XSF Discussion - 2020-07-16

o/

0. Welcome

!

Apologies for not sending minutes nor an agenda

Any particular items to discuss today?

I added two items, one this week, one last week

Ok, I'll take the items from Trello anyway.

I'll write the minutes from my last meeting right after this one.

1. Minute taker

will do

Thanks!

2. Reevaluate the process for accepting XEP contributions

I see this is still on Trello, but I think we handled this, right?

Yeah this should have been answered

iirc jonas’ was happy with the outcome.

Right

I think so, too

3. Figure out how to reconcile mission statement's goals with neutral stance

This is one of pep.'s items (struggling with the punctiation here).

There's a description in there that could be useful

I'm not sure if I follow. Is the question on whether the mission statement was officially adopted by Board?

ralphm, the mission statement says (first sentence): > The mission of the XMPP Standards Foundation (XSF) is to build an open, secure, feature-rich, decentralized infrastructure for real-time communication and collaboration over the Internet.

Yes, I agree this is and has been our mission.

Which to me means we want to encourage open and decentralized infrastructures

Right.

We want to *build* open and decentralized infrastructure

That's what it says

MattJ, we're not building infrastructures anyway we're building protocols

infrastructure is a generic term, meaning it's something that can be a foundation (in this case, the protocol)

This might be nitpicking, but doesn't defining the technologies, in the forms of our specifications, amount to building?

ralphm, might be nitpick but it's an important nuance

Which to me would mean there are use-cases to encourage more than others

And causes to support (to which the XSF is never answering even if it's right in our mission)

I think the rest of the mission statement neatly explains what we mean with 'build'. And yes, it is opinionated.

It's not in our mission, though I think you're trying to read it that way

MattJ, I'm not hiding I would like it to be that way (I hope it's known by now), though I do indeed read it this way anyway

Sorry, I don't read it that way

I.e. we do design for a decentralized architecture, and we do prefer people to operate their implementations in an open fashion, as part of the so-called Jabber network.

But because we believe in the openness of the protocols and technologies, we also not object to proprietary setups.

ralphm, but we don't actually care if the main usage of XMPP has nothing to do with decentralization, is the message we're sending?

Well, my theory here is that by not excluding that, we encourage the openness of what we do.

I'm happy for it to be, as long as it's clear

Do we

I'm not sure how to respond to that.

I have my own answer, and it's "no". But I would like to have an official answer someday :)

Answer to what exactly? I don't understand what's actionable here

Are you saying the mission statement is in conflict with what we do?

I was commenting on ralphm's statement "we encourage the openess of what we do"

MattJ, yes that's my original question, how to reconsile both

Both what?

what we do with our mission statement

I don't see any conflict

I think the alternative position is to somehow attempt to lock down what XMPP technologies can be used for. This seems like a non-starter.

ralphm, I'm not saying "lock-down", I'm saying encourage more one way than another

There are multiple ways to "promote", you can either put less resources on one side or more on the other

I don't see a problem with encouraging open protocols and decentralized systems, while at the same time not being concerned with using those technologies in other ways.

Do you think we've been putting resources on one side but not the other? I don't

MattJ, well then we're effectively not promoting anything are we. Where are these values of openness and decentralization

Or would you think this might be a problem if we put up testimonials of entities building systems, based on XMPP, which are not open or decentralized?

The fact that the XSF is in the minority of standards organisations that openly publishes its standards, doesn't require membership to participate, etc.

ralphm, that'd be an example yeah. I might prefer not to I think.

If you think that's not "open", I don't know what to say

What problem are we trying to solve here?

My own view here is that by showing that we build useful technologies (they are being used), our work has merit and a higher likelihood of furthering our mission.

We seem to be juggling with words mostly - what's the desired outcome here?

Guus, to me, either admit we're not actually promoting "openness" and "decentralization", or do it

I think we already do.

Well, we are.

pep.'s desired outcome is that the XSF becomes more involved with political causes and promotion of open-source decentralized communication (pep. correct me if I'm wrong please)

I don't think we do

You apparently want to do it in different/more ways. Please elaborate on those

Possibly not optimally, but we definitely promote openness and decentralization

pep., there's some enterprise evaluating XMPP as a basis for their closed solution. Would you prefer them going for something else?

MattJ, that sounds more or less correct, with a nuance on "involved", one because we are already, whether we like it or not (don't tell me openness and decentralization have nothing to do with politics), and also because our involvement doesn't have to be all or nothing

(My assumption being that opening the solution is usually not an option at that point, and the only question might be whether or not to base things on XMPP.)

E.g. by our presence at conferences and other events, swag we have handed out/sold, news letter, messages on social media.

