XSF Discussion - 2019-03-27


  1. Ge0rG

    Guus, Seve: https://opensourcedesign.net/jobs/jobs/2019-03-19-design-of-badges-for-different-xmpp-compliance-levels

  2. Seve

    Ge0rG, goooood! I'll add it to the newsletter so we can help spread the word

  3. Ge0rG

    I've also created a strawman set of badges at https://op-co.de/tmp/xmpp-compliance-badges/

  4. Ge0rG

    jonas’: do you want me to bump the version number for s/412/410/ in 0412?

  5. jonas’

    what?

  6. jonas’

    -ENOCTX

  7. Ge0rG

    jonas’: my last mail to standards regarding the typo in the 0412 changelog

  8. Kev

    All XEP changes /should/ have a version bump, yes.

  9. Kev

    Editorial level, obviously.

  10. jonas’

    changes in the revision log are weird though

  11. Guus

    Good job Ge0rG

  12. Guus

    Although https://op-co.de/tmp/xmpp-compliance-badges/ gives me nothing but broken images

  13. Guus

    https://xmpp.igniterealtime.org:7483/httpfileupload/e788d7b8-1101-4142-8fb4-b6ca55e46647/hxxMFi9-Sjy3ylTXZZsCZA.jpg

  14. Seve

    Same here

  15. Ge0rG

    I'm sorry, it looks like my rsync job went wrong. Everything becomes very complicated if you are capped to 64kbit/s.

  16. Guus

    64K should be enough for everyone, Ge0rG ...

  17. Ge0rG

    I've regenerated the index, does https://op-co.de/tmp/xmpp-compliance-badges/ work and show six badges per line now?

  18. Seve

    Yes Ge0rG

  19. Seve

    Except one, Ge0rG

  20. Seve

    Advanced IM 2016

  21. Ge0rG

    jonas’: do I need to make a new changelog block as well then?

  22. Guus

    Looks great, Ge0rG !

  23. Guus

    My suggestion is to fade to grey, instead of red, over time.

  24. Ge0rG

    Seve: thanks, fixing it now

  25. Guus

    red implies 'broken'

  26. jonas’

    Ge0rG, ...... maybe wait for the inevitable LC feedback

  27. MattJ

    Hmm yeah, if you can keep the contrast right, I'd vote for grey as well

  28. Ge0rG

    fixed advanced-im-2016.svg

  29. pep.

    Guss, yeah I'd also prefer that. I don't mind another color, just not red

  30. Guus

    pep. exactly. Not red. Could be something else than grey.

  31. jonas’

    I wonder if it should be `[Logo] Client | Core/Advanced/... 2019` instead of `[Logo] Core Client | 2019`

  32. Guus

    (although 'fade to black' does imply outdated stuff)

  33. Ge0rG

    feel free to play around on your own. Generator URL is https://img.shields.io/endpoint.svg?url=https://op-co.de/tmp/cs2016.json&label=Advanced%20IM - clone the JSON <https://op-co.de/tmp/cs2016.json> to your own webspace, change the content

  34. jonas’

    because the part of the suite which is implemented is kind of the "value" of the badge, not necessarily the key?

  35. Ge0rG

    If you want to regenerate all badges, https://op-co.de/tmp/tags.sh and https://op-co.de/tmp/tags.txt

  36. Ge0rG

    we also don't have categories for combinations in the CS

  37. Ge0rG

    like Mobile IM

  38. Ge0rG

    otherwise I'd love to link to the specific sections of 0412

  39. Ge0rG

    it would be possible to mark them "XMPP Client: Advanced IM" or whatever :D

  40. Ge0rG

    shades of grey would conflict with the grey background of the left pane

  41. jonas’

    I think there’s precedent for using grey on the rhs of the badge

  42. MattJ

    Light grey wouldn't?

  43. jonas’

    a light gray I think

  44. Ge0rG

    jonas’: there was an LC and it did generate feedback. I really don't want to have another round of "two weeks" in here re 0412

  45. MattJ

    shields.io has a 'lightgrey' option

  46. MattJ

    aliased to 'inactive'

  47. jonas’

    Ge0rG, ah, I see

  48. jonas’

    I can just fix the typo when I assign 1.0

  49. Ge0rG

    I think that the size of the badge and the lack of the string "XMPP" is a larger issue than the color of the rhs

  50. Ge0rG

    jonas’: that sounds like the most elegant solution. Feel free to use this commit: https://github.com/ge0rg/xeps/commit/ec31053591a7fdf98b54680450dbe1fba026d5da

  51. Ge0rG

    before we bikeshed the color of the year, can we please bikeshed the color of the left pane? I want to have a cheap color for "Core" and a premium color for "Advanced" please.

  52. Ge0rG

    or maybe advaced should have golden text?

  53. Seve

    Ge0rG, all the categories, are just as a mockup thing, to see which ones get selected or your view on this is to use all those categories?

  54. Ge0rG

    Seve: I've created all categories that are sensible in my eyes

  55. Ge0rG

    e.g. "Mobile Web" doesn't make much sense

  56. theTedd

    Compliance Suites Badges for All!

