https://logs.xmpp.org/council/2022-02-16?p=h#2022-02-16-be3c28ae5ed36ed4

1) Roll Call
Present: Travis, Georg, Daniel, Jonas, Larma

2) Agenda Bashing
Jonas checks whether Larma got any reply regarding taking ownership of XEP-0272 
(Multiparty Jingle (Muji)) - Larma hasn't yet posted to the Standards List, as 
was agreed last week.

3) Editor's Update
* Updated XEP-0004 (Data Forms)
* Updated XEP-0060 (Publish-Subscribe)
* Proposed XMPP Extension: MUC Affiliations Versioning

4a) PR #1158 (Obsolete XEP-0156 and add warnings) - 
https://github.com/xsf/xeps/pull/1158
Daniel checks that this will essentially make BOSH obsolete as it would no 
longer be discoverable - the author, Travis, notes that it does remove BOSH, 
but he also has an alternative proposal to remove everything else, while adding 
BOSH.

Travis: +0 (this removes bosh, so my alternate proposal is remove everything 
else but adding bosh)
Jonas: -1 (we should keep BOSH discoverable)
Georg: -1 (because of BOSH)
Daniel: -1 (i'm on board with getting rid of dns and getting rid of http, but i 
do think we need to keep bosh)
Larma: -1 (don't think we need to keep BOSH forever, but probably still need it 
today)

Travis intends to rework the PR for next week.

Daniel thinks that it's interesting W3C EventSource is still around, even 
though WebSocket exists, but that's probably a discussion for another day.

4b) PR #1159 (Obsolete and update Security Considerations for XEP-0138 and 
XEP-0229) - https://github.com/xsf/xeps/pull/1159
Jonas would like to point out that writing a MUST NOT in an Obsolete document 
seems pointless.
Georg agrees with obsoleting, but says that "this method is deemed insecure and 
MUST NOT be used" is a normative change and policy must not be enforced with 
protocol.
Sam understands the reasoning, but has also deployed it on large projects and 
found it to be extremely beneficial - Travis notes that lots of insecure things 
are useful.
Jonas would be happier if, instead of changing the normative text, a huge 
security notice were added at the top of the document and in the place where 
the normative text would've been changed, otherwise the "MUST NOT" part seems 
obsolete - the author, Travis, explains that the "MUST NOT" is targeted at the 
method specifically, as he expects a new compression method may be used with 
the negotiation being resurrected - Jonas still thinks it seems off.
Georg would be okay with obsoleting and adding a big red warning in the 
Security Considerations - Jonas agrees - Travis is happy to change it and will 
prepare an update for next week.

Travis: +1 (as author)
Larma: +1
Georg: -1 (because of the MUST NOT)
Daniel: -1 (either just obsolete it or just add a security warning)
Jonas: -1 (what Georg says)

Sam has a brainwave: maybe there could be a non-normative "Editorial Notes" 
section at the top of XEPs which doesn't require a version update since it's 
not actually part of the XEP - Jonas thinks that such things should definitely 
be versioned - Sam says it would still be versioned in Git, and could be 
updated by Editor or Council.

4c) PR #1163 (XEP-0045: Remove some more mentions of GC 1.0) - 
https://github.com/xsf/xeps/pull/1163
Georg doesn't think that "[citation needed]" is appropriate in a XEP - Editor 
will leave a note.

Daniel: [on-list] (probably fine, just want to double check later)
Georg: [on-list]
Larma: [on-list] (same as Daniel)
Jonas: [on-list]
Travis: [on-list] (would be be +1 if not for [citation needed])

4d) PR #727 (Obsolete some deferred XEPs) - https://github.com/xsf/xeps/pull/727
This PR moves three currently Deferred XEPs to Obsolete; Jonas suggests voting 
for each XEP separately - Daniel agrees.

