XMPP Council - 2017-04-26

seems it's about time

1) roll call

I am here

Different account Daniel is also here

Hi.

Dave Cridland, ping

Tobias: s2s to gultsch.de seems to be stuck. Since prosody doesn't retry after the connection failed once or something like that?

Sorry. Hello.

inputmice, it should do that, though.

inputmice, strange...should report that to mattj

2) Minute taker

I can do it

seems JC isn't here

Link Mauve: I dont know. We have seen weird behavior with other prosody instances in the past. Anyway we should probably look into this after the meeting

SamWhited, thanks

Sure.

3) Our SHA-1 migration plan

I haven't yet synced with Link Mauve on where we are and what needs to be done. Will do so right after this meeting if Link Mauve has some time.

Yup, good idea, I haven’t done any progress on my side either.

4) Date of next

same day of week, next week, same time of day?

Wfm

Works for me.

WFM

Date of next set for 2017-05-03T16:00Z

great

5) AOB

I think the voting period for XEP-0334 ended, right SamWhited?

Yes, any news on resuming the process of publishing? IIRC Kev said it was done, and SamWhited that it wasn’t.

Yes, the voting period is over for that

and with SamWhited's -1 vote it won't proceed AFAIK

so back to the drawing board I guress

*guess

I will commit to looking through the relavant XEPs and submitting patches to replace the hints in 0334 if that would be desirable, since I was the one who blocked it

That's primarily carbons and mam I guess

great..that sound like a way forward

And in carbons that's a simple delete

And mam requires another ns bump :-(

Did we ever have a discussion on maybe not bumping the ns while a xep is in experimental and use a tmp namespace for the entire time.

i don't think had such a discussion, or at least I don't remember it

Because I feel that conditions that deal with the slight differences are easier to do than handling six different namespaces

Especially if you look at the recent changes in jingle ft and mam

inputmice, but that way we would end up with incompatible implementations, which would have no way of knowing what is wrong with other ones.

Can't you do both? Eg. just handle those slight differences, and then add a few more namespaces to the switch statement that directs JFT stanzas to the correct parser (or whatever)

i guess the only way is to write implementations that can deal nicely with multiple versions

This might be better for after the meeting or on list though

SamWhited: yes

Let's move on

Any other AOB?

16:08:43 Link Mauve> Yes, any news on resuming the process of publishing? IIRC Kev said it was done, and SamWhited that it wasn’t.

yeah..there're lots of stuff missing on eos...totally forgot about it last week...will take a look at it

I think it would probably be easier to get Kev's docker thing setup than it would be to reverse engineer the special snowflake environment that we were using before, but that's just a hunch.

And since I'm not the one doing it ¯\_(ツ)_/¯

happy with waiting for kev to do a virtualenv in docker in a vmware in a virtualbox

You should probably keep track of all that on a blockchain or something too

:D

any other AOB?

Doesn't look like it.

thanks everybody

chat logs are down, so no link in the minutes today

thanks for writing the minutes