There is a very large corpus of FLOSS software out there serving everything from individual itches to whole industries. Any project that is important to someone’s bottom line is likely to have paid developers working on it but often alongside hobbyists.

The project I predominately work on is about 90% paid developers but from lots of different companies and organisations. Practically though the developers don’t care about the affiliation of the other developers they work with but the ideas and patches they bring to the project.

For portability Vulkan is the way (it also gets you GPU compute for free without needing vendor libraries). That said the ruttabaga encapsulation is useful for things like Wayland over virtio-gpu which is useful for some use cases.

I should note for even closer to native performance you want virtio-gpu with native context. Patches for that are currently being reviewed on the mailing list: https://patchew.org/QEMU/20241024233355.136867-1-dmitry.osipenko@collabora.com/

FEX redirects graphics library calls to their native equivalents. This substantially reduces the amount of translated code you need to execute.

slp did a nice demo at KVM Forum last month. https://kvm-forum.qemu.org/2024/The_many_faces_of_virtio-gpu_F4XtKDi.pdf and https://youtu.be/10Ztv0UI5I0?si=19KPcA6wGbXM3IsS

Quite. Go to the big services that know how to moderate and maintain (and importantly pay for) a public square. But also encourage the interesting ones enable federation for wider coverage.

There are some advantages to algorithms for discovery - it’s certainly is more user friendly. It’s just a shame they tend to enshitify or become toxic. Bluesky seem to offer an API of sorts to plug in feeds you create. Perhaps open algorithms are more accountable?