Matrix proponents would do well to learn from what is happening to Firefox and Mozilla.
Element do 90% of the development in the Matrix ecosystem - including the only server that can run at scale (I wont say "work") and the only homeserver designed to replace it (not production ready).
If Element were to come on hard times (as they have in the past, with near disastrous results), Matrix will simply fold.
@kline What do you suggest as alternative for a potentially struggling company? No company?
Never used Onlyfans but that "only" implies it's 100% commercial enterprise while open-source is freely available with *voluntary* donations.
What I'm just trying to highlight - after spending like 20 years in open-source community as user and developer - is that donations do increase chances of an open-source project surviving in long-term, simply because people behind it have lives too, and the most user-visible work is usually the most boring too.
So I have seen tons of really brilliant projects that were abandoned simply because the developer wasn't able to continue working like a slave to satisfy continuous stream of demands from bastard users who didn't give a fuck to even say "thanks", not to mention donate $2 pm to a project they use and benefit from on daily basis.
Therefore, whenever I see a project I use and like and don't want it to disappear, I just start donating to it, using any means available.
@kravietz @anornymorse @arjen I think what would be better is to build a community of developers who can scratch their own itch and spread the load of bastard users.
Predicating your project on getting big enough to make cash quick is a bad strategy.
But my whole point is *not* about making profit.
People are motivated by different things, some like the community that forms around a project, some like the experience and can't do community. You can't tell them "oh just build a community".
Some project do commercialize and close their code completely, usually as a desperate, last resort when their work is being used to make profit by bastard companies, as it happened with Grsecurity or Nessus for example.
A good case study is OpenSSL project which was developed initially by one guy (EAY) then taken over by others and, having a very permissive license, was immediately commercialised by some companies, and used by literally everyone on the Internet, while the developers were funded in a typical trickle-down model like "oh we earned $1m on your code, but our shareholders won't approve $1k for you if you publish your code for free, but hey, here's $100 donation!"
@kravietz @anornymorse @arjen this whole discussion and the original post is about building resilience of software projects. Being profitable but still super centralised on one individual or group is the issue. Some people try and fix this with money (as someone is less likely to leave when they're being paid), but it's never going to get fixed unless you dampen the bus factor.
@kravietz @anornymorse @arjen specifically, the bus factor of Matrix right now is exactly 1x for-profit commercial entity, Element.
Everyone throwing $2/mo at element doesn't fix this.
You can consider a situation where Element - as a commercial, for-profit entity - is going to do something morally dubious, and core developers consider leaving and taking the GPL code away. The situation is significantly different if they know there's 1000 users who pay $2/mo which means they at least don't have to pay for infrastructure from their own pockets.
@kravietz @anornymorse @arjen matrix has never, ever had grass roots, non commercial development, but I suppose a patreon will just make that happen
You're raising absolutely valid point about project management but it's a bit of a separate issue.
A project like Matrix or OpenSSL is not only code but also issue management, project management, security, documentation, user support, website, business logic maintenance & updates etc etc. Projects like Nmap, Nessus or Suricata also require very frequent rule updates. All that tends to be extremely boring and routine work.
Volunteers tend to lose interest in these.
@kravietz @anornymorse @arjen alright, but I think that these issues are completely unrelated to Matrix being centralised and are probably best discussed in a thread of their own.
I don't think that the solution proposed here ( https://social.privacytools.io/@kravietz/104680879548972793 ) is a practical one that actually addresses the issue this thread is about.
So the OpenSSL code gradually went into a legacy nightmare and the wake up call was series of terrible vulnerabilities back in 2010's. So a number of "phew, you guys write crap code, let me show you how it's done" projects was spun off, some of them funded by corporations, who previously didn't give a fuck about the code because - you guessed it, they could have it for free.
