XSF Discussion - 2020-04-02

Hi

Sarathoma, hi

Ok so I have a question

ask away

how do I register for a jabber account

like what site is if

1. pick a provider 2. register with a provider 3. ??? 4. PROFIT

Sarathoma, "xmpp.org" doesn't provide accounts

What’s a provider

I just need a account with validcc . Su

then make an account with htem

they should have a website explaining how to do it

just like email

I cant.. says I need to have a jabber

https://www.google.com/search?hl=en&q=jabber%20account

(sorry, it’s early morning here, I haven’t had my coffee yet, and I genuinely think that google can do a better job at that than I can right now)

Also I'm pretty sure validcc is an illegal market site

I assumed the same

> How to Create a Jabber Account: 15 Steps (with Pictures ... So many steps...

I should’ve sent them to Conversations.

🙈🙉🙊

> 🙈🙉🙊 Yes, I understand people may run into issues with xmpp or have questions. this should not be the case anymore, but is actually

https://xmpp.org/getting-started/

Let's try to improve that, instead of pointing people to third party instructions.

having linked that would probalby also have been good

Guus: Yes, I agree with you. We can link to thrid parties for sure. But it should be somthing there aswell 👍 I will look if I have some ideas

isn't it perfect already?

Ge0rG: Nothing is ever perfect 😉

Ge0rG: How can it be perfect if they dont list Yaxim or Bruno??

I'll walk myself out.

Ge0rG: What do you mean? 😅 (srsly I dont get it)

emus: I was being sarcastic and failed

One thing we definitively should do is to somehow put it the link to "Getting started" more present on the main page

Ge0rG: I thought so, but I still didnt understand it 😅 ✏

I'm afraid putting this on the main page is just going to accentuate the pains we've encountered just trying to sort out active/inactive implementations.. because we can't decide on what's neutral and what isn't

the only safe thing is to link to lists of lists of servers and clients.

I think that's a really bad idea

"here user, you might be able to find something in there somewhere, good luck"

pep.: I've argued that point in the past as well, yes. But listing implementations is an obvious violation of the XSF's neutrality.

I remember 404city being very loud about his server not being #1 on the server list on xmpp.org

> pep.: Then we can just do an "official" hostet thing of XSF if we want to be neutral I know that provcative ✏

And yes, I've given up getting yaxim and/or yax.im onto that list

pep.: was not you state ment sorry

emus, I think for this type of user, it’s more important to have a support room which is prominently linked and web-chat-joinable on search.jabber.network

I’m 99% certain that those people join here via s.j.n

the influx started when s.j.n added the webchat links

Who is going to do the support for carders and terrorists tho?

> pep.: I've argued that point in the past as well, yes. But listing implementations is an obvious violation of the XSF's neutrality. But do list? 🤔

> Who is going to do the support for carders and terrorists tho? Cmon guys, actually we were talking about how to improve

emus: surprisingly, xmpp is mostly used by the underground

Ge0rG: I know, Im talking about the other side

Well 80% of the traffic through the anon mucs is exactly those people

> emus: surprisingly, xmpp is mostly used by the underground _people who have something to hide_

I wonder what would happen when I ask a friend "Make a xmpp account and send me a message" nothing else

people who have zero clue about opsec

emus: you'll get a long list of findings that are well-known and that nobody cares to fix

Daniel: Yes, but crininals will always find a way. The discussion was about normal newcommers

emus: we need a landing page with two buttons: "Are you a criminal? [ Yes ] [ No ]"

> emus: you'll get a long list of findings that are well-known and that nobody cares to fix Ok, what else? ✏

.... thank you

[yes] [no] [inshallah]

the first hit for "Make an xmpp account" is xmpp.jp

I wonder what they do right to pop up there

was there any progress on the badges front bTW?

> Ge0rG> I've argued that point in the past as well, yes. But listing implementations is an obvious violation of the XSF's neutrality. hence why I think the XSF is not fit for this

the neutrality thing is a pain

maybe we are doing it wrong anyway. how many people wake up to the question "how do I join xmpp?"

Ge0rG: true

maybe

> hence why I think the XSF is not fit for this Can you explain? At least for me? 😅😄

> maybe we are doing it wrong anyway. how many people wake up to the question "how do I join xmpp?" So you think we need to advertise xmpp in general?

If so I think the attempts are hardly connected ✏

emus: no, we are targeting the wrong people

emus: what we need is marketing to small/large teams interested in better control over the data and availability, like Slack / Teams

emus: also we need easier ways for an existing user to invite friends

like... XEP-0401

And good clients

like Pidgin!

