XSF Discussion - 2022-03-04


  1. wurstsalat

    Link Mauve, is your DOAP -> JSON processor available somewhere on the interwebs? ;) I couldn't find it in your regular repos

  2. Zash

    Reject JSON, embrace CSV! `xml2 < mything.doap | 2csv implements @rdf:resource xmpp:status xmpp:version xmpp:since xmpp:until xmpp:note > mything.csv`

  3. wurstsalat

    TIL hugo can parse CSV as well. But we'd need some processing to get somewhere near this one: https://github.com/xsf/xmpp.org/blob/98281879854ce8a7b6466dbd4564cf7e442a13fb/data/clients_doap.json

  4. emus

    https://blog.documentfoundation.org/blog/2022/02/28/tender-to-cleanup-and-further-improve-odf-conformance-202202-01/ I would like to discuss what they do here with an xmpp perspective. Maybe we cannot tender and pay. But we could officially suggest ideas/developments and tender for donations. we have the fiscal hosting instance already. So why not use it the other way round? (suggest keys projects/developments to work and donate for) This could be really interesting als to guide things a bit. I know everyones like to break things in piece and hurdles before they are realised - so lets focus if this is a great idea ☺😉 (and of course it is a great idea 😛)

  5. emus

    *and it would offer to improve general things in XMPP single projects usually don't tackle

  6. Sam

    I don't really understand what you're suggesting; can you give a specific example of a thing the XSF might do?

  7. Sam

    It sounded like you were suggesting paying someone to fix things in certain projects, but then you say "Maybe we cannot tender and pay" (which I think is probably correct).

  8. emus

    Ok, let me try again. (First of all I assume we have no budget in xsf to pay development) Then i thought we could, as TDF does, suggest some core XMPP development on a meta level we were looking for years, but no one does it or it lags a lot. I think of for example xmpp.net (maybe a bad example) issues and development resources (no offence jonas', I really appreciate). I is an important project, but if there are few resource to fix things its not good of course. So maybe make this a key development we want to see being up & running for the community and xmpp in general. ensure such infrastructure is great or turn the website into hugo or so (wurstsalat was quicker of course and many thanks again). I dont wanna discuss such examples. it should just tell that there are important things to do for the network - but it would need some higher level attention or guidance. this is what we could lead a bit. I hope my point got more clear

  9. Sam

    Ah, I see, internal XSF stuff, not random XMPP related projects. Got it.

  10. emus

    well not xsf internal (only). thibgs that the network is missing on a higher scale too. maybe we think we need a new library in the ABC language or so - but no one does it then we could suggest that and requirements and call for funding. dont know if thats a good example, but I want to show my idea

  11. moparisthebest

    If no one has written a library in ABC language then it's clearly not needed

  12. Ellenor Malik

    or it is and you need to do it

  13. moparisthebest

    I still remain unconvinced XMPP libraries are useful in any language since every client writes their own anyway, and I guess that's the basic problem with this, everyone has different opinions on what is needed

  14. Sam

    emus: sorry, I see why you didn't want to give specific examples now; everybody focuses on those instead of what you were actually wanting to discuss :)

  15. Kev

    Most clients have their own, but many things that aren't primarily XMPP chat clients use libraries.

  16. Sam

    But yes, I see what you're saying now.

  17. emus

    🙈🙈🙈

  18. emus

    Another example: https://blog.documentfoundation.org/blog/2021/08/03/tender-to-implement-autoupdater-202108-01/

  19. emus

    they ask for an autoupdater, which I think is a key and useful feature

  20. moparisthebest

    That's what a package manager is for...

  21. Guus

    > Most clients have their own, but many things that aren't primarily XMPP chat clients use libraries. Most public IM XMPP clients, maybe.

  22. moparisthebest

    See we can't even agree on the basics :)

  23. Guus

    Non web, even

  24. emus

    so could we ask for things we clearly lag in general e.g. create doap file for these clients that are missing them: 1, 2, 3... rework XEP-123 (I know thid is contrary to the idea and editor, but maybe there is a fair way as many other get not paid - maybe we should pay all work!1!!)

  25. Sam

    I don't know if the XSF itself would want (or is allowed) to do this sort of thing, but I'd be open to helping start small donation pools that can be used to give grants to XMPP related software that wants to do some specific thing we establish a grant for

  26. Sam

    Even if the XSF doesn't want to, eg. pay for specific XEPs to get updated (which I can imagine they wouldn't), that doesn't stop us as a community from taking donations and offering someone a grant to do the work

  27. emus

    Yes, and the first step is not money, the first step would be to define tasls

  28. emus

    Yes, and the first step is not money, the first step would be to define tasks

  29. moparisthebest

    Yea seems to fit better in another org is my gut feeling

  30. moparisthebest

    Even if the other org is mostly the same people

  31. emus

    A very good example (even we dont have the need atm) Expand the XEP website by feature 1, 2, 3

  32. moparisthebest

    emus: isn't that what Snikket is doing actually? They paid tigase to implement things in siskin

  33. emus

    Yes, but they focus on their software and out of their funding (which all is fine)

  34. emus

    look at my previous example

  35. moparisthebest

    I feel like the XSF has the opposite problem, people do the work and it just never gets merged

  36. emus

    its a xsf but also xmpp important topic (without being a client or so)

  37. moparisthebest

    Where's the nice doap client list that's been done for months?

  38. emus

    moparisthebest: then we have our first idea maybe :D

  39. emus

    at least we have a problem

  40. moparisthebest

    Is the website using Hugo now? Did that get merged?

