XSF Discussion - 2019-05-15

emus and I are thinking about doing some kind of introductory presentation for people that never heard xmpp before. Are there any latex/pp/... Templates that already exist?

COM8 there's a list of (old) presentations on the wiki that might give you some ideas: https://wiki.xmpp.org/web/Presentations

> When? January. Gotta love those. Also it's missing my subversive anti xmpp presentation.

so if you submit something to the inbox, council votes on it and it gets experimental or rejected

how does on submit a protoxep that can be worked on?

Technically nothing prevents you from working on it and resubmit it? Council should provide feedback as to how improve it, if there is room for improvement, or blockers, or at least why they rejected it

so are you saying there is no protoxep stage?

experimental = protoxep?

no t really sure what protoxep means

lovetox: what about pausing work for the three weeks that the XEP is in council?

Ge0rG, not sure what you are trying to say

im refering to the email dave sent to the list

Let me read that first, then

he talks about a protoxep stage, a stage where feedback is gathered and work on a xep is made

im not aware of any xeps in that stage, and wondered if he meant experimental?

lovetox: there is an inbox for XEPs, and once something is submitted there, the Council will vote whether to accept it or not. Those in this stage are called protoXEP

After accepting, the protoXEP becomes a real XEP with a number, in Experimental state

yeah so if you submit a xep to inbox, this means council can outright reject it and there is not much work in this stage you can do

see for example ATT protoxep a few weeks ago

It got feedback that can be dealt with

And nothing prevents the author from pushing changes

lovetox: it's a little bit awkward because the author shouldn't change a protoXEP currently under review, yes

there are a few criteria for acceptance to Experimental, and ATT failed one of them (a clear answer to the question "what is this about?")

if your XEP is rough, but can answer that question (and has no other flaws which prevent moving to Experimental), it’ll be moved to Experimental and that’s where the main development should happen

It's usually pretty hard to get a protoXEP /not/ be accepted to Experimental. Mostly just "Is obvious what it's trying to do, isn't obviously broken, and isn't duplicating something we've already got without improving it".

Kev, ha, now I’ve got you: https://github.com/xsf/xeps/pull/750 pls review

Will try to look tomorrow.

Kev: improving or taking a different shot at the same problem I'd say

Well. It *should* be pretty hard to get rejected. Perhaps it isn't right now, and that would be a problem.

We would know if the XSF had only one solution per problem

pep., it’s not about duplicating solutions, but duplicating specific solutions

XEP-0390 was accepted despite XEP-0115 existing

But also, if there's discussion around improving a ProtoXEP, that's usually a sign it should be adopted, given a number, and improved within the XSF.

jonas’: I don't understand the difference you want to make

In this particular case, there's an existing full stanza encryption XEP, of course. But I don't see that as a bar for any other.

I think discussion alone may be an indication, but isn't conclusive, in the case that something is horribly broken, which might sometimes happen.

e.g. a security related XEP where the security model is obviously broken.

I think in that case accepting it to Experimental isn't the right thing, because of the potential for harm if anyone did try to implement it.

But other than that example, pretty much.

I am all for an IETF I-D like process for ProtoXEPs: Just upload the ProtoXEP, get an stable versioned identifier and have something to point people at

Since i really feel like the XSF is missing something like that

The Experimental stage is what is meant to correspond to the IETF 's draft stage.

So what's the equivalent to 'bis' documents?

We just give things a number sooner. The difference is that we act as a working group, and the initial vote is like an apron to a working group draft.

dwd, not sure about that. The requirements are definetly very different

Adoption. Not apron...

Flow, they shouldn't be much different to WG adoption.

dwd possibly, but then it is not what like what I had in mind

Flow: our process is just fine, IMO, as I tried to put in words in my email to standards@.

Aaaw yeeeaaah, XSL! https://xmpp.org/extensions/xep-0030.xml

Can we not fake .html and serve text/xml instead? so we don't have to convert anything ourselves :-°

Technically, yes.

Could also be rendered on the fly by the web server

So many options!

Clearly it should be rendered client side with webgl and rust compiled to wasm

I even heard pep. was working on that

wat