XSF Discussion - 2023-03-01

jcbrand: what are your plans for 425 moderation? Are you going to remove the usage of the 'retraced' element or are you going to keep it but just not in way that is compatible with 424 for the receiver?

Hmm.. Is it allowed for IQ stanzas of type 'error' to contain child elements other than the child element from the associated request, and an `error` element?

no

IQs are rather strict in that way

Daniel: I plan on incorporating the feedback soon. I'm a bit behind at work and need to focus on other things first. I lean towards the latter. Keeping the `retracted` element. I also think I'm going to put the `retracted` element back inside the `moderated` element. The reason being that I still believe moderation can be more general purpose than just retractions. For example in some cases one might want to allow moderators to edit messages, then you'd put a `replace` element inside `moderated` instead of `retracted`.

Or perhaps like the Twitter feature where they allow community-generated explanations/context to be added to certain controversial tweets.

You could perhaps attach something like that to a message.

Although I'm not sure that's a moderation use-case

Retracted inside moderated sounds better to me than the other way round

Yes, I agree. I switched it around because I though (based on feedback) that we could get XEP-0424 compatibility like that ✏

I no longer think trying to get XEP-424 compatibility is desirable

And doesn't look like it's feasible either

OK. Yes I agree. And no rush. I just wanted to clarify that this PR is still being worked upon because I think the editor almost merged it yesterday

Maybe put a draft label on it?

Well it has a `needs council` badge I see, although I don't think that's necessary since it's experimental

It's not Experimental it's proposed. But council will vote on getting it back to experimental today

ok

But after our vote today editor _could_ merge it hence the suggestion of flagging it as draft or something

Maybe I can make the changes today already but I don't want to promise it now

Yeah I marked both my PRs as draft now

If I've not screwed up too badly, my triage wrapper script should ignore draft PRs, so thanks.

Kev: If it's not like that already, I'd suggest that you also don't automatically merge PRs with unresolved threads.

Might mean that people need to be nagged to resolve their threads, but it prevents something that's still being discussed from being merged

Discussions shouldn't be happening on GitHub anyway, so that shouldn't be an issue.

I think it was here that we talked about moderation and stuff; a reasonable improvement to clients would be to automatically hide all previous messages sent by a banned participant, while still letting the user display the messages somehow if they want to know the reason of the ban after the fact.

Perhaps we should draft a new modernxmpp page about moderation workflows?

I don't think all banning should lead to messages not showing

Moderation and banning are two separate things

Daniel, the only case I remember for users being banned without having sent bad messages were when they reconnect in loop and flood a room with presences.

And in this case they get unbanned after they’ve fixed their setup.

We don’t have any event for unbanning someone though. :/

In every other case, the ban was in direct reaction to something they sent, so requiring a click to display the messages seems ok to me.

But if you moderate every message they send, how far do you go back? Really every message?

I read that as a purely client side thing

so it could be done ad-hoc when building the view

The moderate command of the clix tool basically takes a time range along with optional nickname and body contents to match on

Link Mauve: xmpp:abuse@joinjabber.org?join for moderation and more

I don’t currently have the bandwidth to join many more rooms.

But thanks for the invitation. :)

:-O

Link Mauve not joining one more room :o

Anyway, that was also to say we're thinking about it at joinjabber

pep., currently. :p

Kev: technical feedback on the diff can and does happen on GitHub

Link Mauve: are you on 75 baud? Also, that should still be ok?

:-D

:D

The mental bandwidth, should I add. :p

ralphm, I'm fairly certain that 75 baud is not sufficient anymore for the amount of MUCs Link Mauve is in.

People don’t write that much actually. :p

Link Mauve, chat states, presence.

This would all eventually balance, during night time for instance.

you sure about that? 75 baud is really not a lot

I have disabled systemd’s bandwidth accountance so I don’t have data any more. :(

the text-only content of programming@ since 2018 is approximately 0.079 bytes/second

(a bit less, because this includes the full timestamp, which is generally not included in messages)

so 75 baud, with *only* text, would suffice for approximately 1000 MUCs

Indeed

Link Mauve: have a look at Netdata (shameless plug)

(for your systemd metrics)

Careful. Does it run on a Nintendo Switch?

You don't know what you're getting into :P

MattJ: likely yes. We have ARM support.

Time to have XMPP Brotli compression XEP then

(EXI is more efficient for XML I believe, BTW)

But with other complications

(Thanks to complications & problems, we have work… What if there were no complications to solve?) Side-note about EXI: It’s ment for resource contrained networks, and is estimated to compress XML approx. 7 times more efficiently than LZW. Can be combined with UDP transport and DTLS, and thus create interesting alternatives to the more thightly coupled CoAP and LWM2M protocols. (Also permit huge amounts of connected resource-constrained devices to the same machine, as UDP & DTLS do not require socket connections.)

Hey, I don’t run my infra on any Nintendo hardware atm!

