XSF Discussion - 2018-09-20


  1. ralphm

    Steve Kille: yes, that's why I said: you can have multiple different identities and disco extension forms in one result.

  2. ralphm

    Not convinced yet you need node='mix' "to make this work".

  3. jonas’

    indeed, merged information could work just as well

  4. ralphm

    The problem I have with the current solution is that I can't just discover from an unknown JID that it is a MIX room. I'd have to know, somehow, for the MIX specific information.

  5. Steve Kille

    ralphm: I don't have time to look at the MIX spec this week and consider this point. I will review next week. Shall I respond in the MUC or by email to the list?

  6. ralphm

    I was already drafting it

  7. Steve Kille

    ta

  8. ralphm

    To be found at https://mail.jabber.org/pipermail/standards/2018-September/035355.html

  9. Steve Kille

    ralphm: ta. Will respond next week

  10. ralphm

    Cool

  11. Maranda

    SamWhited: battery usage reduction with heavy muc usage is consistent between 50-60% on my device with those optimizations using Conversations btw

  12. Ge0rG

    Maranda: what optimizations?

  13. Maranda

    I just had to adjust the queue max buffer amount to make it less detrimental to notifications as well

  14. Ge0rG

    I don't understand

  15. Maranda

    Ge0rG: gimme the time to type will you?

  16. Maranda

    🤸‍♂️

  17. Ge0rG

    Maranda: sorry, I misunderstood your sentence as a response to my question. If only we had threads!

  18. Maranda

    Ge0rG: presence deduplicated queuing, filtering of bodyless message payloads, and queuing of all groupchat type messages except mentions basically (anything not meeting criterias triggers a flush)

  19. Maranda

    Ge0rG: that's the optimization

  20. Ge0rG

    Maranda: okay, nothing surprising here. What's the baseline you are measuring against?

  21. Maranda

    (CSI obviously)

  22. Ge0rG

    Maranda: also do you flush MUC OMEMO immediately?

  23. Ge0rG

    Maranda: CSI is not a baseline, it's a signaling mechanism

  24. jonas’

    (the "(CSI obviously)" was obviously not meant in response to your baseline request)

  25. jonas’

    (because it had parenthesis and doesn’t make sense as an answer)

  26. Ge0rG

    Maybe I should remove myself from the discussion then, I fail to process what I'm reading.

  27. Maranda

    Ge0rG: for now I just tested on my server and gathered numbers from my own usage as I said, hard to find people willing to be sample subjects 😜

  28. Ge0rG

    Maranda: you make a prosody module, I deploy it.

  29. Maranda

    Ge0rG: 😆

  30. Maranda

    I implement both (part of) SIFT and CSI in the same module while Prosody has several submodules hookin' on mod_csi so I'd have to look if porting of portions of the current code is viable

  31. Ge0rG

    Maranda: what do you need SIFT for?

  32. MattJ

    Ha, SIFT

  33. Maranda

    Ge0rG: allowing clients to decide which stanzas to filter but since there're 0 implementations as usual I guess: "nothing"

  34. Maranda

    SIFT was supposed to replace part of the featureset provided by Privacy Lists afair

  35. !xsf_martin

    > Maranda: you make a prosody module, I deploy it. Me too

  36. pep.

    > Ge0rG> Maranda: sorry, I misunderstood your sentence as a response to my question. If only we had threads! haha.

  37. MattJ

    Who do we have today?

  38. ralphm

    Hi!

  39. MattJ

    \o

  40. ralphm bangs gavel

  41. nyco

    _o/

  42. Guus

    hi!

  43. MattJ

    Ok, all we need now is Martin to appear :)

  44. ralphm set the topic to

    XSF Board Meeting | Logs: http://logs.xmpp.org/xsf/ | Agenda https://trello.com/b/Dn6IQOu0/board-meetings

  45. ralphm

    0. Welcome + Agenda

  46. ralphm

    Welcome all. Apologies for not being here for a fit.

  47. ralphm

    bit

  48. ralphm

    Things should be more quiet (schedule-wise) going forward.

  49. nyco

    no pb

  50. ralphm

    Anything in particular we should discuss today?

  51. ralphm

    Elections come to mind

  52. nyco

    Board prios... for next term

  53. Guus

    one sec, someone at door

  54. nyco

    https://trello.com/c/Al4GHsYy/311-proposed-prios-for-the-2018-2019-board-based-on-survey

  55. ralphm nods

  56. jonas’

    how’s the survey going?

