What Is Required to Build a Video Chat Messenger from Scratch?
-
I personally believe that Ubuntu Touch will go from 0 to 100 if it had a good, SECURE, video chat messenger.
In another thread, it was noted that the main problem with starting a video chat app is that it requires centralised servers and that costs money -- and would require staff/infrastructure, etc.
So if that's the case (and I do believe it to be so), the obvious way forward is to have people pay for the app. I do believe people will.
How about we start small. Let's get the MAXIMUM users we possibly can one one server. Then, launch a good marketing campaign to cause that to grow on a cloud infrastructure or balanced servers or something. I don't know, I'm not a sys admin; but surely we can develop something that works and will be scaleable later, without too much trouble?
Mark my words: If we get ONE private, end to end encrypted video chat messenger running, that will make all the difference in the world for UT.
UT is such a great and polished OS, it's just sad to see it fading out the back door.
I think if one video chat app doesn't get built, UT touch is over.
But .. If there's a will, there's a way.
So we just need to get cracking on building a Linux video app from scratch. OR, work very closely with another FOSS project that has most of it already done -- and fork it.
We can start with just ONE server. If we can get all that functional, we can branch out. People will pay. Even if it's via crowdfunding. Everyone who contributes to crowdfunding gets access to the first server. The second and future servers will be by subscription.
We can set up a website and social media for the effort. It will grow, believe me. There's a LOT of support for UT. It just needs a bit more of a push to get across the line.
Link to social media, crowdfunding.. create YouTube videos. Get some PROPER marketing behind the effort. I'm capable of building all that myself. I'll do it for free. I do marketing, Facebook ads, Google ads, websites, etc all day every day for a living. It's my job. I can help.
-
@mrlen
In the thread you're referring to, Florian said that the easiest way is to get someone motivated to implement videochat in Teleports.
The first step will be audio calls as the API already exists and servers too.
For video, IDK if the API has been released by Telegram, it's still in beta IIRC.The main problem is the man power.
There are changes for me professionally and I may step in depending on the next couple of weeks. But right now I'm not available. But I'm following the videochat topic closely for some time now. -
@mrlen we don't need to implement a new service to do video chat. We just need to support the existing ones, which are already working well on other platforms and are widely used, for example Signal and Telegram. So, to add the videochat feature in Axolotl and Teleports, the unofficial UT clients for the two services. I am a contributor to Axolotl, and to be able to do video chat, this issue in Morph need to be solved first: https://github.com/ubports/morph-browser/issues/399
-
Okies guys, I'll check out telegram more closely. Ubuntu Touch will be so awesome if we can at least get video chat running. I can think of about 10 people I could easily convince to carry a Nexus 5 or Oneplus One phone if it had video chat.
I have never been a telegram user, but I'll definitely check it out and become familiar with it.
-
@fla said in What Is Required to Build a Video Chat Messenger from Scratch?:
@mrlen we don't need to implement a new service to do video chat. We just need to support the existing ones, which are already working well on other platforms and are widely used, for example Signal and Telegram. So, to add the videochat feature in Axolotl and Teleports, the unofficial UT clients for the two services. I am a contributor to Axolotl, and to be able to do video chat, this issue in Morph need to be solved first: https://github.com/ubports/morph-browser/issues/399
Has any progress been made yet?
-
Sorry to say but I dont think so. The core team is working only on 20.04 upgrade since 16.04 is end of life by now. That will consume all resources for some months, we have a few hundred deb packages to be updated, tested and built, and then solve all the issues with the new root image.
So if not someone from the community dives into the lowlevel issues there is not progress soon.For TELEports: We know that we could solve it at least for outgoing calls. But there is again zero manpower, and we need to adapt 3 or 4 external libraries to UT first, before we can even start with the main app. Telegram has a very proprietary way to implement voice and video calls.
-
We had uMatricks that supported video and audio chat, but I think that used an old api or something? Idr but what I do know is that it doesnt work on modern devices, I think it never got updated to work on Ubuntu 16.04, but I wonder if some of that code is reusable?
-
@phoenixlandpirat said in What Is Required to Build a Video Chat Messenger from Scratch?:
We had uMatricks that supported video and audio chat, but I think that used an old api or something? Idr but what I do know is that it doesnt work on modern devices, I think it never got updated to work on Ubuntu 16.04, but I wonder if some of that code is reusable?
If I remember correctly it was because Canonical spend a lot of time creating a custom version of the browser that allowed doing this stuff on Mir. But this older browser was dropped at some point by Ubports as it was too much effort to maintain and qtwebengine promised to allow the same at some point on Wayland.
Edit: This is really the main issue I think. Most messengers get away by using an embedded browser and WebRTC to do video chat. Doing something like that from scratch is really hard.
