jdev - 2022-12-12

  1. nicoco

    edhelas and Zash: "which message ID to use?" → https://lite.framacalc.org/9igspeka0w-9y47 -- I may have got something wrong, especially message retraction which only mention "origin-id", but since origin-id is not mandatory for clients, maybe it's reasonable to use the message id for 1:1 chats instead?

  2. Zash

    nicoco, would you kindly forget all about origin-id

  3. nicoco

    Zash: what do you mean? I'm just reporting what is in the XEP

  4. Zash

    oh no

  5. Zash

    nicoco, origin-id and message-id should be considered the same thing, and the later preferred except in very very rare circumstances where a MUC might be rewriting IDs. But then in MUC using an archive-id is better

  6. nicoco

    archive-id‽ stanza-id you mean? I figured so about origin-id, but just reported what I read in the message retraction XEP.

  7. Zash

    "lowed child has many names" ... ✎

  8. Zash

    "loved child has many names" ... ✏

  9. Zash

    stanza-id when MAM is available is the archive-id 🙂

  10. nicoco

    hahaha. even if I'm not affected by origin-id, I'm glad you clarified this. did I understand this right? → no origin-id should be sent by clients in 1:1, and clients should only use origin-id in if the muc does not have the "jabber.org/protocol/muc#stable_id" feature.

  11. Zash

    MAM had its own id in an earlier version, but it got broken out

  12. nicoco

    examples are not normative, but I guess this example should be changed to remove the origin-id mention? https://xmpp.org/extensions/xep-0424.html#example-3 It sure looks like origin-id is mandatory; also origin-id and message-id differ in this example, which is not how that's supposed to work, right? ✎

  13. nicoco

    examples are not normative, but I guess this example should be changed to remove the origin-id mention? https://xmpp.org/extensions/xep-0424.html#example-3 It looks like origin-id is mandatory; also origin-id and message-id differ in this example, which is not how that's supposed to work, right? ✏

  14. nicoco

    mmmm but https://xmpp.org/extensions/xep-0422.html is explicitely saying that origin-is mandatory for referring to messages. it's a mess, really ✎

  15. nicoco

    mmmm but https://xmpp.org/extensions/xep-0422.html is explicitely saying that origin-id is mandatory for referring to messages. it's a mess, really ✏

  16. pep.

    It's a mess because people disagree on using origin-id :p

  17. Kev

    Origin-id just needs to be balefired and everything will be ok :)

  18. Ge0rG

    I vaguely remember a Council member refusing to stabilize 0359 on those grounds. Apparently it's in "abandoned" state since.

  19. Zash

    Proper use of balefire 👍

  20. nicoco

    my gateway-developer input (which is not the most important input, I know): my code would be cleaner if we used "1:1→message id and MUC→stanza-id". ✎

  21. nicoco

    my gateway-developer input (which is not the most important input, I know): my code would look cleaner if we used "1:1→message id and MUC→stanza-id". ✏

  22. pep.

    It's not just about code cleanness though, or how easy it is to write, that's how we end up with http upload, wait.. ✎

  23. pep.

    It's not just about code cleanness though, or how easy it is to write, that's how we end up with http upload, or styling, wait.. ✏