  57. MattJ

    jonas’, results were posted here quite a while back

  58. MattJ

    Current discussion is what actions to take based on the results

  59. jonas’

    oh

  60. jonas’

    I must’ve missed them

  61. jonas’

    were they also posted to members@?

  62. MattJ

    It's nearing the end of this term, but it would be good to get a head-start with some recommendations for the next Board

  63. jonas’

    if not, posting them to members@ would be nice

  64. MattJ

    Not sure, if not I can dig them up in a bit

  65. ralphm

    Since we're kind of there:

  66. MattJ

    The minutes should have linked to them, but our minutes have been a bit sketchy as of late

  67. ralphm

    1. Survey

  68. Guus

    (back)

  69. ralphm

    I read the summary on the Trello ticket. Seems reasonable.

  70. MattJ

    Yeah, I think we generally agreed with that. Concrete actions are harder to recommend

  71. ralphm

    Except I think the Board's first priority is always what is listed as 2 here: Organisation (9) => bureaucracy, funding

  72. MattJ

    A lot of things circle back to our finance situation, at the end of the day

  73. ralphm

    For 3 (standards development) we have Council and Editors

  74. MattJ

    I think the key is for Board to enable these things to happen

  75. ralphm

    I am indeed not sure what Board specifically can do itself for the other two points.

  76. ralphm

    Right

  77. ralphm

    We'd still need other people to make thing happen. E.g. I am really happy with the newsletter by JC and others

  78. MattJ

    Well I think the big thing the survey highlighted for me is that everyone agrees there should be more IRL events

  79. ralphm

    So basically we need to spice up SCAM

  80. MattJ

    pep. organised an unofficial hackathon, which was great... and hopefully we will have some more

  81. MattJ

    It was unclear whether these events can draw from SCAM, so I think that is one thing we can discuss

  82. Guus

    Actually, the should be

  83. MattJ

    I'm not even sure it's unclear to me, just that it could be more explicit

  84. Guus

    (able to draw from SCAM)

  85. ralphm

    Well, at least for promotion

  86. Guus

    but we have not much of protocol nor history, which makes it hard for people to know what to expect from SCAM

  87. MattJ

    I don't want to speak for pep., but for example I suspect it could be clearer what process he would have needed to go through

  88. ralphm

    if there are things needed, like materials, why not see if the XSF can help out there

  89. MattJ

    I think things turned out just fine as they were (Collabora sponsored it generally), but for future reference at least it would be good to get this more organised

  90. nyco

    I think and believe we (XSF) should let go some stuff I mean software we are a bunch of low level developers, doing low level stuff, like protocols we are not bad at it (could be better, always) I think and believe we need to catalyse higher steps in the software stacks like bring in developers who are not protocol-minded or protocol-oriented we need software developers who are into better UX and UI we need to test and improve things end-to-end it may be time to revive "Modern XMPP" except there should be no "XMPP" in the title because once again that focusses on the protocol let's focus on chat and chat alone, start from here, improve things, one step at a time

  91. Guus

    Agreed - it's on one of my many to-do lists (with SCAM hat on)

  92. Guus

    what do you mean with "letting go software" ?

  93. MattJ

    nyco, "Modern XMPP" (if you refer to my talk some time ago) is not dead, nor is it an XSF activity

  94. nyco

    I know

  95. edhelas

    XMPP 2.0

  96. MattJ

    and since it targets developers, neither should it be rebranded (though a user facing brand is a separate issue that the XSF may or may not want to tackle)

  97. ralphm

    nyco: yes, I think the XSF remains primarily a standards organisation, and there's always been debate if we can/should be at the center of development

  98. nyco

    not "XMPP 2.0", because it focusses on the protocol

  99. ralphm

    We made a more or less explicit choice here when we went from Jabber Software Foundation to XMPP Standards Foundation.

  100. nyco

    XSJ used to be JSF

  101. nyco

    oh the glorious days

  102. MattJ

    There is a lot of uncertainty about the Jabber trademark... some people use it, some refuse to

  103. Guus

    ralphm: I don't understand, or see that debate. We're not developing software as the XSF - apart from some tooling for our own benefit?

  104. nyco

    we can start an informal movement, see where it goes

  105. Ge0rG

    ralphm: was it an explicit decision back then?

  106. ralphm

    Guus: well, for example, should we promote certain implementations?Tthat's what has come up numerous times.

  107. Ge0rG

    nyco: do you have the man power for that?

