail, I was still here!

  242. jonas’

    wurstsalat, interesting idea

  243. jonas’

    pep., moohahahaha

  244. eta


  245. Ge0rG

    I think it would be good to have one or two badges per implementation in that list

  246. wurstsalat

    eta, Description of a Project (an xml file)

  247. Ge0rG

    eta: example at https://yaxim.org/doap/yaxim.rdf.xml

  248. pep.

    eta, https://lab.louiz.org/poezio/poezio/-/blob/master/data/doap.xml

  249. eta


  250. pep.

    It's not used anywhere yet though

  251. pep.

    Just linked from json files in the xmpp.org repo

  252. jonas’

    problem with project-hosted DOAP is that the website build would then depend on the availability of that XML file

  253. pep.

    Until someday somebody does something

  254. jonas’

    which is not cool

  255. jonas’

    and gets less cooler with every additional project doing that

  256. pep.

    jonas’, yeah we might have to cache slightly :/

  257. jonas’

    which requires extra infrastructure

  258. edhelas

    got also mine there https://github.com/movim/movim/blob/master/doap.xml

  259. Ge0rG

    jonas’: would a subdirectory full of doap files inside the data/ directory be better?

  260. jonas’

    Ge0rG, no, because then projects would have two places where they need to keep their DOAP up-to-date

  261. Ge0rG

    for active clients it might be a significant Editor burden to keep up with the PRs

  262. pep.

    btw Link Mauve, https://linkmauve.fr/ns/xmpp-doap, is this supposed to parse error

  263. jonas’

    which is about as bad as the current listing situation

  264. jonas’

    Ge0rG, Editor is not responsible for xmpp.org

  265. jonas’

    Commteam is

  266. jonas’


  267. Ge0rG

    So what would be a serious, usable, approach to keep badges on the implementation list?

  268. Daniel

    Automate with doap is the only thing that will scale imho

  269. jcbrand

    Ge0rG, jonas’ hello, I'm now here and awake

  270. Ge0rG

    jcbrand: we sold you to the IETF to set up a smooth-running converse.js instance so that they don't need to use slack.

  271. jcbrand

    From my reading, you want to help the IETF by hosting a server and webchat (and maybe other things) for them?

  272. Ge0rG

    jcbrand: I sent https://op-co.de/tmp/glen-xmpp-mail.txt to the IETF xmpp admin. Let's see what happens next

  273. jcbrand

    Sounds like it's for a good cause

  274. Ge0rG

    That thread on ietf@ is alread three days old, so we probably should also chime in in a friendly way there and give them a taste of modern XMPP.

  275. jonas’

    jcbrand, good morning!

  276. Ge0rG

    I don't know if we can give them Conversations gift codes or something ;)

  277. jcbrand

    FWIW, I'd be happy to help out and set something up for them.

  278. eta whispers "use pix-art"

  279. Ge0rG

    jcbrand: thanks very much, I'll keep that in mind

  280. Ge0rG

    is the converse instance on xmpp.org up-to-date?

  281. jcbrand

    Probably not, but I never touched it. If someone gives me access, I can try to help there as well. I intend to release Converse 7 soonish (before the end of the month) and can then update it in various places. Having access to deployments like that can also help me test for specific use-cases so that I know something's not broken.

  282. jcbrand

    (before releasing)

  283. Ge0rG

    jcbrand: I have a 5.0.3 instance with a slightly cumbersome path setup running on https://yaxim.org/chat/ where I'd be glad to have a hand

  284. Ge0rG

    I know I had issues with it, as I want it to work for both mobile and desktop, and I wrote about it in https://github.com/conversejs/converse.js/issues/1057

  285. jonas’

    jcbrand, end of month is in less than 24 hours ... are you sure? :)

  286. jcbrand

    jonas’ ahem... end of September

  287. jcbrand

    Ge0rG Yeah, I remember your tickets about mobile issues. I can try to help with updating it to Converse 7, and capturing tickets for issues. I can't promise fixing mobile issues timeously though.

  288. Ge0rG

    I don't rememeber anymore what I did (and if at all) to make 5.0.3 work.

  289. Ge0rG

    Though I must admit, the adium renewal is a good example of the "maintainer must do an update to ensure up-to-date implementation" going wrong.

  290. jonas’

    Ge0rG, that CC escalated *quickly*

  291. Ge0rG

    jonas’: oh yes indeed

  292. jonas’

    but nice and extensive response

  293. pep.

    Are you going to keep the suspense longer? :P

  294. Ge0rG

    pep.: nobody is going to paste from private conversations.

  295. Zash


  296. pep.

    What is this doing here then

  297. moparisthebest

    You could use snikket instead, for the invite stuff

  298. Ge0rG

    TL;DR: we were thanked for our offer but Glen is not the right person, so the mail was forwarded to multiple IETF representatives

  299. jonas’

    TL;DR: not quite the correct person to address, but he CC’d a bunch of other people who are the correct addressees and may have opinions on matters

  300. Ge0rG


  301. jonas’

    good that we read the same email :D

  302. Ge0rG

    Meh. https://github.com/ietf/www.ietf.org

  303. jonas’


  304. Ge0rG

    I thought about submitting a PR to exterminate pidgin

  305. Ge0rG

    and Adium is... well.. as questionable

  306. pep.

    Especially with the existence of alternatives

  307. Ge0rG

    which alternatives are there?

  308. pep.

    BeagleIM iirc

  309. pep.

    And monal maybe?

  310. mdosch

    > Previously issue tracking for new www.ietf.org IETF website. Please send issues to webmaster@ietf.org So email instead of PR :)

  311. Ge0rG

    mdosch: that doesn't scale well

  312. pep.

    Send a patch to that email address*

  313. pep.

    Ah wait there's no code

  314. Ge0rG

    a patch for the HTML?

  315. Zash

    Convert the HTML to Markdown and send a patch for that.

  316. Ge0rG converts Zash to markdown

  317. Zash

    Submit an Internet-Draft

  318. moparisthebest

    yea why don't web browsers have integrated markdown processors yet

  319. Zash

    Why don

  320. moparisthebest

    I'm sorry for suggesting that already

  321. pep.

    They have a subset of markdown

  322. Zash

    Why don't the at least show the plain text?

  323. wurstsalat adds XEP-0441 to DOAP file, but https://xmpp.org/extensions/xep-0441.html still 404s

  324. jonas’

    wurstsalat, oh sorry

  325. jonas’

    wurstsalat, there I fixed it

  326. wurstsalat

    Thank you jonas’ :)

  327. Ge0rG

    We should have called it MAMa, because of how well it cares for our messages

  328. lovetox