XSF Discussion - 2021-10-08

Was the last Office Hour recorded?

sam ^ ?

I think we should make a strategy how to lobby for xmpp..

Sounds great

The way to get things done is to do them

I think people are more open for xmpp then we believe. But there are two aspects. To many clients. And to many servers.

Peiple like it more simple.

So some of the client and server projects should stop development and be taken down?

To much energy wasted on to many clients and all are not "perfect"

So which ones should be allowed to continue, and which ones should we force to stop development? (and how can we stop them?)

MattJ: wrong aproach

Then explain to me what you're suggesting the right approach is :)

I am thinking about something like a organization of develipers. A scrum . today we make your house tomorrow we make my house.

I don't see how that helps

There will still be "too many clients" and "too many servers"

if only there was a project which tried to gather a good client for each platform and pair them with a well-configured and easy to install server

bonus if that project can gather some funding in order to support those clients in becoming better.

I thought to organize the developers somehow into some kind of association or foundation, and then they can vote on what the priorities are. I think it is important to have priorities, and to proceed methodically, whose house will be developed first can then be voted on

christian: Feel free to create such an organisation :)

What's that, some kind of XMPP Software Foundation?

christian, how will you incentivize developers working in their freetime on stuff to work on stuff they don't enjoy working on?

(if that stuff gets prioritized by the majority)

Wow strong reaction

I mean that's the gist of it, isn't it?

Its a unfinished ideea. Sorry.

you need to get people to work on the not-so-fun things to get polished clients

christian: getting people to do thing they don't like to do, for free, is hard

christian, I can totally relate to your point, splitting up the (very) scarce manpower into several competing projects (with the result that none of them work really well) feels like an insane waste. Problem is, there's no solution. Except maybe with lots of $$$ 🙂

Also, is it really a waste if people *want* to do what they're doing with their spare time? :)

christian, _suggesting_ a dev to work on $otherproject instead is pointless, the dev was aware of $otherproject before your suggestion and has his reasons to prefer his own. _Forcing_ a dev obviously won't work.

Kev, if the question is "what's the optimal allocation of our scarce resources to maximize outcome" then having several competing projects doesn't seem like the answer to me, right. If you ask a different question then it may just be the right answer indeed 🙂

(BTW it's not just about spare time projects in my book.)

Also skills. Imagine that the first thing the org votes is that "XMPP on iOS is terrible and should be our #1 priority", but it's impossible to develop for iOS without Apple hardware, an Apple developer account ($$$) and knowledge of the Apple APIs, OS and ecosystem

and most XMPP developers lack some or all these requirements

(and this is separate from the motivation to actually work on this ecosystem even if they do have the skills and resources - I've spoken to many iOS devs who just don't want to spend their free time working on iOS stuff)

That's stuff that actually _could_ be tackled by some organization/whatever tho. I think incentives is the real issue.

> (BTW it's not just about spare time projects in my book.) Sure, but it's probably unlikely that Process One are going to sign up to do some work on Isode's M-Link :)

It should be explained to the developers that we are in a vicious circle, bad clients -> few users -> little money -> even fewer developers -> even worse clients -> even fewer users etc... I think you should explain to the developers that the network is in trouble.

Bad clients?

Kev, right, I'm just babbling about why exactly this seems unrealstic. Might be partly due to how market economics work, which doesn't necessarily yield optimal allocation.

christian, I don't think developers need explanations. They're probably aware of the problems you see.

mdosch: yes, we must polish the clients. ✏

I think if the point is "If all the XMPP devs worked together they could potentially make better (but fewer) projects" it's fundamentally likely to be correct. Getting that to happen sounds like an insurmountable problem, but maybe it just needs someone smarter and better with people than me.

And if all of humanity worked together we could probably solve all kinds of problems.

Quite.

Zash: But I believe that developers are on a higher level than politicians and the military. We should succeed in things that they do not succeed in

christian, question is, who is "we". You? 🙂

We all here.

huhn: it was recorded, I just need to finish editing and post it. I'm off work today, so soon

christian, I've often seen users / community members stating great (or not so great) ideas about what others (i.e., developers) should do different from what they're doing right now. That's the part that doesn't work. If they wanted to do things differently, they would just do it. The assumption that you have a great idea for them they didn't think of themselves is usually wrong.

christian, so what makes more sense is thinking about what you could do _yourself_. (Which was MattJ's initial response.)

Be the change you want to see in the ecosystem

Which is what many of us are doing daily, and no amount of voting by others is going to stop that :)

Daniel: I use and recommend your client even though the blabber people are my friends. Maybe we should start here.

Holger: I am on telegram a lot (and many others) trying to persuade people to move. Success rate is 5...10%. You must see me as a kind of "sales department".

Make good things and talk about it.

Telling people to switch mostly just makes people dig in and become defensive in my experience.

I am the 'talk about it' department

No, not at all. The ux is the big thumbstone. Clients are designed for nerds

Not for dummies

The interfaces are no "toys"

christian, we have social media accounts and we are currently polishing the XMPP website and also advertise clients better. That also means that potential developers can see existing projects. We are also happy to publish content and news on XMPP

but also support developing

by providing helpful information

If anyone feels offended I am sorry. Did not mean to attack anyone. Everyone muddles on as before and we remain friends.