  108. nyco

    not the XSF

  109. MattJ

    ralphm, the survey fairly clearly indicated that people value the XSF's neutrality

  110. Guus

    Ralphm: but we do not, and agreed recently to will not change that.

  111. nyco

    Ge0rG that's what I am working on

  112. ralphm

    Ge0rG: from my recollection, yes

  113. ralphm

    Guus: indeed, and I think that supports what nyco said in his wall-of-text

  114. Ge0rG

    I've called out for a new Jabber Software Foundation to care about UX and branding and supporting developers for some years now.

  115. nyco

    Ge0rG I know, let's just do it

  116. ralphm

    Just like our protocol is distributed, there's no reason for a software organisation to exist next to the XSF

  117. Guus

    I'm not against people doing things like that - but that'll be out of scope of the XSF, won't it?

  118. MattJ

    Ge0rG, calling out helps a lot :)

  119. Guus

    (so by all means, go for it, but lets not make it a topic here?)

  120. MattJ

    ralphm, assume you meant "not to exist"

  121. ralphm

    *not* to exist

  122. nyco

    there's no reason for a software organisation NOT to exist next to the XSF

  123. nyco

    Guus indeed, I think it is out of the scope of the XSF, not the XSF to decide

  124. ralphm

    (well, we got that clear, phew)

  125. MattJ

    I think everyone agrees with this, and pep. and I have been looking into it a little

  126. ralphm

    I have some ideas, too, but not as part of the XSF indeed.

  127. Guus

    if we all agree that this is not a matter for the XSF and thus its board - let's move on? 🙂

  128. MattJ

    Yep :)

  129. nyco

    yes

  130. ralphm

    So, going back to the topic, we can conclude that we *do* want to continue supporting community events

  131. ralphm

    For now, I'm looking forward to FOSDEM again.

  132. pep.

    Sorry I'm at work, can't follow all this, I'll try to reply later if necessary

  133. Guus

    FOSDEM + summit

  134. ralphm

    As a FYI: the real-time peeps have requested a Realtime Devroom again, and I will do the same for the Lounge

  135. ralphm

    2. FOSDEM + Summit

  136. MattJ

    Thanks ralphm

  137. ralphm

    Guus: let's get kicking for this, hopefully preparing better and getting more people there

  138. Guus

    agreed

  139. ralphm

    Ideas welcome, of course

  140. Guus

    let's do that in SCAM context

  141. Guus

    (I want to ping Seve about some of his ideas he had last year)

  142. ralphm

    scam@muc.xmpp.org as always

  143. Guus

    agreed

  144. ralphm

    3. Elections

  145. Guus

    When is our tenure over?

  146. ralphm

    It is September again, so I guess we need to invoke Alex to start up the process

  147. ralphm

    Early November I think

  148. Guus

    ok

  149. ralphm

    4. AOB

  150. ralphm

    Anything else?

  151. nyco

    POSS ?

  152. nyco

    Paris Open Source Summit

  153. nyco

    or next meeting?

  154. Guus

    nyco, take that to SCAM?

  155. ralphm

    SCAM first indeed

  156. Guus

    no need for the board to decide on that, I think?

  157. ralphm

    then Board if needed

  158. nyco

    oh yes, of course, sorry

  159. Guus

    no AOB from me

  160. ralphm

    5. Date of Next

  161. ralphm

    +1W

  162. ralphm

    6. Close

  163. nyco

    +1

  164. ralphm

    Thanks all!

  165. nyco

    thx!

  166. Guus

    wfm

  167. ralphm bangs gavel

  168. MattJ

    Thanks ralphm

  169. ralphm

    Good to be back :-D

  170. ralphm set the topic to

    XSF Discussion | Logs: http://logs.xmpp.org/xsf/ | Agenda https://trello.com/b/Dn6IQOu0/board-meetings

  171. MattJ

    Good to have you back :)

  172. Guus

    plz apply same magic to Martin

  173. ralphm

    Oh, who can make minutes?

  174. MattJ

    I'll add it to my backlog

  175. MattJ

    Once I get going it's fine, I'll probably push them all out in a batch once I'm done with what I'm currently doing

  176. ralphm

    yay

  177. nyco

    thx MattJ

  178. jonas’

    MattJ, the last minutes sent to members@ only had "Draft a membership survey on priorities." :(

  179. jonas’

    so... yeah

  180. jonas’

    please re-send the survey results to members@ when you get around to it

  181. jonas’

    ralphm, good points, I like it, +1 (on XEP-0369, disco#info things)