  57. theTedd

    https://bitbucket.org/mrtedd/compliance-badges/src

  58. theTedd

    coincidentally similar to Georg's, but different

  59. Ge0rG

    I think IM+ is the brand name of some IM client/platform</bikeshed>

  60. Ge0rG

    (the website is still loading)

  61. theTedd

    Ge0rG, yes, I think so too -- it's just a shorthand instead of adding "advanced" everywhere; other suggetions welcome

  62. Ge0rG

    Golden text color for Advanced?

  63. Ge0rG

    theTedd: is your proposal as strawman as mine?

  64. theTedd

    I'm open to design suggestions - I was going to fiddle with the colours later; the main part was generating them automatically

  65. Ge0rG

    theTedd: I've made a very unprofessional attempt at splitting the text into two lines at https://op-co.de/tmp/core-client-2019.svg

  66. Ge0rG

    Didn't look into the actual SVG generator for how flexible it is, this is manually "designed" in Inkscape

  67. theTedd

    I saw, but it's a little squashed; I think badges should be short and simple

  68. Ge0rG

    I very much like the "+" notion for shortening the text, except for the Trademark problem

  69. theTedd

    * is an easy alternative

  70. Ge0rG

    * has a very different meaning (where's the footnote?)

  71. Ge0rG

    I don't see anything else with a positive connotation...

  72. MattJ

    ++

  73. Ge0rG

    https://upload.yax.im/upload/vLVYbPWkMwCOs5bL/cat.png

  74. theTedd

    I could add emoji characters :D

  75. Ge0rG

    theTedd: are you reading my mind?

  76. theTedd

    sorry, bad habit

  77. Ge0rG

    `XMPP:2019 | Client🙈 Web🤢`

  78. theTedd

    Client⁺

  79. theTedd

    (superscript +)

  80. Ge0rG

    Client➕

  81. wurstsalat

    > `XMPP:2019 | Client🙈 Web🤢` :D

  82. Ge0rG

    Almost looks like a grave tomb. Fits the general status of xmpp

  83. Guus

    At some point, I'm driving over there to take your keyboard away from you.

  84. theTedd

    ♕ Client

  85. moparisthebest

    https://gizmodo.com/why-you-should-stop-using-telegram-right-now-1782557415 "you should ditch silo X for silo Y" no mention of XMPP :'(

  86. Ge0rG

    > The app also leaks metadata “like a champion,” I'm sure XMPP can beat it on _that_ front.

  87. moparisthebest

    what's signals counter, "we don't leak metadata, we just have it all ourselves" ?

  88. Ge0rG

    moparisthebest: they are working really hard to avoid as much metadata as possible. at least they claim to do so

  89. moparisthebest

    yea that's the problem "they claim" not verifiable at all

  90. moparisthebest

    that article kind of smells like a facebook-funded hit piece hehe

  91. moparisthebest

    "switch to a messenger approved by your facebook overlords"

  92. Ge0rG

    I'm sure facebook doesn't even need to pay for this kind of marketing.

  93. Wiktor

    w.r.t. Signal and privacy https://news.ycombinator.com/item?id=18234849

  94. Zash

    Didn't they use hardvare stuff to prove what server code they were running?

  95. pep.

    You'd still need to make sure that you're running the correct client binary

  96. jonas’

    luckily, reproducible builds are a solved problem /s

  97. MattJ

    Zash, did they? Reference for that?

  98. MattJ

    I can't imagine how you could possibly prove it

  99. Zash

    I read it on the Internet.

  100. MattJ

    :)

  101. jonas’

    something about SGX

  102. jonas’

    and remote attestation

  103. pep.

    https://signal.org/blog/private-contact-discovery/ ?

  104. Zash

    Indeed mentioned there

  105. Zash

    But is that only used for the contact discovery?

  106. pep.

    I haven't read the code, but that's what the HN thread above seems to show

  107. moparisthebest

    oh well as long as they verified it's secure with a black-box intel extension that also can't be verified secure I guess it's secure?

  108. moparisthebest

    o_O

  109. MattJ

    Yup

  110. Wiktor

    > But is that only used for the contact discovery? As far as I know it's not actually used. Just a proof of concept.

  111. jonas’

    hey, it’s too early for another markup flamewar

  112. Seve

    Noo, never enough!

  113. Zash

    It was enough when the first person typed *something* on a type writer

  114. MattJ

    Did the typewriter render it as bold or italic?

  115. moparisthebest

    bold if you hit the keys hard enough, assuming mechanical

  116. Ge0rG

    did you know that you can press backspace on the typewriter once for each letter and then re-type for bold markup?

  117. moparisthebest

    I'm awaiting your XEP for that Ge0rG

  118. MattJ

    I've been doing that on XMPP for years

  119. Ge0rG

    XEP^H^H^HXEP! Look how bold I am!

  120. dwd

    Ge0rG, Usually X^HXE^HEP^HP, but yeah. Doesn't [gn]roff still do that?

  121. Ge0rG

    dwd: I'm sure you are going to get more boldness with my approach.

  122. Ge0rG

    slightly related, I was just editing the badge SVG generated by mrtedd in the browser's inspector.