4d i) Obsolete XEP-0008 (IQ-Based Avatars) - 
https://xmpp.org/extensions/xep-0008.html
Jonas: +1
Georg: +1
Daniel: [on-list]
Larma: +1
Travis: +1

4d ii) Obsolete XEP-0038 (Icon Styles) - 
https://xmpp.org/extensions/xep-0038.html
Larma: +1
Jonas: [on-list]
Travis: +1
Georg: +1
Daniel: [on-list]

Georg would still like a XEP for mapping the ASCII smiley to Unicode - Jonas 
thinks that's better suited to ModernXMPP - Pep suggests using 'jabber:x:data'.

4d iii) Obsolete XEP-0051 (Connection Transfer) - 
https://xmpp.org/extensions/xep-0051.html
Larma: +1
Travis: +1 (with prejudice because it needs major security considerations)
Jonas: +1 (think this is best addressed with <see-other-host/> stream error in 
RFC 6120, which also talks about the corresponding security considerations)
Daniel: [on-list]
Georg: +1 (given <see-other-host>)

Georg notes that it's been Deferred for over a decade, and asks if anybody is 
using it - Travis hopes not, or at least that they're using it in a secure way.

4e) Proposed XMPP Extension: MUC Affiliations Versioning - 
https://xmpp.org/extensions/inbox/muc-affiliations-versioning.html
Daniel asks about using attribute namespaces - remembers there was a huge 
debate, but not the outcome - Jonas says the outcome was that some people are 
using XML libraries that can't handle them; Sam has seen multiple 
implementations that will break, even though their XML parser technically has 
namespace support. Pep notes that they're used in XEP-0103 (URL Address 
Information) and referenced in XEP-0449 (Stickers).

Jonas: [on-list]
Larma: +1 (as author)
Travis: [on-list] (gut reaction is to run from attribute namespaces)
Georg: [on-list]
Daniel: +1

Pep (also author) doesn't think use of attribute namespaces should be an 
barrier for Experimental - Jonas agrees.

Daniel starts thinking bigger: versioning not just affiliations, but presence 
and roles as well - Pep and Larma point to XEP-0311 (MUC Fast Reconnect) and 
XEP-0436 (MUC presence versioning) for doing exactly that.
Georg thinks it would be awesome to have a differential membership update 
mechanism for huge MUCs.

5) Pending Votes
Georg votes on PubSub Type Filtering: +1

6) Date of Next
2022-02-23 1600 UTC

7) AOB
Jonas had an idea, but Daniel notes the time and checks if everyone is fine 
with extending the meeting - nobody starts rioting.

Jonas notes that the past two years have brought increased usage of A/V 
technology and was wondering whether Council would like to migrate to 
conducting meetings in audio and possibly video; invites people to think about 
it to discuss next week - Georg is against it, not just for auditability 
reasons; Daniel was considering proposing something similar, maybe for the 
first meeting of each month. Larma isn't entirely against it, but doesn't like 
the idea of using non-standardised XMPP for this. Travis would prefer text, but 
is fine with audio if everyone else is.
Sam notes that minutes would need to be written live, unless the meeting is 
recorded - Jonas volunteers to write proper online minutes for auditability 
(already has been doing that for various work meetings.)
Daniel thanks Jonas for the suggestion, but with Georg seemingly against it, 
maybe it's something to reconsider at a later date. Jonas would like a proper 
discussion next week when there is more time.

8) Close
Thank you everyone. Thanks Daniel! Thanks!


Kev thinks "Editorial Notes" is a fine idea, but also thinks there's no reason 
not to version them (version numbers are cheap, and the last part already 
represents editorial changes) - Sam was aiming to avoid a bump on a 
ten-year-old Final XEP just for fixing a typo.

Pep would be against A/V meetings, but maybe there can be trials to see how it 
goes; Kev thinks it's useful for Council if that's their only activity (e.g. 
not on a train), while text is more useful for people following along or 
reading later - maybe a poll would be useful to see how many people actually 
read the logs.

_______________________________________________
Standards mailing list
Info: https://mail.jabber.org/mailman/listinfo/standards
Unsubscribe: standards-unsubscr...@xmpp.org
_______________________________________________

Reply via email to