Their claim is that regular Synapse will also get performance improvements, just only the ones for small instances, which might have been a credible claim if it weren't for the fact that I *am* on a small instance and it performs like shit just because I'm in a lot of rooms, which incidentally has always been the metric for "big instance" with Matrix, so y'know
@hans I'm sure they want that. But that's of no concern to the broader ecosystem, and this kind of thing is precisely what wasn't supposed to happen. They have been warned many times that their business and developmental model is unsustainable (with assistance offered), and if they insist on continuing it anyway, then I have little sympathy for the decisions that they feel they need to eventually make.
The server in question is running with workers (couldn't keep up in monolithic mode) but I do not administrate it.
@hans "What I understand is that if Element can reliably turn a profit every year, development of Matrix (and the broader ecosystem as a result) is secure."
That is how Element frames it. In practice, this is quite far from the truth - Element's stewardship has caused a *large* amount of volunteers to burn out or bounce off. I have personally spoken to many of them.
Volunteers which would have been able to provide that ecosystem security, which Element is infamously failing to provide, with their continuous technical and stewardship failures, and the increasing closedness of what are functionally the reference implementations.
There was never a lack of volunteer interest or labour availability. The problem has always been that Element did not provide the space for it to grow, and insisted on a highly centralized development model that only works with large amounts of funding.
@hans Or to put it more bluntly: Element has found a solution to a problem that Element has created.
Are you running your server in monolithic mode or with workers?