## Intro

Mass Psychology And Open Source
Showing: Jitsi

by

Christoph Lohmann <20h@r-36.net>

## What Is This About?

* When Corona happened something new came.

* People had to fast adapt to home office.
	* Laptops, even really old ones, were bought for much money.
	* Webcams were sold out.
	* Video conferencing was needed asap.

* School systems had to adapt to online courses.

* Universities had to adapt to online lectures.

* Everyone wanting to meet someone had to adapt too.

## The Cloud Is A Lie

* This brought Microsoft Teams to a halt and outage.
	* The cloud does not scale.
	* It is all a big lie.

* Microsoft Azure was full
	* It is all a big lie.

* Google Cloud was full
	* It is all a big lie.

## Data Scandals Happened

* People were first moving towards the commercial alternatives:
	* Microsoft Teams
	* Zoom

* Zoom failed miserably, had numerous scandals.

* Microsoft Teams is pure lock-in and only corporate slave holders still
  use it.

## The Jitsi Side

* Volunteers set up jitsi instances to allow people video conferencing.

* This is what the standard setup of Jitsi does.
	* Open some video conferencing.

	http://some.jitsi.instance.com/someroom

	* After closing, the room is gone.

## The Teaching Side

* Digital learning got a big push too.

* Here the open source solution was
	* https://bigbluebutton.org/
	* Had a big exploit too.

## Summary For Now

* In a big new change, open source can win, when it is ready.
	* Low cost.
	* Easily deployable.
	* Just works.

Do you see paralells to the bitreich philosophy?

## Getting Deeper

* The whole web conferencing stuff is mostly basted on WebRTC.
	* And ugly apps. But this is not part of the talk.

* WebRTC is an extension for browsers so clients can talk to eachother.

* STUN is used to get connection parameters between clients.

* SIP is most of the times used for signalling, but over websocket.

* For media transport (S)RTP is used.

* All codec are allowed for audio.
	* OPUS and G.711 recommended.

* Inbetween a broker is used.

## Bloat in Jitsi

* Besides just working, jitsi is an ugly beast.

* Based on Java.

* Much encoding/decoding is done in Javascript on client-side.

* Runs for internal stuff some XMPP server written in lua.

* Requires the whole VPS or server to be just for jitsi due to dependencies.

* Is only easy in docker-like container thinking.

## Can We Add Bitreich?

* WebRTC can be used to get web browsers to connect to something. 

* WebRTC can be used to get web browsers to connect to eachother.

* All WebRTC browsers can talk to eachother over a raw bytestream.

## A Simple Idea

* Compile ffmpeg into webasm.

* Run some simple VM in each webbrowser.

* Use WebRTC as pipes for some emulated shell.

* Allow easy streaming and scripting.

* Make a platform easily modifiable for video and audio conferencing.

* Take over the world.

* Compare a simple VM image to the average JS library.

## Mass Psychology

* Adaptation of things normally is not based on ideology.
	* The masses do not care about saving the world.
	* The masses do not care about dolphins.
	* The masses do not care about eco-friendly.

* Adaptation is based on a practical need.
	* In case of crisis even faster.

## Mass Psychology And Social Networks

* When will be the moment for moving Facebook/Twitter to Mastodon or something
  open, federated?

Hint:
	* Twitter is making a deficit besides having a huge user growth.
	* Facebook is not really used by the youth.

## Mass Psychology And Mobile Phones

* At the conference we had Maemo Leste.
	* Why isn't it in the mass market?
	* Why is the ugly amateurish Android used?
		* Whoever said to use Java for display was a fool.

* Can the Android bloat hell help?
	* Google failed at Android versioning and deployment.
	* The build system is still python2.
	* Modularisation of Android is impossible.
	* Google does not really care.

* What if users ever realize how they are spied on?
	* In Germany the data perverts show no mercy anymore.
		* Will people wake up?

## Mass Psychology And Open Source

How can Open Source win in such situations?

* It is cheap.
* It just works.
* It has no (obvious) cost.
* It is stable in the long run.

How can Open Source lose in such situation?

* For extended features there is no support.
	* Corporations require this.
* If something is new, corporations evolve faster in development.
	* Open Source wins in the long run by being stable.

## Dialectics And Mass Psychology

Change is always a chance and can be used to the positive and negative.

* Use it to the positive!

## Questions?

Do you have any questions?

## Thanks

Thank you for listening.

Comments please send them to:

	Christoph Lohmann <20h@r-36.net>