XMPP Summit - 2020-02-02


  1. moparisthebest

    pep., fair , also I think spellcheck might have mangled the message in parens or I don't get it

  2. Daniel

    > In germany you don't have the right on your picture anymore when you attend public events like concerts, summits and so on. I guess it's more or less the same in other European countries. The summit arguably isn't a public event

  3. Daniel

    It's also doesn't matter if it's legal or not. It's just not cool

  4. MattJ

    There is no barrier to entry, it's open to anyone who says they want to attend. Not sure if there is a specific definition of "public event", but it fits mine

  5. MattJ

    Anyway, I think two things: if you're at such an event, you can't assume that the default is that no photos will be taken, you should always check (if you care)

  6. MattJ

    Regardless, I think we should make the policy clear in future events, and I'd personally prefer that policy to be "no photos without notifying the room first"

  7. MattJ

    And wow I overslept

  8. Daniel

    Most people know everyone else. I know and are known by more people at summit than at the average party. While the theoretically barrier to entry is low it practice its just 'us'

  9. Daniel

    Most people know everyone else. I know and are known by more people at summit than at the average party. While the theoretically barrier to entry is low it practice it's just 'us'

  10. Daniel

    > Regardless, I think we should make the policy clear in future events, and I'd personally prefer that policy to be "no photos without notifying the room first" Yes making it explicit is the obvious solution to avoid any confusion

  11. pep.

    Personally I don't want to involve legalities in this, as Daniel says it's "not cool". Ultimately I want the burden of checking on those taking the pictures. I've been to quite a few events now where you'll get a different color lanyard if you don't want to appear in pictures (and they are careful to ask you when you arrive). Others like CCC it's just normal to ask.

  12. pep.

    and if you take a picture with lots of people in it you just blur/hide faces

  13. pep.

    (at least)

  14. pep.

    but in any case explicit is better than implicit.

  15. Daniel

    At previous summits people asked before taking a group picture too. So it's very likely that the person in question just forgot about it and there was no harm intended. I guess that's why the ietf puts up a note well in front of every single session as a friendly reminder.

  16. winfried

    My talk on XMPP will be in appr. 15 minutes in UD2.208 (room is tucked away quite far). There are still plenty of seats free.

  17. eevvoor

    I agree to make it explicit. I do definitely not want photos from me, and I cannot check all the time. So this schould be the default.

  18. eevvoor

    I agree to make it explicit. I do definitely not want photos from me, and I cannot check all the time. So this should be the default.

  19. eevvoor

    Can we avoid googledocs in future for the organisation, to talk about another privacy issue ... ?

  20. pep.

    yep I'd also like that. it's a bit less easy to settle than just saying "please ask before taking pictures" though. baby steps

  21. nyco

    Two hoodies sold: where do we report/acount that?

  22. jonas’

    before it’s too late, I already wish everyone safe travels when returning from Summit/FOSDEM.

  23. jonas’

    especially the now-ex EU folks.

  24. nyco

    thx a lot, safe travel to all as well, it's been awesome this year!

  25. vanitasvitae

    The matrix guys' cross signing solution is basically our master key stuff

  26. vanitasvitae

    :D

  27. vanitasvitae

    Exactly what we discussed :P

  28. nyco

    Dino is covered on Hacker News, congrats!

  29. vanitasvitae

    I had a very nice talk with Matthew.

  30. winfried

    vanitasvitae: any interesting results from that talk?

  31. vanitasvitae

    Mostly that we should embrace our similarities and should learn from another instead of having this "messenger war".

  32. vanitasvitae

    But we also agreed that this war is mostly fueled by a few. And most of those are non-developer members of the community.

  33. winfried

    Yeah, I got Matthew on board for a comparison of Matrix and XMPP highlighing the strenghs of each project in stead of breaking down....

  34. nyco

    Matthew sparks and maintains this war... At POSS, he wrote and said XMPP was "monolythic"

  35. nyco

    I had a nice chat with Manu, their CTO, who couldn'd come at POSS: very polite

  36. winfried

    The only thing I can find in XMPP that is monolithic and that isn't monolithic in Matrix is the management of a MUC/MIX, but it is an interesting question if the matrix model is better....

  37. zash

    What does that even mean?