I should write and deploy a mod_pidgin_warning module that will disco#info or 0092 query each new client and send a warning message to install dino/gajim

Ge0rG, Daniel: ok, I will think about it. One picture I had in mind once I hear about the neutrality thing, and I dont wanna blame, was a bit the mother is not accepting their childs 🤔 I know thats likely wrong but just saying how it felt at one point. I also wonder how we want to be absolutely neutral? each member is follwing their own personal interests, in the end too, isnt it? Maybe we need to find a way to deal with not being neutral in the end? But I'm getting philosophical, I hope you can follow what I am wondering about ☺

But still, good that we actually discuss that

emus: the XSF is limiting the number of members from a single organization.

emus: so having different member interests is a good thing because we can figure out what's best for all

Ge0rG: Yes I got to know that 😅

emus: :P

Yes, true

emus: also if you have three children, you can treat them all equally.. but the XSF has... dozens? hundreds?

Yes, thats not the correct picture obviously. just a hint.

With focusing to the standards in general it felts like the actual outcome are a bit apart, thats why I got that picture and was wondering since I knew about xmpp etc.. But I know this is intentional, but brings us back to the discussion start

Start a separate organization, like a advocacy group? And a lobby group. And a software foundation.

Zash: I know this is in the talk for long time, but I think its a waste of energy to split this

Hi ralphm, goffi, on https://xmpp.org/about/technology-overview.html the Idavoll component is now a dead link, and sat_pubsub is missing.

You may want to change that.

Is JC here? It appears that he is making changes to his ProtoXEP which already got proposed. IIRC we try to discourage that, for good reasons, as it introduces ambiguity which version of the XEP is under vote. But still, it shows that the ProtoXEP/inbox process should be reworked. I'd really like to have a phase where people can present their ProtoXEPs in HTML under a stable identifier while incooperating the feedback from the community (i.e., what JC's appearantly currently doing)

Link Mauve: oh, I need to fix that. Looks like a configuration error. Thanks.

Check the Converse.js room?

flow, ↑

Is Prosody missing from the list of "servers [that] include built-in support for PubSub or PEP" ?

Zash, thanks

flow, we could, with minimal effort, modify the process to simply allow development as ProtoXEP under 0.0.x (where the editor will just hit Merge if it builds) and have the author explicitly ask council to vote on it.

jonas’, i think what we really need is a central platform where people can work on ProtoXEPs, similar to IETF I-D system

like... xsf github?

Ge0rG, the point is that you want those ProtoXEPs rendered, not in XML

and stable identifiers for versions

flow, so? ✏

Zash, ideally stable identifiers to immutable versions yes

but I could live without it for a start

editor hits merge, possibly even with a bot which auto-merges all the stuff in inbox/

(if it builds)

ah, stable identifiers for versions in inbox are a different and tricky requirement

not going to happen without incremental builds in my book

and incremental builds are not going to happen with the current infrastructure

I don't think they are tricky or something, just require a bit more effort

something something dvcs revision hash

what's wrong with having the author work on the XML off-repo while the proto XEP is being voted on?

what we could do easily is mkdir incubating-xeps/ in the xeps repo

Ge0rG, nothing, but that’s not what people want

I don't get people

flow, why that if we already have mkdir inbox?

and allow people to submit xeps there and have those xeps published via the normal git xml xeps to html process

that’s what’s happening already

we just don’t have HTML-rendered-versioning for those

jonas’, i would keep inbox/ as it is: for protoxep submission

I don’t think we need a separation

KISS

I think we do, while it also think it's KISS

alright, that doesn’t matter either way

if we want those to be HTML-rendered versioned, it’s going to be a pain

or even HTML-rendered is going to be a pain for fast-moving incubations

with a 1h+ cloud build process ✏

Ge0rG, what's wrong is that authors do usually not do it. And even if, it would be nice if the XSF provided facilities for that, so that not every author has to do it

jonas’: cp protoxep-0.0.1.xml protoxep-0.0.2.xml ?

Just do the magic xml trick that makes https://xmpp.org/extensions/xep-0001.xml work and render

(this being the actual XEP file)

flow, not on my watch ✏

editor just has to ensure that existing things do not get modified after they have been published

we have a version control system

I’m not going to endorse keeping different versions of the same document in the same commit

ahh, there goes the KISS principle, na? ;)

we need per-xep git repositories!

Is there some web server magic you could do to get stuff out of the repo directly and render it?

Zash, not going to happen

at least that’s my last bit of info from iteam; no builds on the server.

jonas’, I note that we do exactly this for attic XEPs

flow, yes, I find it terrible

I don’t want this in the xeps repository, period.

jonas’: I mean as closely as possible to serve data straight out of the git repo, no "builds".

Zash, I guess it still qualifies if you run an XSLT on the server side.

