Modern XMPP project discussion - 2022-06-09

Hello, Did anyone think of using the Telegram Desktop* and making it compatible with XMPP, or making it XMPP-only client? Is there a discussion about it? Most of the UX is good, so I think it's worth a try. * and maybe the App too

Are they actually decoupled? If not i think it would be faster to just write a new client

Schimon Movim is inspired from this client for some things in the UI :)

One if the issues of not having a "telegram" style client in xmpp is that the part of the community that can program is very cli, barebones oriented so there hasnt been that much work on it

Yeah movim is the closest one. Maybe converse too

https://movim.eu/img/chatroom.png

https://external-content.duckduckgo.com/iu/?u=https%3A%2F%2F9to5linux.com%2Fwp-content%2Fuploads%2F2020%2F07%2Ftelegram22-1-1.jpg&f=1&nofb=1

edhelas, thank you!

Is it web based-only or also available for desktop?

It's a PWA :)

So you can install it as an "App"

PWA?

Progressive Web App

edhelas: do we have app now?

Damnit 😬

I don't get what's unclear about what I said

Schimon, unfortunately it's a lot of work to just "adapt" a client like that. The Telegram client will make a lot of assumptions about how Telegram works, that won't be suitable for XMPP.

I'm not saying it couldn't be done, if someone wanted to work on it. But it wouldn't be a small project. ✏

I would fund it

Not really interested in learning android dev

> MattJ: > 2022-06-09 03:30 (GMT+03:00) > Schimon, unfortunately it's a lot of work to just "adapt" a client like that. The Telegram client will make a lot of assumptions about how Telegram works, that won't be suitable for XMPP. Yes, _only the initial work_. The rest would be easy. I only want the UI/UX. I think such client will bring a substantial amout of new users to XMPP IMHO

Schimon, feel free to work on that, but be aware of the difficulties of taking a codebase made for a very different protocol and “just” changing the underlying protocol.

I understand. This is a collaborate project.

What is stopping you from getting started with that?

> What is stopping you from getting started with that? Because: * I don't think I'm qualified to begin this task. * and also job and other chores.

Schimon, who would be more qualified to start a project than someone who wants to do it? :)

As a question because i am vaguely interested in something like that: Lets say you removed the telegram part. What library would you replace it with? For xmpp support. Like i am looking maybe something starting in kotlin maybe. Since it has more support and such. Would conversations be able to be split back and front and "glue" it to a UI?

Or is there some other library

Besides the tigase kotlin library which i am definetily interested in

Link Mauve, I know. I know. Same for feature request and bug fixes. I think it was you who told me that almost a decade ago.

I just dont want to reimplement the whole world. Just the front :)

jajaja

Very possible. ^^

I don’t remember everything I said a decade ago though. :D

msavoritias, I undersatand. The thing is that Telegram, with all the resources they have, they essetially have brought us an open source UI/UX product and examples* so the Telegram Desktop and App are great opportunity for XMPP community to lure in more users, using Telegram already-made UI/UX. * Regardless the fact that it's not a good network, IMHO

Zash, however, the best I think I can do at this stage is to attempt to organize a group of people that either interested to participate or to be updated with the progress.

> I don’t remember everything I said a decade ago though. :D I tend to remember only the good things in people.

Me too!

Schimon: true but: 1. Its java. 2. If i can just plug a library its gonna be faster probably

> Me too! I think that's a wisedom I learned from you too ;)

> Schimon: true but: > 1. Its java. > 2. If i can just plug a library its gonna be faster probably Sure. Look, all I want is the UI skeleton. All the rest doesn't matter

If I were you, I’d start by looking at the code of that client, maybe contribute a few bits here and there to get familiar with the codebase.

Link Mauve: for telegram you mean?

It will also help you understand which parts you can reuse (or want to reuse), and which parts you want to burn with fire.

msavoritias, yes.

Or anything really. :)

Good idea

Last note on Telegram: * They offer nothing new. * They seem to attempt to overdo WhatsApp. * They seem to attempt to be a replacement to Facebook * To offer themselves as an alternative, they must make giveaways (i.e. open-source client and special features) that they wouldn't do otherwise (i.e. if they were the biggest in the market). * They do have a big back of money. These are the main reasons why I'm critical for the UI/UX and why I find it so valuable.

Another option is to extract the design bits you find interesting, and implement them in another existing client.

It was how Conversations got first created fyi, Daniel pretty much 1:1 copied Hangouts’s current design, and voilà it became Conversations.

Of course, design is an iterative and ever-changing process, so even if Hangouts was the pinnacle of design back then, the landscape evolved a lot since then.

> Link Mauve: > 2022-06-09 08:16 (GMT+03:00) > It was how Conversations got first created fyi, Daniel pretty much 1:1 copied Hangouts’s current design, and voilà it became Conversations. He's not an example. Everytjing he touches turns into gold :D

Anyone else could pick an application and copy its UI.

I did that with Discord → Converse in one afternoon with great success.

Just changing the colours and hiding some information.

Suddenly people didn’t call it “eww, not modern”.

Link Mauve: but what would you use as a library for android? Any ideas?

msavoritias, I’m not well-versed in Android development, sorry.

I don’t even use that OS.

No worries me neither

Just saw the need for a client with stickers and a modern ui and everything. And i am wondering how to go about it

> Link Mauve: > 2022-06-09 08:17 (GMT+03:00) > Of course, design is an iterative and ever-changing process, so even if Hangouts was the pinnacle of design back then, the landscape evolved a lot since then. I agree, but not sure. I recsll having arguments with Yann about Gajim's UI, but it was difficult to persuade him, unless I've a desiger(s) that will explain my opinion. However, current Telegram UI, sjould serve us for a longer period of time, namely because it was made by a company that looks into manipulating isers into using it's network. Contrsry to them, we, in XMPP, introduce freedom.

msavoritias, stickers on XMPP is mostly not a UI problem, but a protocol problem.

I recently implemented a sticker selector for poezio for instance: https://bouah.net/2022/04/updates-from-the-poezio-ecosystem/

> Link Mauve: > 2022-06-09 08:21 (GMT+03:00) > I don’t even use that OS. What do you use? Please advise, as I want to buy a new device with XMPP support

But it’s not any better than a file picker, stickers on other protocols come in packs, which you can either buy or copy from your contacts, and that’s the missing part.

Link Mauve: i know about poezio yeah :D

> Link Mauve: > 2022-06-09 08:23 (GMT+03:00) > msavoritias, stickers on XMPP is mostly not a UI problem, but a protocol problem. FYI, I didn't refer to stickers, nor do I think they're useful

XEP-0449 is a step in that direction, but still lacking somewhat.

Schimon, I replied to msavoritias here.

I know

I personally think they’re useful to convey emotions.

I still love you

D’aww. <3

;) <3

Jajaja

https://linkmauve.fr/file_share/-4YTkeuWZNYPJmqixCyCV0G_/UI_EmotionIcon21.png

It's cute

Schimon: so telegram is not worth it a according to jdev channel

I tend to agree

Im just conflicted if i go with java or kotlin

Kotlin