I'm not aware that anyone feels offended, don't worry. But you raised some proposals and others raised questions/concerns with those proposals. You're right that we're all muddling on doing the things we feel are important, I hope you do also :)

Another problem is that it can feel like little progress is being made when you only see a small part of it, but there is a lot of activity going on around the XMPP ecosystem. I don't think the network is in such danger as you described.

In that regard I recommend to subscribe to the newsletter through on of the channels 😉 😛

(and help distribute it)

👍️

oops ✏

Okay, next time when test users will tell me that xmpp has not so modern and up-to-date clients, and are not intuitive enough, I will disagree, and say that it just seems that way, because you don't have the overview of what's going on in the ecosystem.

Well, I think we are not arguing against this point, that many clients lag one or the other feature, but we currently have no better idea part from that most here love to keep it open. There are also many commercially developed client applications or XMPP implementation. However, many go with leaving the open side of the network behind them ✏

emus: we had pidgin. A legend, and we let it die.

Pidgin ain't dead tho

There's a *lot* of work to do to bring its XMPP support up to date, but it's not dead ✏

I think Pidgin is not what many connect with XMPP anymore, it had XMPP as one of the implementations.

unfortunately, it *is* in fact what many connect with XMPP

it is a cautionary tale

not a legend

Zash, actually I am interested, where can one follow it? jonas' yes agreed ✏

"we" didn't let anything die. "we" don't currently develop any software. To be honest, Pidgin was never that great, and its XMPP implementation had many issues from the start.

> Zash> Pidgin ain't dead tho I think you mean it's... resting

Beautiful plumage!

🥁

Heh

What is dead may never die!

Nice to talk to you. Have a nice day

Going back through, that presentation was *really* rough (I mean, more so than mine normally are). Everything is all kinds of out of order and badly explained. Sorry about that. Still, it's available now (or will be shortly): https://youtu.be/lprIwxyPY2E

/cc huhn

Thank you Sam. I think it is a good start and we need to evaluate more anyway. People should get attention to it and provide feedback now

you must cut it, it has 1 hour and 20 minutes and after 24 minutes it is over, after this it has black creen until 45 when it starts again from the beginning and other at 1:12 and then 8 minutes from your desctop ...

Sam:

Yah, no idea how that happened, I'm exporting a new one now.

another one who hates me now :))

What?

I was joking !!!

Okay; I guess I missed some context, IDGI. Thanks for the report anyways!

christian: No one hates you. If you feel refused it is maybe because often people tell to change things, but all here are volunteers and already try to but it usually does not come with that we or others delegate to each others. At least we ask to do things kindly or rather expect that changes are made by the one who claims them. So, this is why it is not so easy here to just drop a recommendation and expect it to be picked up right away. If you want to propose clear and feasible suggestion in communication you can do so in the github repository https://github.com/xsf/xmpp.org ✏

> I think if the point is "If all the XMPP devs worked together they could potentially make better (but fewer) projects" it's fundamentally likely to be correct.

At least 3

How many different client and server implementations does whatever the 1 company behind matrix is now named have now? ✏

So that's likely untrue then, I mean, anecdotally

There are too many Matrix clients! I'm just gonna stick to XMPP!

Problem with Matrix is you need to pick the right combination of client + server, otherwise nothing works.

emus, Hoo, hold the horses. :)))) I'm here because the group is called "Discussion", and I don't want to give orders to anyone. I'm trying to stimulate a conversation with pros and cons, of course, and I'm not at all hurt if I don't get my way. I would be the happiest person on earth if I could make clear and feasible suggestions. But I am a practical person and I try to steer you by bringing you feedback "from the front". If I talk to someone and try to convince them to use XMPP, and they openly tell me why they don't like it, and I hear the same thing from several people, then I have the gall to tell you here. What you make of it is then again a whole other thing

I sincerely hope that you do not want to go the matrix way. It is a bad way. It leads to nowhere. We should take IRC and NEWS as a model. Speaking of role models. But make everything a little more playful, clients need to be like toys. People like to play.

pidgins aren't real - they're government surveillance drones!

christian, nobody is taking offence to what you say, or trying to insult you. I think we all understand and agree with your underlying point - that it would be better if everyone put their collective efforts into a small number of projects - that would definitely produce nicer, more polished clients; the issue is not that people don't agree with this point.

The issue is one of motivation. People have limited free time and their motivation to work on their own projects (instead of someone else's) is that it matches their own vision. And since everyone's vision is different, if they were working on the same project then they would all be pulling in different directions - and that brings its own set of progress limiting issues.