  38. MattJ

    vanitasvitae, oh, it was you who leaked the FOSDEM invite link :D

  39. vanitasvitae

    Was that a mistake? :D

  40. MattJ

    I didn't want it to go viral and have to deal with that - I was trying to limit it to people who came in person to the stand

  41. MattJ

    I even had backup QR/NFC tags prepared in case I needed to revoke the invite token :)

  42. MattJ

    FWIW I don't mind, everything is fine :)

  43. MattJ

    I guessed it had happened when I saw e.g. debacle sign up, who wasn't even at FOSDEM

  44. winfried

    zash: that is the better question 😁

  45. MattJ

    FWIW I don't think there is a "war", I also don't think that "monolithic" is a bad thing, but I *also* don't see any way you could describe XMPP as monolithic

  46. MattJ

    I'd describe it as quite the opposite: quite fragmented

  47. MattJ

    Matrix is monolithic: they have one spec (well, one c2s, one s2s) and one implementation

  48. winfried

    Yeah, so many variables to compare on, so many ways to judge them...

  49. larma

    MattJ: but dendrite and all the non-riot clients

  50. winfried

    Still wondering what I got myself into...

  51. MattJ

    dendrite doesn't work, and pretty much none of the non-riot clients impelement everything (and also riot implements a bunch of stuff that isn't in the spec)

  52. MattJ

    winfried, it's a very difficult task indeed

  53. larma

    XMPP Clients also don't implement everything

  54. larma

    And don't they use dendrite for their p2p matrix thingee?

  55. MattJ

    We don't expect them to :)

  56. winfried

    MattJ: yeah, but good comparison would still help me....

  57. MattJ

    One of the Matrix selling points is that they only have one spec, and no mess of extensions

  58. larma

    Stop expecting it from matrix clients then

  59. MattJ

    I'm not expecting it, it's basically what they advertise as one of the primary benefits of Matrix over XMPP

  60. larma

    Do they still claim that? I think they mostly stopped doing that since around Matrix 1.0

  61. MattJ

    I personally think it's inevitable if you have an open ecosystem (as we and Matrix do)

  62. MattJ

    So now there is no difference between Matrix and XMPP? :)

  63. zash

    Instead of a mess of extension proposals, they have a mess of change proposals. Completely different!

  64. larma

    > We think of Matrix and XMPP as being quite different; at its core Matrix can be thought of as an eventually consistent global JSON db with an HTTP API and pubsub semantics - whilst XMPP can be thought of as a message passing protocol.

  65. larma

    Except that xmpp also has pubsub sementics in mix

  66. larma

    But we don't have a JSON based http api!

  67. larma

    Huge difference

  68. winfried

    Absolutely!

  69. MattJ

    zash would like to disagree

  70. pep.

    larma: yet? :p

  71. pep.

    and yeah, mod_rest

  72. zash

    mod_rest !

  73. MattJ

    Prosody has multiple, latest (and most comprehensive): https://modules.prosody.im/mod_rest

  74. larma

    That's non standard!

  75. zash

    Que confusion with ejabberds mod_rest

  76. MattJ

    and I think most other servers have one too - we just need a standard :)

  77. larma

    But yeah so basically no difference then

  78. MattJ

    (I personally think a standard would be very very sensible)

  79. pep.

    that was on the summit list

  80. MattJ

    It's one of the topics I don't think we reached

  81. MattJ

    We need a whole week!

  82. pep.

    I wasn't ready to accept http personally :x

  83. pep.

    and JSON.

  84. MattJ

    You don't have to, but it's convenient for some people/use-cases

  85. larma

    I personally would prefer if we used XML instead of json...

  86. larma

    For the rest api

  87. zash

    mod_rest does both

  88. larma

    zash: talking about the standard to be ;)

  89. larma

    The summit topic was unfortunately mentioning json, which might have scared some off

  90. ivucica

    XEP-0295 pls

  91. Ge0rG

    User defined data transfer please?

  92. MattJ

    Can't tell if joking

  93. Ge0rG

    http://catb.org/jargon/html/H/ha-ha-only-serious.html

  94. Ge0rG

    I should have suggested fixing the name into "developer defined data transfer"

  95. Ge0rG

    > Dichlorodiphenyltrichloroethane, commonly known as DDT, is a colorless, tasteless, and almost odorless crystalline chemical compound,[5] an organochlorine. Originally developed as an insecticide, it became infamous for its environmental impacts. Toxic. Poisonous. Also a Russian rock band.

  96. Intosi

    Hello home. Thanks for a very productive Summit and your help at FOSDEM!

  97. MattJ

    Glad you got home safe (and in good time) - agreed, this was a good few days, thanks to everyone who participated at the summit and at the stand

  98. vanitasvitae

    I agree. I had a great time at fosdem and the summit was especially nice this year. Thanks to everyone who participated in it and made it such a pleasant experience!

  99. pep.

    I haven't had a look at the minutes since summit. I'll try to dump that in the wiki tomorrow