  41. emus

    moparisthebest: yes, more of those requests :)

  42. emus

    moparisthebest: yes it is. just an outdated example

  43. emus

    but wurstsalat would have deserved some funding - as everyone does!

  44. Sam

    I am still very much against doing the DOAP thing, and wouldn't vote for or donate to that one if we were to do this, but it seems I'm alone on that.

  45. emus

    Sam: that is fine

  46. emus

    just examples

  47. emus

    to get my point. we all could vote on such "tendering"

  48. Sam

    One I'd like to see is updated editor tooling; half the reason I stopped doing editor things is it was *extremely* time consuming and there were lots of little tasks that could be automated (and most of the existing tooling was broken or hard to setup).

  49. Sam

    Except the one time I tried to update it we came up with a plan, I started going things, and then was told "no, stop, if we're going to do these minor fixes we need to also do a million other things and rewrite the world from scratch" and then we bikeshedded forever until I left :)

  50. Sam

    And automated deploys for the website. I know the iteam says it's not a problem for them to manually update, but it's a problem for everyone else that we make changes and then have to figure out who to ping and that person might not be available

  51. Sam

    Ooh, more ideas: having the attic be automatically generated from Git history instead of being a thing we have to manually maintain (and is therefore never maintained)

  52. emus

    > Sam escribió: > And automated deploys for the website. I know the iteam says it's not a problem for them to manually update, but it's a problem for everyone else that we make changes and then have to figure out who to ping and that person might not be available 😅 its working great manually so far - but yes!

  53. emus

    Okay, maybe I can start to query for such ideas and maybe we get into some good direction with this (maybe there are even better approaches?)

  54. Sam

    I'm with moparisthebest that I suspect we'd want this to be a separate organization from the XSF (that the XSF could donate to itself if it wanted). This has some overlap with re-starting a JSF style organization that focuses more on software too.

  55. jonas’

    Sam, I wish we could do that for the attic, unfortunately we don't have a good enough revision history for that

  56. jonas’

    I started to tag things properly, but I wouldn't bet that that's consistent

  57. jonas’

    (and especially in the past—I've recently been more strict about that as you know—people added revision blocks not as the *last* commit of their update, so even going by that is not going to cut it)

  58. jonas’

    I think I started automation around that already and decided it was not worth the effort :/

  59. jonas’

    having the attic update based on *new* tags (keeping current content as-is) would be great tho

  60. Sam

    Yah, it would be hard. We'd also have to figure out a lot of edge cases I'm sure.

  61. Sam

    I was thinking just scanning all commits and not using tags

  62. jonas’

    scanning commits is going to yield incorrect results unless you come up with a much better heuristic than "revision block was added here"

  63. Sam

    Probably, but it might be "good enough"

  64. jonas’

    it's not

  65. jonas’

    it would be a step backward from the current attic for sure

  66. jonas’

    I'd go with "keep the current attic as static stuff, and starting at commit X, build the attic from tags" or so

  67. jonas’

    I'd go with "keep the current attic as static stuff, and starting at commit X, add more stuff from tags" or so

  68. jonas’

    tags have value outside the attic, so enforcing their use would be an advantage

  69. Sam

    Why wouldn't it be good enough/

  70. Sam

    (although yah, we could definitely do the "start from what we have now" way too, that seems fine)

  71. emus

    I dont think this is about software by hard. It is about strategic and developing xmpp as a network, infrastructure, idea and tool (in general). That can be software, but it is more that we maybe should consider. and I dont see why we as xsf cannot suggest developments?

  72. Sam

    Did anyone say they shouldn't suggest things?

  73. emus

    You were reglecting this JSF thing

  74. emus

    You were reflecting this JSF thing

  75. emus

    was before my time

  76. Sam

    I'm not saying we should do that, just that people have been talking about it for a long time and this idea seems to have some overlap (in that it would partially be about funding software)

  77. Sam

    This is more limited in scope maybe, but I think there's overlap with the idea.

  78. emus

    Hmm, but let me repeat: Its not only about software development. actually it could be anything. Redesign the XMPP logo (Im explicitly against it atm 😊) Make a fancy xmpp video!!!!! things we would like to see, but are not provided so far

  79. Sam

    I know, not saying it's exactly the same thing, just that there's overlap and it might be worth considering how they're related.

  80. msavoritias

    Maybe even something like a bullet board kind of thing? Were the ecosystem and invidual projects could post. The ecosystem areas or new things they thing are benneficial. And the individual projects where they need help.

  81. Sam

    That specific idea sounds like reddit; maybe we could run our own lemmy (federated forums/link sharing/reddit like thing) instance or something for that

  82. moparisthebest

    the strength of XMPP and the XSF is that it's so extensible everyone can have different or even conflicting ideas/goals/etc and it all mostly just works

  83. moparisthebest

    I don't consider this a downside really, but that also kinda means the XSF making cohesive decisions about anything else won't work so well

  84. msavoritias

    Sam: i had something like this in mind actually: https://ccs.getmonero.org/ Disreguard the topic of the page. More as platform example

  85. emus

    > msavoritias escribió: > Maybe even something like a bullet board kind of thing? Were the ecosystem and invidual projects could post. > The ecosystem areas or new things they thing are benneficial. > And the individual projects where they need help. could also be a thing!

  86. emus

    > moparisthebest escribió: > the strength of XMPP and the XSF is that it's so extensible everyone can have different or even conflicting ideas/goals/etc and it all mostly just works We can start with the simple, positive (for all) and low-hanging fruits