XMPP Council - 2022-06-22

  1. daniel

    30 minute warning

  2. moparisthebest

    I'll still be here

  3. jonas’

    I probably will

  4. Ge0rG

    good evening!

  5. jonas’


  6. moparisthebest


  7. moparisthebest

    jonas’, gesundheit

  8. daniel

    It’s time

  9. daniel

    1) Roll call

  10. Ge0rG ,o/

  11. daniel

    do we have larma?

  12. larma

    Not really

  13. daniel

    i guess not?

  14. daniel

    2) Agenda bashing

  15. daniel

    i didn’t send out an explicit agenda for today; but we are going to use last weeks agenda

  16. daniel

    3) Editors update

  17. daniel

    a) Proposed XMPP Extension: WebSocket S2S (https://xmpp.org/extensions/inbox/websocket-s2s.html)

  18. daniel

    b) Proposed XMPP Extension: XMPP over QUIC (https://xmpp.org/extensions/inbox/xmpp-over-quic.html)

  19. larma

    Had no time to read but love the initiative :)

  20. daniel

    quick reminder to the editor (no rush I know you are on holiday) that we wanted to start a last call

  21. daniel

    for 215

  22. daniel

    4) Items for voting

  23. daniel

    b) Proposed XMPP Extension: XMPP over QUIC (https://xmpp.org/extensions/inbox/xmpp-over-quic.html)

  24. jonas’

    daniel, thanks, I noted it in the spreadsheet of doom and will hopefully remember to look at it next week

  25. moparisthebest

    +1 (obviously)

  26. Ge0rG


  27. jonas’

    why do we start with 4b?

  28. daniel

    as a developer I understand the "just do everything rfc xyz is doing except" but I'm not sure this is correct form

  29. daniel

    copy paste error. sorry

  30. jonas’

    I like the idea, +1.

  31. daniel

    ignore the above feedback because that's for 4a

  32. daniel

    i’m +1 on xmpp over quic

  33. moparisthebest

    the risk of copy+pasting from the other RFC is that something is missed or wrong, I guess

  34. daniel

    i'm assuming larma is on list too

  35. daniel

    a) Proposed XMPP Extension: WebSocket S2S (https://xmpp.org/extensions/inbox/websocket-s2s.html)

  36. moparisthebest


  37. daniel

    now my feedback applies: as a developer I understand the "just do everything rfc xyz is doing except" but I'm not sure this is correct form

  38. jonas’

    I... I have ethical issues with this and I'll need to ponder on-list.

  39. jonas’

    A proper section on "motivation" may convince me, but I note that this is missing completely.

  40. moparisthebest

    I think I put that https://xmpp.org/extensions/inbox/websocket-s2s.html#usecases

  41. Ge0rG


  42. jonas’

    I don't scroll that far to find motivation :-).

  43. daniel

    does anyone else feel that this should rather go into a new version of rfc 7395?

  44. jonas’

    ah that too, yes

  45. daniel

    and/or be an informational xep?

  46. daniel

    plus it creates ietfs namespaces?

  47. moparisthebest

    eventually they should probably both be RFCs, I'm completly unfamiliar with that process though

  48. daniel

    which i think we are not allowed to do?

  49. jonas’

    we're most certainly not

  50. daniel

    moparisthebest, i understand that that's why I was offering "informational" xep as a cheap alternative

  51. moparisthebest

    it's that or live with a tmp namespace forever :)

  52. daniel

    with a different namespace

  53. jonas’

    moparisthebest, could also use urn:xmpp:...?

  54. jonas’

    (which we own)

  55. daniel

    I think in it's current form I have to be -1

  56. moparisthebest

    could also just use jabber:server ? :P

  57. moparisthebest

    re: "just do what RFC X says" I think there is prior art there, ie https://datatracker.ietf.org/doc/html/rfc7858 which says "just do TLS like RFC X and then just do DNS like RFC Y"

  58. moparisthebest

    daniel, just based on the namespace or what else?

  59. daniel

    a combination of the namespace thing (which would be an easy) fix. and it just not feeling right. I'm not familiar with the "prior art" you mentioned. I guess i should read into that

  60. daniel

    maybe "on list" until i checked that out

  61. moparisthebest

    re: namespace any objections to jabber:server ?

  62. daniel

    but fwiw I don’t think creating an RFC (with your name not a number. is there a term for experimental rfc?) is super hard

  63. moparisthebest

    yea I gotta look into that one day...

  64. daniel

    that just feels like a more proper way to handle that. or as I said make it informational to get around the hacky nature

  65. daniel

    but i also don’t want to be the only one blocking this. so let's see how the other votes turn out

  66. jonas’

    I'm happy to be a second blocker if you need one ;)

  67. moparisthebest

    just making it informational so we can ignore things seems like a hack

  68. jonas’

    yeah, I agree on that moparisthebest

  69. jonas’

    I don't think making it informational solves anything here

  70. daniel

    ok; but let's move on. i'll read up on the other rfcs moparisthebest mentioned and vote on list

  71. daniel

    5) Pending votes none

  72. daniel

    6) Date of next

  73. jonas’

    +1w wfm

  74. daniel

    +1w wfm

  75. moparisthebest

    +1w wfm

  76. daniel

    7) AOB

  77. jonas’

    none from me

  78. daniel

    8) Close thank you all

  79. jonas’

    thanks daniel

  80. moparisthebest

    thanks all !

  81. Kev

    Various things have migrated from XEP to RFC, or back again, in the past, FWIW.

  82. Kev

    So to my mind unless Council/XSF lacks the ability to review it sensibly (which is sometimes the case with security things), I don't think something starting as a XEP and migrating over would be the end of the world.

  83. moparisthebest

    thanks Kev that's how I'd prefer to do it anyway :)

  84. moparisthebest

    re: s2s websocket namespace, I do include the contact of who to register namespaces with on there, unsure if it works like ALPN where the editor just rings them up and they register it or not though

  85. Zash

    The starting point to look for who's responsible for some urn:* would be https://www.iana.org/assignments/urn-namespaces/urn-namespaces.xhtml

  86. moparisthebest

    So it's just expert review and look who the team leader is: > Peter Saint-Andre

  87. moparisthebest

    We can probably contact that guy!!!! :)

  88. moparisthebest

    Turns out jabber: isn't registered, we are all breaking the law!

  89. Zash

    Yeah, it's from before anyone knew about needing to register these things.

  90. Zash

    I feel like I've seen mention of it somewhere as a thing that was used without being registered, hysterical raisins etc.

  91. moparisthebest

    > To date, the XML > namespaces defined within the Jabber/XMPP community have used names > of the form "jabber:*" (deprecated since early 2002) https://www.rfc-editor.org/rfc/rfc4854.html

  92. moparisthebest