The last time I did that was like a decade ago, on a Wii.

Ah no, I also had my Switch as a build server for my phone and other servers for a while, until I found out about Oracle Cloud.

Might migrate some bits and pieces of that infra to a dedicated Wii U someday.

Peter Waher, are there any free and open source libraries these days for EXI?

which aren't java

But I still have a few things to fix in the kernel before I can use it properly, for instance hardware AES-CBC-128 support is still not merged in the kernel.

(It also isn’t really used any more…)

ralphm, that looks cool, thanks!

Peter Waher: do you have experience in running xmpp over exi?

Hi, I wanted to discuss if we should consider Google Season of Docs for writing XEPs? https://developers.google.com/season-of-docs/docs/get-started?hl=en

I've pondered it in the past, but came to the conclusion that we're not in a position we could provide a sensible framework for it for XEPs. More likely would be providing user docs for individual projects, I think. (Others may disagree)

Daniel: Yes, we have experience with XMPP & EXI.

Kev: So you believe it only works for software?

I mentored a master student to create a plugin to OpenFire a while back (Java), but not sure what happened to that project.

Jonas: Not that I’m aware of. I personally work with a commercial broker and libraries, in C# (many of which are open source however, yet commercial for commercial use). ✏

> : So you believe it only works for software? No, I don't think that's true. I don't think the XSF is is a position that it could sensibly support work on XEPs under that program - but I accept others might disagree. I think that the amount of domain-specific knowledge needed to work on XEPs in a non-trivial way is more than one would reasonably be aquired for the program.

Kev: but they say you need to have a past of technical writing, so it can be one that has the domain knowledge already

maybe lets ask in a differnet way: Can anybody think of how it could work in xmpp?

I meant XMPP domain knowledge, rather than writing knowledge - but I'm happy to accept that I'm being pessimistic about it if someone else can make it work :)

Hmm not sure if I understand your concern. Maybe elaborate where the idea does not conform with us? > Google Season of Docs supports documentation in open source by: > Providing funds to open source organizations to use for documentation projects > Providing guides and support for open source organizations to help them understand their documentation needs > Collecting data from open source organizations to better understand documentation impact > Publishing case studies from open source organizations to share best practices > Google Season of Docs seeks to empower open source organizations to understand their documentation needs, to create documentation to fill those needs, to measure the effect and impact of their documentation, and, in the spirit of open source, share what they've learned to help guide other projects. Google Season of Docs seeks to bring more technical writers into open source through funding their work with open source projects and organizations.

I see development of xeps also as a project or a type of documentation personally

and btw, in google summer of code we may get 500-1500$ - in GSoD its 5.000-10.000$ - but I hope this is see as a positive side effect, not the intention^^

What would Documentation on XEPs consist of exactly?

Anyway, even if it does not apply for XSF, it may apply for xmpp projects, here is some inspirational list from 2022: https://developers.google.com/season-of-docs/docs/2022/participants?hl=en

Does that require global understanding of XMPP or is it a set of tasks that can be achieved only by reading single documents at a time

pep.: Id specificly know, what would we like to see would be a good start? ✏

> pep.: > 2023-03-01 03:18 (GMT+01:00) > Does that require global understanding of XMPP or is it a set of tasks that can be achieved only by reading single documents at a time I assume it certainly does, but the participation expects you have a writing history. its nor like GSoC for OS newcomers

It's not about the "Writing history" that I'm worried but about the "Knowing XMPP"

pep.: but I assume we habe about 50 people here knowing xmpp, right?

« Hey, go read 6120/6121/6122 and then 0045, 0060, and also XX, YY, ZZ. Then we can talk"

emus, but this Google thing is only for students no?

pep.: no, I dont think so

> - Do you have technical writing skills and experience? Google Season of Docs is not intended to teach basic technical writing skills. Are you self-motivated and organized? > - Do you have time to participate during the six months of the program? Project sizes vary, but range from a commitment of 5-30 hours per week during the program. > - Are you interested in gaining experience working in open source? Are you willing to devote time to learning a project’s processes and community norms? > - Are you willing to work with the organization administrator to refine their project proposals, develop timelines, assess and compile metrics, and help the organization as a whole understand the value of technical writing? ✏

That is something to discuss > You should also meet the program eligibility requirements, including being able to create an Open Collective account. Please note that you must be legally able to receive funds for work performed in your place of residence. The Season of Docs program administrators are unable to provide tax advice or answer questions about your work eligibility.

Ok, then it's already slightly more into XSF's reach. Maybe send a call on Mastodon and all to see if anybody is interested?

But what would that mean exactly, to work on specs as a technical writer

> pep.: > 2023-03-01 03:24 (GMT+01:00) > Ok, then it's already slightly more into XSF's reach. Maybe send a call on Mastodon and all to see if anybody is interested? Letz maybe first start via members@

I'm not in there, you go ahead :)

ok I see

> But what would that mean exactly, to work on specs as a technical writer I cannot tell, everyone is invited to state their propsoals :)

I'm not 100% sure I know what a technical writer does.. Apart from writing technical things :/

Writer with a tech background? Documentation? That's not too reductive?

