XSF Discussion - 2020-11-04


  1. Tim

    Is the XMPP logo free to use in websites next to xmpp: links, the way, say, the GitHub and Twitter logos are sometimes used next to links to GitHub and Twitter accounts?

  2. MattJ

    Tim, go for it :)

  3. MattJ

    I just switched from subscribing to planet.jabber.org's Atom feed to the RSS feed after I noticed my feed reader was missing a number of items

  4. MattJ

    They seem to be in the feed ok, but then the W3C validator throws up a handful of issues (likely issues with the source posts)

  5. Ge0rG

    MattJ: write some Javascript to fix the issues and inject it from your own blog?

  6. MattJ

    Good idea

  7. Tim

    MattJ: Thanks! It's in the footer at https://circulex.nz/

  8. Guus

    Is there any client left that depends on stuff like the crossdomain.xml file, or using the http://www.jabber.com/streams/flash namespace? From what I can tell, this is old Flash-related stuff. I'm considering to drop support.

  9. jonas’

    Flash is very EOL at the end of 2020. Go for it.

  10. Guus

    Yeah, I'm all for killing Flash with fire. I'm primarily worried if something else is piggybacking on those mechanisms though.

  11. jonas’

    nothing matches streams/flash in xeps

  12. jonas’

    nothing matches crossdomain

  13. jonas’

    kill it with fire

  14. jonas’

    if anything piggybacks on it, it is wildly nonstandard and piggybacking on a technology from the previous decade (no matter which way you use to count decades? :)

  15. jonas’

    if anything piggybacks on it, it is wildly nonstandard and piggybacking on a technology from the previous decade (no matter which way you use to count decades) :)

  16. jonas’

    if anything piggybacks on it, it is wildly nonstandard and piggybacking on a nonstandard technology from the previous decade (no matter which way you use to count decades) :)

  17. Guus

    Thanks 🙂

  18. Guus

    Yuk

  19. Guus

    // We need to allow the NULL character, however, for Flash XMLSocket clients to work.

  20. Ge0rG

    No matter how I count, it's technology from two decades ago ;)

  21. Guus

    Doesn't mean that people aren't still using it though...

  22. Zash

    Grep your forum for hints maybe?

  23. Ge0rG

    it's one of the few projects that have an even worse security track record than libpurple.

  24. Guus

    There's actually a couple of hits on 'flash' in our forum. Half of which related to people asking if that's a security issue, the other half questioning why their Flash client isn't working with Openfire. :-S

  25. Guus

    (and a gazillion of hits on notifications that are 'flashing' somewhere)

  26. moparisthebest

    > Yeah, I'm all for killing Flash with fire.

  27. moparisthebest

    killing it with... Openfire ?

  28. moparisthebest walks himself out

  29. Guus

    I'm now annoyed with myself that I didn't think of that.

  30. emus

    moparisthebest: 😁

  31. marc

    Ge0rG: why does 389 rely on unauthenticated IQs?

  32. Ge0rG

    marc: https://xmpp.org/extensions/xep-0389.html#sect-idm46110484684112

  33. marc

    I don't understand this part but it is an option not necessary, no?

  34. marc

    I mean i don't understand it's use case

  35. Ge0rG

    marc: I haven't read it either, just saw that IQs are used after stream negotiation, which I read as before auth

  36. Ge0rG

    or maybe before binding? Dunno

  37. marc

    I don't know but it is not necessary IMO

  38. Ge0rG

    but the question is: is it supported?

  39. marc

    No, you say it is necessary for 389 which is probably not the case

  40. marc

    Ge0rG: talked to sam, it is after auth

  41. Ge0rG

    marc: well, I wrote that it "relies" on them, but in that case I was wrong.

  42. marc

    👍

  43. marc

    It's for the case an admin wants to create an account on the server, for example

  44. Ge0rG

    I'm surprised that Sam didn't point that out

  45. marc

    He missed that point

  46. marc

    Ge0rG: if I understand it correctly the entire on boarding process can be based on 389

  47. Ge0rG

    marc: that's great. Implement it, write it down, submit a XEP

  48. Ge0rG

    We can add it to the list in the updated 0401

  49. Ge0rG

    I'm sure that it will be added to smack 4.5 one day, which I'm going to integrate into yaxim around 2023

  50. Ge0rG

    Ah, Sam has clarified the intended use case as well

  51. marc

    Ge0rG: pre-auth token is not implemented in smack either ;)

  52. Ge0rG

    marc: no, but iq is, and adding a custom element is rather simple

  53. Ge0rG

    marc: https://github.com/yaxim-org/yaxim/blob/master/src/org/yaxim/androidclient/packet/PreAuth.java

  54. marc

    Ge0rG: I have no comparison but smack is quite extensible isn't it?

  55. Ge0rG

    https://github.com/yaxim-org/yaxim/blob/master/src/org/yaxim/androidclient/service/SmackableImp.java#L855-L857 is the actual implementation

  56. Ge0rG

    marc: yes it is, but implementing one new IQ is an order of magnitude or two easier than implementing a new IBR protocol