I'm not sure XMPP is necessarily the best solution technically, but I think it's in our interest if they go for XMPP in that it might yield implementation improvements, help with marketing, and so on.

What's a specific action you'd like the XSF to take? Promoting sounds a bit vague?

So I'm not sure why we'd discourage such a usage.

I don't think we're discussing discouraging anything?

Guus, I was asking pep. about his motivations.

Well as an example "Decentralization" is not even mentioned on the main page.

rather than engage in other activities more

I'd personally want to support free software solutions, for example. Giving them resources that they don't have compared to private solutions

(Personally I appreciate if the XSF just concentrates on specs and that's it.)

pep., it is mentioned here, though: https://xmpp.org/about/technology-overview.html

ralphm, yeah..

I'd also want the XSF to be slightly more vocal and take stances on these two points (openness and decentralization) online

There's only so much that you can put on a front page. I don't see this as a problem.

What concrete things do you believe we should do then?

Things like the letter I sent to the list 1-2 weeks ago. That was the perfect example

I'm not asking to sign it now, I realized after the fact it was from 2019, but I would have wanted to sign it then

https://mail.jabber.org/pipermail/members/2020-June/009168.html

Put out a tweet saying "Federate or Die!" (this is not a joke, we've had people with t-shirts like that)

heh, I'd be happy if the XSF got there someday, even though I doubt it ever will

I personally not interested in such activism.

Underneath a snake?

ralphm, what are you even doing here then. I'm actually curious what is everybody doing here if there is not a hint of activism in what you do

Why do we even care about openness or decentralization

Anyway, I'm happy to have had this discussion with board with not so many bumps compared to previous times. I see nobody actually wants to change

pep., there's a difference between working on technologies and believing in openness and decentralization, and signing letters demanding things.

As as organization openly saying it has as mission to build an open and decentralize infrastructure, I wonder where you place the line

pep., if by change you mean: have more activism, as is more common in the Free Software movement, than indeed, I do not have that preference.

I personally 200% believe in decentralized and open communication tech for everyone

and I don't want an activist standards organisation

I want a standards organisation that serves as a place to collaborate on the protocol, with anyone who wants to do so

Indeed

I prefer the XSF to not engage in activism either. I'm not saying that there should not be activism, but I don't think it'd be particularly beneficial for XMPP if the XSF started doing that.

Ditto

Idem.

Which doesn't mean individuals or groups of people within the XMPP community couldn't, of course.

pep., FWIW my main motivation is decentralization. So I'm interested in good specs and good implementations as they help with that. In a non-ideal world like ours, even closed solutions using the same specs/implementations might help with that. Activism/marketing might obviously help as well, but that's a separate task.

Holger, you might have heard me wrong, I haven't (yet?) said kill all closed source solutions

secondary: adding 'do activism' on our plate will spread our already thin resources even thinner. This might get complex, fast: what activism do we do, what's our wording, etc, etc. I'd just prefer to not be involved at all. ✏

Can we agree to disagree here?

Anyway. yeah, next

Unsure to what extend we need to re-hash arguments?

Guus: well, I guess it is good to make it clear where we stand.

I don't think we've had this discussion so explicitly yet.

Sure. I feel we now have, to a large degree. ✏

And I feel that pep. has been hinting around this for a while. I think it is great that expresses his view point, and that we can agree to disagree.

Since we took our time:

4. AOB

> Put out a tweet saying "Federate or Die!" Having said all the above, occasional Tweets such as this one would be cool 😂

I'll probably be not available for meetings in the next 3 or 4 weeks.

Agenda items I've been meaning to add: come up with a roadmap. iirc there was one already that has never been finished?

Holger, I'm happy to follow your new twitter account

🙂

pep., can you put that on trello, then?

will do

thanks

5. Date of Next

+1W

6. Close

Thanks all!

Thanks

bye!

(Hadn't even noticed I'm babbling into your meeting, sorry 'bout that.)

Holger, I think input is good

from anybody involved

Holger: please do this more often

Reminds me, application time …

Input from "the floor" is awesome and we should have more of it.

Any news about the SCRAM order? SCRAM-SHA-256-PLUS > SCRAM-SHA-256 > SCRAM-SHA-1-PLUS > SCRAM-SHA-1 Recall by stpeter (June 2019) : "When using the SASL SCRAM mechanism, the SCRAM-SHA-256-PLUS variant SHOULD be preferred over the SCRAM-SHA-256 variant, and SHA-256 variants [RFC7677] SHOULD be preferred over SHA-1 variants [RFC5802])" -> https://tools.ietf.org/html/rfc8600

What news are you expecting?