"writing technical things" is pretty much it yeah. It's a broad field :)

Writer who is able to talk to devs or other tech people to find out how things work then write coherently about that

> singpolyma: > 2023-03-01 03:30 (GMT+01:00) > Writer who is able to talk to devs or other tech people to find out how things work then write coherently about that And I think thats what every author of a xep od tech doc is

Sure, it's what every xep is trying to be anyway. A description of how some tech works ✏

Of how some tech should work*, or how we think it works (for "informational")

So.. a sponsored tech writer would.. do what? Improve existing spec's wording? Or rather think about how tings can be improved to present the information?

Is tagging also part of the thing? Now that it's possible to tag specs

As said - no clue, we can come up with proposals

I guess for the presenting part it's more up to a designer

"we" who

If we have no clue

maybe lets look at the summit 25 (2023) topics ✏

Any can help me finding access to the twitterdeck legacy view? I didnt manage to find how I made it last time

emus: Do you mean https://tweetdeck.twitter.com/ ?

jcbrand: yes, but its legacy view

@emus if I understand your point and it can be helpful, you can search via Nitter here: https://nitter.nicfab.eu/

nicola: I don't fully get what you are suggesting?

@emus I understand that you need to search something on Twitter

If you need to search your tweets, you can use Nitter and put your account. The platform will list all of your tweets

nicola: ok thats great - however, I actually need to send a remote tweet via tweetdeck

but it does not work with the new interface anymore

https://jabbers.one:5281/upload/hDg4xgTXY7g9hjhyfE9CsYQL/20230301_162054670_9856..jpg

found it

@emus can't you switch to the old one?

thanks for the hnt

hint

Wojtek: ^

yeah, saw your image after hitting send :-)

Sorry emus, I wasn't saying that the XSF isn't eligible (I haven't checked), but that I don't think we'd be able to manage it. But, again, maybe I'm wrong.

Kev: But what does it require? I can do the application. what we cannot manage about writing xeps? 😅

its not about getting new people onboard

I don't think we're in a position to suitably guide people on writing XEPs - what needs to go in them, etc. Maybe I'm wrong

Didn't we find that ChatGPT could do this for us? :)

But how did we produce more than 400 xeps?

noone was guided to write a xep really so far?

emus, I think the main reason people write XEPs is to fix something lacking in the protocol, because they have an interest in there.

Getting someone to write XEPs because they get paid to write documents, would require to describe a bunch of features we do want and teaching them all about XMPP, or something like that.

Unless you manage to find someone already involved, that’s unlikely to be a good match imo.

I guess the whole point of this wouldn't be to write XEPs then but to improve things about them

I know I've got a couple of XEPs that I ended up needing to write after the Summit - but trying to tell someone else what needed to be in them would undoubtedly be significantly more work for me than writing them myself (and I'm struggling to find the time just to write them).

Because yeah I agree, it'd be weird to have someone write specs just because

Well, leave the money point out and ask: - is there work most agreed or wish to be done? I remember there were 6-8 tech topics at the wall at the summit - Is anyone out there being capable to do without requirement to do so or was planned anyway? I mean, how many development things do we have in the pipeline? If we have absolutely no clue we might use the same track to study and outline this as GSoD seem to pay also for evaluation studies 😃 (Besides when people write xeps in their spare time why not use this chance when other got paid for it via their employer?)

If you want to get down in "what is important and what isn't, at the XSF" territory (to know where to spend the money) I think you're in for a lot of fun :)

Well, I see there are many topics with different views, but I bet there is also a lot common sense, too --> thats what I look for

Ooh, just spotted that Community > Service Providers on XMPP.org links to a dead page on xmpp.work

PR welcome ;)

Definitely will, if nobody else gets there first.

Thanks!

Is that the right place for "Service Providers"?

I think that section is just gone now

It's more "if you're a service provider, you might get work here" :)

xmpp.work recently changed hands and got some updates

IIRC it was like listing XMPP consultants (like me!) and similar service providers

PR is up

Thanks!

I posted about each project: Monal: https://twitter.com/xmpp/status/1631049299646709761 https://fosstodon.org/@xmpp/109950233974730123 Dino: https://twitter.com/xmpp/status/1631050907897978880 https://fosstodon.org/@xmpp/109950259189174110 Moxxy: https://fosstodon.org/@xmpp/109950267243301508 https://twitter.com/xmpp/status/1631051422597799936

would there be any interest in working with hello tux to get decent shirts with the xmpp logo? https://www.hellotux.com/

i've been working with hellotux to improve their shirt quality and/or do custom embroidered shirts on shirts I send them