Unless you insist on application of xslt in the web server counting as building :)

I don't see an issue, has served us well in the past years, and we could establish something for unsubmitted ProtoXEPs within a few hours. And it doesn't mean that we can not change/improve the process later on

might even be worse, because it happens on each request

jonas’, or clientside! which works already!

flow, note that xep-attic even has rendered HTML files in it, which is worse. xep-attic is a place I don’t look too closely at and which I don’t treat as a place for version control, but as an append-only storage. See the commit log if you wanna get an idea. xeps is for version control. I’m not going to endorse it being abused as a data dump, even in a subdirectory.

If you want a free-for-all playground space that can probably be done in a separate repository.

though I’m still not sure it’s a good idea, and I’m hell as sure not maintaining it manually

I could live with that. The infrastructure for that is already there, and I would be willing to maintain it

it isn’t

and to be honest, before we think about adding new infrastructure, we should get the existing stuff in order

we still don’t have a working registry

and I’m not happy with editor team members adding more stuff to maintain, infrastructure wise, before we have the actually right now needed extremely important stuff broken for years

well I do consider xep development of upmost importance

I think having our current xeps usable is more important than the legally questionable corner case of developing a XEP outside XSF IPR, but sure

legally questionable?

sure; anyone contributing to the discussion in protoxep stage could claim IP on the modified document

since the protoxep only enters the regime of the XSF IPR statement once it has been approved by $approvingBody

that’d have to be sorted out, too

but discussion in protoxep stage is already happening

(even if I would consider it an issue or even legally questionable)

that something is happening doesn’t make it legally unquestionable.

doesn’t mean we have to officially endorse it until that is sorted out

besides, we could simply have the IPR statement being taken on submission

we could, though that might raise the bar for some submitters

they’re giving away their IP before they know whether the document will even be handled under the XSF umbrella

how does that raise the bar, first time submitters always have to sign IPR

but that’s really not a question the editors should handle

flow, they have to, but the IP signover only happens on approval as Experimental

if the approvingbody rejects your XEP, you keep your IP

that won’t work if we need the IPR to be effective before approval because of the incubator space

because then non-author people have already contributed under IPR rules.

again, #notmydepartment editor wise

going back to work

be it then, if this is an issue for the submitter, then he/she can't use the incubator space

to summarise my position: incubator space is nice-to-have, needs IPR resolution, not going to endorse any infrastructure which keeps revisions as individual files in xeps, not endorsing anyone working on additional document infrastructure until we have our already broken existing infrastructure sorted out.

73

it's also news to me that the IPR is only assigned on acceptance

flow dwd is looking for you in open_chat (he can't join this muc for some reason)

> jonas’> I think for this type of user, it’s more important to have a support room which is prominently linked and web-chat-joinable on search.jabber.network > I’m 99% certain that those people join here via s.j.n I'm not sure that's an easy correlation to make. I'd be interested in stats of people clicking on webchat links, or stats for xmpp.org/chat referrers

is xmpp.org/chat linked anywhere?

s.j.n search results?

and wiki

and.. xmpp.org

didn’t know about the last two

The JS snippet that handles ?room was lost tho

Guus [13:46]: > he can't join this muc for some reason I put my bet on asymmetric s2s

!

Hello, all good? :)

MattJ, Guus

0. Welcome

Hi! Hope you are all fine. Who do we have?

.

Hey

All here. Yay

:)

1. Minute taker

Same as usual, I can do afterwards if nobody..

Thanks pep.

(note: I don't like taking minutes as much as the next one)

2. Jonas => iteam

I motion we add Jonas Schäfer to our Infrastructure Team.

+1

+1

+1

+1

+1

Motion carries. Congrats jonas’, and good luck!

Thanks jonas’ for MattJ :P

Haha, same! Glad to see some more help

3. Commitments

Regarding sponsorship, I've _just_ sent the updated text for the e-mail. Sorry about the delay. I'll send it out tomorrow.

Regarding comms person, I learned from peter that we don't have any template contract and we'll probably need to hire someone to hire someone (a lawyer)

I'll also modify it slightly to request new sponsors, and send it to members@, standards@, operators@.

Indeed. However:

> Regarding comms person, I learned from peter that we don't have any template contract and we'll probably need to hire someone to hire someone (a lawyer) I would prefer to focus on defining the job itself and responsibilities

I've received a suggestion for a company that would be willing to help out. I've been introduced, but will follow up first before I can share more.

Seve, we can do both in parallel

oh dear

thanks for the trust

We should know how to do contracts anyway

Just to know what to expect, if it fits the budget, etc

Seve, yes that's the information I'm trying to gather if you read the trello task :)

ralphm that's exciting - in what capacity can they help exactly?

jonas’, I'm going to assume this is still about adding you to iteam.

Guus: helping out with marketing

I should've been more clear in my question 🙂

I've asked marketing people I know as well, started drafting requirements for the job position

I think it is smart to research all of the above.

We can use all the help.

is this a marketing-oriented org? What's their motiviation to help?

Thanks pep. for looking into it.