Some people think that it is: SCRAM-SHA-256-PLUS > SCRAM-SHA-1-PLUS > SCRAM-SHA-256 > SCRAM-SHA-1 or SCRAM-SHA-256 > SCRAM-SHA-1 > SCRAM-SHA-256-PLUS > SCRAM-SHA-1-PLUS or other... ✏

-> https://github.com/xsf/xeps/issues/944

that seems odd. I’d prefer -PLUS if available, no matter the hash...

but it's got a bigger number

I read the discussion. Due to the recent desicion making on communication software in my locale, I have the following questions: If there is someone evaluating their communication, do they actually consider XMPP as a protocol to use? If yes, because they knew it by chance? If not, why not? I think thats the case for most evaluations. Is the actually use of XMPP part of the XSF efforts in general? Where do I "go" as a company/developer to evaluate this technology for my purposes? In case they do, do we actually have the chance to take part of their evaluation? I think those are a questions I personally wonder about if I read the mission statement.

People evaluate products. Not protocols.

Well, some people evaluate protocols, but they are the minority :)

I think what Mozilla has publicly documented is something I see when interacting with those kind of decision makers

Right, I think there was a surprising number of surprised people in this community when that happened

Wrt to people evaluating protocols I'm under the impression that we aren't even doing that terrible of a job

An open sane process, and no, "XMPP" cannot just magically be adopted by Mozilla

> People evaluate products. Not protocols. Ok fine, but then product creators do evaluate on this. Then lets go one step back and ask what those conclude? And then ask do we want to be part of that evaluation or leave them with that on their own?

Like I said. - or tried to say - I think we are doing OK on that front

Im not sure about that. And regarding creating an open and decentralised infrastructure/protocol, I think it almost completely failed (no offence), when measuring the result that is almost everyone is using closed and centralised solutions only. Few are acutally considering to join or even giving anything back to it or using it as it has been designed (open, decentralised). I also wonder why it is so hard (at least it feels like this to me) to question if we actually follow the mission or anything we head for.

One point that could be made and rejoins what emus is saying, if we don't push for open and decentralized implementations, then the protocol is only going to be used for closed solutions, just as it is today

(And what I've been saying anyway)

Because yes indeed people evaluate products not protocols. And XMPP is nowhere to be found

I'm happy to see Snikket becoming a thing, but we have to admit that free software XMPP implementations that can compete with today's competitors are very scarce

"The protocol is only going to be used for closed solutions, just as it is today" - seriously?

Mostly?

I can point you at the various pages on xmpp.org

That mean nothing

My family aren't listed on xmpp.org

*means

So 3 nerds using XMPP + Nintendo, Riot Games, etc.

There are tens of thousands of users on public servers

At least

And yet when Mozilla looks at the state of the art we're not even considered

And a few companies build stuff on top and you say XMPP is only used for closed stuff?

(Note that my issue is not that there is no closed source product using XMPP that Mozilla can use either) ✏

XMPP would have been considered had someone proposed it

What solution?

You tell me :)

Well I think we can agree there is nothing competing with Riot (Element), Mattermost, etc.

Agreed

So here we are

I specifically chose not to take Snikket in that direction (at least for now)

And the XSF doesn't seem to mind

Feel free to work on filling that gap

The XSF is not a software or service provider

That's exactly the kind of comment I don't want, "do it yourself". That's just saying "it's not our fault" (to me it's pretty close to "if there's no product it must be that the market has decided it's better this way")

If you don't do it, who do you expect?

"The XSF"?

The volunteers who all have day jobs?

MattJ, I'd be happy for example to have the XSF sponsor things like Snikket

Or things in domains they think is worth encouraging

(according to the mission)

and what if they think they'd rather send money to isode instead?

Snikket doesn't need XSF funds... the XSF doesn't have enough (it might right now but it's not sustainable)

MattJ, I'm not saying 100%

If you go down that road the XSF just becomes a fundraising vehicle for open-source XMPP projects

moparisthebest, if it corresponds to the mission, so be it

That would be a nice org to have, but again I'll repeat that the XSF should not be that

my point being, XSF members probably have very different ideas about what is "worth encouraging", so maybe it's best for it to just encourage nothing

Members voted 2-3 years ago in that direction

moparisthebest, fortunately there's a mission document under which we're all supposed to be gathered (with various interpretations as usual..)

Ok, I retract "voted"

I think there are better sources for funding - if you want it - than the xsf

It was an informal survey, but still

Daniel, I do want the XSF to set directions though

Even if it's not doing the funding

Why?

