XSF Discussion - 2018-10-31


  1. Kev

    Anyone want to find the bit of 6121 that says you receive your other resources' presence when you come online for me? AFAICS after a few minutes hunting it doesn't actually say that other than mentioning an implicit subscription

  2. MattJ

    I haven't looked, but isn't that enough?

  3. fippo

    kev: 4.2.2?

  4. Kev

    fippo: Says about sending to your resources when you come online, not about receiving from them, I think.

  5. Kev

    MattJ: Only if you apply a lot of thinking, I think. The phrasing is """ The user's server MUST also broadcast initial presence from the user's newly available resource to all of the user's available resources, including the resource that generated the presence notification in the first place (i.e., an entity is implicitly subscribed to its own presence). """

  6. Kev

    Unless you already knew that you had to do it, I don't think a reasonable person would read that and realise, would they?

  7. MattJ

    Depends - are you calling all current server devs unreasonable people? :)

  8. MattJ

    I've never heard this cast into doubt before

  9. fippo

    mattj: we just did what jabberd did

  10. Kev

    MattJ: No, I think all the current server devs already knew what needed to be done so it didn't matter than 6121 didn't say it.

  11. Kev

    Or didn't say it in a way people would pick up.

  12. Kev

    I do agree that the 'implicit sub' bit does mean to do this (well, actually to send probes to yourself, with an optional optimisation), I was just surprised that we didn't say it explicitly.

  13. fippo

    having an explicit "send presence from all available resources" there would be better indeed. now what is the best way to fix that...

  14. fippo

    4.3.1 is the other wiggly part of this.

  15. Kev

    That lets you wriggle out of the probes for the implicit sub, yeah.

  16. flow

    "now what is the best way to fix that..." report as errata, so that it won't be forgotten in a possibly follow-up rfc?

  17. peter

    Feel free to file an erratum on RFC 6121. It seems clear to me, but I'm probably too close to it even after all these years. :-) Also, you don't have to send probes to your own account, I'd say.

  18. peter

    (Thanks to Fippo for the ping on IRC!)

  19. fippo

    i think "IRC" is a bad word on this channel :-p

  20. peter

    Hey we invented Jabber over IRC. ;-)

  21. Ge0rG

    peter: and it shows in things like MUC. So you can be pinged on IRC to get you summoned to XMPP?

  22. peter

    I'm on irc.mozilla.org all day long.

  23. fippo

    only during he days :-p

  24. Ge0rG

    Oh, I don't think I ever connected there. If only there was federation between IRC networks...

  25. peter

    But I've been logging in here more too - however I use two computers and don't always look at this one. :-)

  26. Zash

    put up a single-network-biboumi

  27. Ge0rG

    Or a multi network biboumi, for that matter. I'm doing 95% of my IRC needs from yaxim / poezio / XMPP

  28. Zash

    I'm doing all my IRC through biboumi now

  29. Ge0rG

    There's a network requiring special modes and another one with client SSL that certificates

  30. Zash

    Did you just use "certificates" as a verb?

  31. Ge0rG

    There's a network requiring special modes and another one with client SSL certificates

  32. Ge0rG

    Ah, LMC to the rescue, now also in yaxim!

  33. SamWhited

    But why did you send that message multiple times, I don't even see the small difference that my eyes glance right over because I've already read your message once and assume it's just a transport problem and it got sent twice like SMS does sometimes?

  34. Zash

    Because that's how Last Message Correction works?

  35. Zash

    Or doesn't work

  36. Ge0rG

    SamWhited: looks like your client fails at LMC

  37. Zash

    Or doesn't work ... (test)

  38. Ge0rG

    LMC in MUCs has some privacy implications

  39. Zash

    Ge0rG: The replacement message doesn't stay in the same position, bug, intentional, limitation?

  40. Ge0rG

    Zash: looks like the timestamp is updated to the correcting timestamp, reordering the messages. I'd say it's a minor bug

  41. Zash

    Ge0rG: More minor than not indicating that any correcting happened.

  42. Ge0rG

    Zash: I'm not sure what you mean.

  43. Zash

    Ge0rG: I don't see that the UI indicates that a message has been replaced.

  44. Ge0rG

    Zash: there is a wiring hand Emoji by the timestamp

  45. Zash

    Ge0rG: Did you add that after Sunday or is the device missing emoji fonts?

  46. Ge0rG

    Zash: ✍. If your phone can't render it, it's not my fault!

  47. Ge0rG

    Zash: I've done significant work for the 0.9.3 release this week. Upgrade now!

  48. Zash

    Test

  49. Zash

    Testing

  50. Zash

    Ge0rG: Ok, works now

  51. Ge0rG

    In the chat database, I'm aggregating LMC into the original message, so it's slightly quirky if something goes wrong, or if the correction lands before the original

  52. Syndace

    The wiki page for membership applications states "responsibilities of XSF members are listed here" but there is no list/link to a list.

  53. Ge0rG

    Syndace: https://xmpp.org/community/membership.html

  54. Syndace

    thanks