I'll need a marketing person anyway to help me draft this more thoroughly, hence my asking marketing people :)

Guus: a company familiar with XMPP and the XSF, that would like to contribute. I'll share more at a later time.

And re lawyer, I also asked people who could know, but if anybody also has an idea I'll take it

why I'm asking: if they wan to contribute/sponsor, we might not need to go down the (lengthy/cumbersome) path of having to draw up contracts.

Guus, I think that's a good skill to have anyway tbh

If it's not marketing it'll be something else

I'd prefer to cross that bridge when we need to.

ralphm: sounds very exciting! Hoping you can share a bit more soon

Ok I can bring the bridge to us if you insist

I'm not against anyone volunteering 🙂

:-D

I meant I can put more work on board :P

Ok. Ongoing.

4. Infrastructure Team budget

While I am not against providing a budget, I'd like to have an overview of a plan for our infrastructure. When we need to pay for services or hardware, getting an OK can be done quickly, and if even more speed is needed, I can approve in my role on the Executive Committee.

I.e. I have no idea of what the size of the budget should be.

Basically, you're making a distinction between "can iteam spend money" and "can iteam have a budget of their own"

yes

And my answers right now are: yes and maybe.

I'm thinking that it is desirable for iteam to have the option to spend money to improve things, if they deem that beneficial.

There isn't a settled plan yet (partly because I was hoping we would hear something about the down server by now)

One problem is that right now, officially, only Board or an Executive Director, can enter into agreements.

The one thing that worries me more is have some "emergency money" for iteam Lead to spend in case of what happened recently. Unless we want that person to pay upfront and then have XSF reimburse the amount.

But we now have a DigitalOcean account that is linked to the XSF account

(bank account)

so although that was done to resolve an emergency situation, I feel like some clarity would be good

Which I approved, so that's ok.

Apart from that, I expect expenses should be more or less the same, on a certain period of time and services.

Seve, I think that's a pretty meh precedure and iteam should at least be able to spend some money itself

pep., I don't disagree that they should. I'd like to get an idea of what kind of cost.

And I'm fine if that means giving the iteam lead a way to enter into agreements

It's looking likely that we may want to add one or two more small servers, and so I want to make sure that any spending growth in this area is acknowledged and accounted for

Sure

ralphm, ^

I plan to work more (incl. with jonas’) to come up with a plan, so if the answer is "come back when you have a plan", that's fine with me

I'm kind of ambiguous about iteam getting a budget. I do not want to lightly change who in the XSF gets to approve entering the XSF into agreements.

What kind of agreements?

I fully trust the current iteam lead - but we do not know what will happen in the future.

pep., adding a server is not entering into new agreements, IMO, so MattJ/iteam should be able to do this.

k

Guus, then "we'll cross that bridge when we get there"? :P

agreements putting cost or responsibilities on the XSF, MattJ

But e.g. if we wanted to sign up for a third-party monitoring service?

We've in general given iteam room to do whatever is required to run our infrastructure. If that incurs cost, that should be fine.

MattJ, you'd need an ok.

sure - I'm totally happy with money being spend.

Ok, good to know

And if that needs to be done in between Board meetings, I'm sure we can.

Maybe we could start using the list someday..

pep., in this particular case, all of the infra was out, so I had direct chats with MattJ :-D

yay for federation

back to the issue at hand: do we need an iteam budget?

(list was still up tho)

Zash: (oh, right)

Guus, it sounds like we (iteam) need to come up with a concrete plan and then get a budget approved for just that plan

Guus: I'm happy to establish a upper bound, but I cannot tell how high it should be

and again if we want to add anything else

MattJ I have no issue with that at all (and I'm not saying I have an issue with having a budget either)

MattJ: or, once you have that plan, use that as the basis for a budget, and adjust if needed.

I'm fine with that, fwiw, just seems a little more work, but what else would we discuss at meetings? :)

So let's put this off for now

Ok.

5. AOB

?

GSoC

student application phase closed

I think I've seen around 8 proposals, of which a couple were of very low quality.

but I've seen a couple that looked promising too.

(Note that I only glanced at them)

but anyways, I wanted to share that update. We have some activity there, we might even get a couple of projects.

(that's it)

Thank you for sharing Guus!

Great, thanks Guus.

what was the last state re Compliance Badges?

That we tried contacting the designed, and haven't succeeded yet.

designer

I'll have another go with that.

maybe we should move on with the ugly non-designed badges in parallel? just in case?

I'll try for a little bit more, and we can consider alternatives after. I know it has taken long.

ralphm: thanks very much!

6. Date of Next

+1W

yep

wfm

7. Close

Thanks for all being here despite DST changes :-D

Thanks

Super, thank you all :)

Thanks all

luckily we're all in the same DST-boat 🙂

Yeah, and arc wasn't around 😋

stop rocking the boat, ralphm 🙂

sponsorship message looks good to me ralphm

tx