The XSF doesn't have the expertise to set directions

XMPP is used in more than IM

MattJ, what's the mission for then

Ultimately it always comes down to individuals doing the work, acquiring the funding or whatever

And even IM cases can be broken into different segments

In the meantime we can ask around to various organizations for their needs. Seems pretty obvious to me what Mozilla was going to go with

seemed*

And if you want to be that individual just be it

Why do you need xsf approval to get funding or start developing a client

Daniel, that's exactly what I don't want to come down to as I said above. To me that means us running away from our sheparding-ish responsabilities (whether we want it or not, it's very much what this looks likes to everybody else outside of the XSF, and even to some inside)

Honestly all your problems stem from wanting the XSF to be something it is not, and would be problematic to transition to

MattJ, "it is not" is a strong phrase. Could be :) ✏

As we have established in many ways by now, it could not be

Oh it could, it's just that some don't want it

Apart from anything else, a significant number of members disagree with you

A significant number of vocal members

I don't understand your obsession with the XSF, it is a volunteer org and does what its members do

You complain they aren't doing what you want, and don't want to do it yourself

Why the need for the XSF at all then if it doesn't even set directions

It is a place to build the protocol

We can do that without a front

Sure

The organisation and process and bank account helps though

Or nobody at FOSDEM would know what XMPP is

Well yes, individuals would do it? Isn't it what this is all about? Individuals :)

but do you as a person pep. actually want to do something? or do you want someone else to do something?

It is

for what ever that something is

Daniel, I'm obviously doing things

But what I'm doing and what the XSF thinks should be done are two different things

ok. but why is that a problem?

Well the XSF is not saying what they think should be done.

Also what I've been thinking about that roadmap again

if you think xmpp needs better clients go build a client. if you think XMPP needs more shit posting on twitter go ahead and do that

The XSF thinking it shouldn't do those things is not the same as thinking those things shouldn't be done

(have to go)

that's a perfect way to word it MattJ (+1)

if i could work full time on Gajim instead of 2 hours a week

it would look much different

lovetox: but would you want?

yes, but i guess thats a theme in life, we want the stuff we cant have

lovetox: that's what the omnious they want you to believe :p

MattJ, and as much as it's true it's irrelevant

So it's that you want the XSF to tell you what to do?

The XSF leadership is you and me and three other people

MattJ, not just me, but yeah

So the XSF says (hypothetically) that right, we need a solution for orgs like Mozilla

Then what?

Shows that at least we care about the state of the world (at least the small part in which we are)

And maybe suggests people to take action in these areas

That's not how open-source works, trust me

And on top of that we could say we'll sponsor that project up to x%

Yeah I don't care about open-source. I care about free software

Freeware then

?

Never mind :)

Sorry for missing the taunt

Picking on my use of the term "open-source" is quite beside the point of discussion, but you can pretend I said FLOSS if you want, it makes no difference

Then I don't agree with your vision of how free software works

I still don't see how we get from agreeing that we need something like riot to having something like riot

People will magically work on it if we suggest it

Daniel, the answer is a benevolent parent company throwing money at you for 4 years and then hooking up with some shady crypto businesses, clearly

Even if we transfer literally all the money in the xsfs bank account to JC and JC quits his job I don't think that would be enough to satisfy Mozilla

(why JC)

Because Converse is probably the closest thing we have to riot

Example web dev, and we are talking about competing with primarily-web projects

And that

Ok, well I'm sorry if you only look at things through the money angle

I thought this was about money

pep.> Shows that at least we care about the state of the world (at least the small part in which we are) pep.> And maybe suggests people to take action in these areas

Money is secondary

But yeah it's indeed an efficient way to say we care

"Money is secondary" is not a thing for most people :)

I personally prefer to work on improving the ecosystem around XMPP. Allow business opportunities to happen, which allows more people to work on XMPP implementations, which is a direct driver for improvements to projects to happen.

https://xmpp.org/about/xsf/roadmap see this for example. I think that was a great idea and I'd like to update it

I mean the XSF newsletter thing is I think something I'd personally like to see more of

the matrix people are big on developer advocacy, and the XSF kinda isn't

are there matrix developers? I thought it was just the 1 client

(I won't mention commteam not accepting my offer to work with a marketing person..)

(to those who'd think "he's just words and no action")

Unrelated, https://www.zdnet.com/article/european-court-strikes-down-eu-us-privacy-shield-citizen-data-transfer-agreement/

I don't exactly know what it implies yet, but this happened.

Actually, talking about "all on individuals", apart from the discussion today, it would be honorable to the active people to support their individual internal work here with, if appropiate, e.g. paid work. So we can keep the good things up and consistent and have an organisation which enables to focus on the actually goals. If everything is done voluntary by individuals thats a question of time till important things break apart? So this should be something that is serving for everyone internally. I also want to warn to exhaust people with that circumstance (everything on individuals). Thats also not really caretaking on the atmosphere if thats the understanding of the work of the people here. (And even if I mistook something, that should be an important factor.)