The way we get people to work towards the same vision in our current world is with money; if you offer developers a pile of money then you can motivate them to work on your project and work towards your vision (they won't necessarily agree with that vision, but that's not their motivation.)

christian: ok there has been a website to organise with developers only actually and I agree that there is an issue to deal with resources. For example this why I try to improve coordination and communication between users, developers and knowledge

https://snikket.org/about/goals/ 🤔

<opinion> lack of polish/features is not an actual problem with XMPP, people use whatever client with whatever features anyone else craps out as long as their friends are there, it's 100% network effect and nothing else </opinion>

unless you actually believe people who say ridiculous things like "if only I could set a picture as a background on my chat I could switch everyone I know to XMPP" but I do not

moparisthebest, while that's largely true, if it's a pile of crap to begin with then people won't jump on for there to be any network effect

"I need a cryptocoin integrated with my messenger or I can't use it" - no one ever

My experience of that is that the honest truth is "if only I could set a picture as a background on my chat I could switch everyone I know to needing a different reason to switch".

Kev, exactly

theTedd, right, but I don't think that's been the case for quite a long time, no one is using pidgin on google talk anymore

moparisthebest, you forgot the infinite marketing budget to bootstrap that network effect

so that's all that is missing, no dev work required !

correct

we're not needed

anyone friends with some VC people who like throwing millions at something with no business model ?

I mean it worked for signal...

me neither...

Sorry, I don't think my parents can give out large enough zero-interest loans to get anywhere.

> anyone friends with some VC people who like throwing millions at something with no business model ? > I mean it worked for signal... > me neither... I think initially it was largely tax payer money that got them started

I don't know how to get that either though

there is a definite difference between a UI designed by a developer and one designed by a designer; we are quite lacking in "ooh, shiny" in terms of client interfaces (which is ultimately what users care about, after basic features are taken care of)

A short story: My wife and our son write each other every day A HEART. Sometimes a larger sometimes a smaller sometimes a kiss. They do not have much to say to each other, The son is in Berlin on assembly she in the Erzgebirge at home, and then they "ping" so on. Telegram and Whatsapp offer such games. And you can show emotion with it. Think about it. I can not convince you to switch to XMPP because you can not send a HEART.

♥

♥

And 60% of the time, that "missing feature" already exists.

yes but wit telegram it is huge and animated

But no, that doesn't fit the "XMPP is dead" narratcive

OH

so if we only add animated hearts then everyone will adopt xmpp

If only the heart was animated I could finally move everyone to Conversations

<large_animated_heart.gif>

Yaxim Emoji Hugification™®

oh and user-settable backgrounds

and cryptocoins

Most romantic show stopper ever.

</skepticism>

Meanwhile my 4 year-old is on XMPP, and successfully sends hearts *all the time*. She is unable to use WhatsApp, Signal or Telegram because she does not have a phone.

> Most romantic show stopper ever. :D

MattJ, I'm slowly using my 12 year-old to WhatsApp after years of XMPP usage 🙁

same MattJ , my 9 year old has been using XMPP for over half her life :P

You are cold-hearted developers and do not understand that there are many mothers/son and mothers/daughter relationships that can only be represented with a huge animated heart. :)))

Mmmmhm

christian, but XMPP and most clients fully support sending huge animated hearts

The older the kids get, the stronger the network effect.

(slowly using? loosing.)

(I mean he obviously always used WA to talk to WA-only contacts. But he's slowly starting to use WA to talk to *me*.)

christian, we are people too - we do understand; the thing you're saying is not possible is _entirely possible right now_

https://emojipedia-us.s3.amazonaws.com/source/skype/289/red-heart_2764-fe0f.png

Zash, that's not animated 😥

okay, okay I shut up :))))))))

theTedd: Clear proof that XMPP is in fact completely dead!

<sad_trombone_music.mp3>

Does Matrix have animated hearts?

Join XMPP - now with AnimatedHearts™ Technology!

That was apparently a PNG. How the heck is it animated?

a Gif?

Only the superior 80's technology in the form of GIF has such amazing cutting-edge features

https://burtrum.org/up/f8842ba4-5763-421b-afac-59a6a9a89f8a/heart.gif

👍

I have revived XMPP! is that acceptable christian ?

yes :)

problem solved, signal and whatsapp are no more!

Zash, there was MNG, but it didn't catch on; though there are some weird animated PNG formats in existence too

there are many graphiX guys who could paint emoticons .. we just have to create the plugins for it

is webp still what all the cool kids are using these days or have they moved on?

webp and hvec

was avif something gfix people scare their kids with or?

*hevc

quiet, or I'll compress you in a container

My wife sends the son a pair of glasses and a broom that means "Did you clean your room? Can I come to check" - he sends back a heart and a girl, which means "girlfriend is here I love you" :))))

Egyptians with their hieroglyphics are making a comeback? ;-)

Wojtek, +1 :))

> <opinion> lack of polish/features is not an actual problem with XMPP, people use whatever client with whatever features anyone else craps out as long as their friends are there, it's 100% network effect and nothing else </opinion> +1, but this also means that the only way to move people to xmpp is by the gentle force of not using silos.

to get a network effect, you first need a critical mass of users; you can't attract those users by network effect, so first you need something shiny

I don't have a iPhone so I can't say personally, but I have heard a lot of complaints about iOS clients.

It's rapidly improving over the past year or so

Monal, Siskin and now Snikket are under active development on iOS