https://logs.xmpp.org/council/2021-04-07?p=h#2021-04-07-7313fa59410db8a5

1) Roll Call
Present: Daniel, Jonas, Zash, Dave
Late to the party: Georg

2) Agenda Bashing
None.

3) Editor's Update
* Last Call on XEP-0313 expired (last week already)
* Last Call on XEP-0280 expired yesterday (2021-04-06)

4a) Advance XEP-0313 (Message Archive Management) - 
https://xmpp.org/extensions/xep-0313.html
Jonas shares Georg's concerns about the Security Considerations section (AOB 
item related.)
Daniel has complicated feelings about this XEP (along with XEP-0280) - both 
feel like temporary fixes to fundamental issues in the protocol; Zash would say 
they feel like pretty large parts of a modern XMPP. Kev suggests they are both 
temporary, and would be surprised if either lasted more than a couple of 
decades. Daniel doesn't think any of this should stop them from advancing, it's 
more an explanation of why they will never be entirely satisfactory.
Zash thinks it might be a good idea to go through the previous Last Calls and 
check whether everything raised has been addressed (not that he is volunteering 
for the job.) Daniel recalls that many of the points raised were that specific 
edge cases weren't covered, but doesn't expect all edge cases will ever be 
fully covered, and that's just something we have to live with - Jonas thinks 
that the problem of "how to get all relevant messages on all devices" is almost 
fully covered by XEP-0313 and XEP-0280, and they're worth moving to Draft based 
on that. Georg has at least checked through the LC feedback for XEP-0280; Jonas 
skimmed through and mostly saw "Bind2" (for the long part at the bottom.)

Zash: [on-list]
Daniel: [on-list]
Jonas: [on-list]
Dave: +1 (the CVE bits can be added afterwards)
Georg: [on-list] (really want to see a discussion of the points I brought up)

Georg is most interested in properly specifying the storage rules in XEP-0313 - 
Zash wonders whether those rules will have to evolve with new XEPs, etc. (also 
applies to Carbons) - Georg agrees and thinks having them under a separate 
namespace is a good idea. Zash was thinking that new XEPs could describe 
whether their payloads should be stored. Daniel thinks the problem with both 
XEPs is that there isn't really good definition of what a message is.
Sam thinks all of this is a reason to advance both XEPs - they will never be 
perfect, and may need to change in the future, but both are already very widely 
implemented so they can't be changed too much (which sounds a lot like Draft.)
Zash believes subtly different rules are needed for MAM, Carbons, CSI, and 
cloud notify, but there is still a lot of overlap, though they will have to 
continue evolving in the future - Georg is very interested in extracting those 
subtle differences.
Jonas doubts this can be solved here, so further discussion is needed.

4b) Advance XEP-0280 (Message Carbons) - 
https://xmpp.org/extensions/xep-0280.html
Daniel: [on-list]
Zash: [on-list]
Jonas: [on-list]
Georg: [pending]
Dave: [pending]

Georg is very much interested in more list feedback, particularly from server 
developers; there wasn't an appropriate amount of feedback to the LC given the 
importance of the XEP, but Georg isn't sure if that just means it's working 
perfectly for everyone - Jonas suggests the community may be worn out after the 
long succession of LCs it has already had. Kev thinks everyone understands that 
Carbons is a necessary evil - Georg thinks it's a hack on top of a hack on top 
of a bad design from the turn of the millennium.
Jonas doubts this can be solved here, so further discussion is needed.

5) Pending Votes
One for Dave on Deprecating XEP-0013 (expires next week.)

6) Date of Next
2021-04-14 1500 UTC

7) AOB
Georg is working on a first draft of adding a CVE element for XEPs that works 
similar to a 'code' block - this isn't strictly a protocol thing (it doesn't 
affect the protocol itself, only implementations), but Georg thinks it goes in 
a similar direction as DOAP. Jonas thinks that having references to past CVEs 
in the Security Considerations is a good a good thing as it shows the severity, 
and rationale for specific rules.
Jonas feels that having it in the document metadata would be wrong because CVEs 
are about implementations, not protocols, so adding the list to the protocol 
seems generally incorrect; additionally, embedding the metadata list at the 
right place in the document would be a non-trivial XSL task. Georg hopes that 
having a custom element will allow for future extraction.

Georg also wanted to discuss his LC comments on both XEP-0313 and XEP-0280, but 
the meeting is already over time; Jonas suggests taking it to the list.

8) Close
Thanks Jonas, everyone, all.

Georg feels he should veto XEP-0313 as is because of the lack of business rules 
- the 0280 rules were a huge step forward, even if not perfect, and having them 
in 0313 would be another huge benefit; and business rules for clients.

Zash recalls talk of dusting off XEP-0226 (Message Stanza Profiles) - Georg 
thinks that counts as volunteering - Zash suddenly has somewhere to be.

_______________________________________________
Standards mailing list
Info: https://mail.jabber.org/mailman/listinfo/standards
Unsubscribe: [email protected]
_______________________________________________

Reply via email to