I was going to advertise it more at some point, but I just announced it here and already we are saturated with new OpenSSL tests that the interpreter cannot digest yet: https://www.mail-archive.com/openssl-dev@openssl.org/msg3930...

My personal trajectory would make no sense if tis-interpreter ended up as an advertisement. It may end up helping convince the general population that it's possible to detect all practical undefined behavior in a useful C program, by giving everyone access to a piece of software that does it along one execution. Then TrustInSoft will only be asking its more dubitative prospects that its other products can do it for all possible input vectors, instead of just one. I'm not even claiming this is an unintended consequence. But tis-interpreter will be a self-contained, useful product for the largest number. You have my word.

Linux has won everywhere except the desktop, and that's for many reasons - one of which is Windows holding on with a vicelike grip. The first time Windows had actual desktop competition from Linux was 2007, with netbooks - they dropped the price of an XP licence to $5 or even $0 just to keep Linux the hell off the things.

So Linux can do most of the job, and it's other factors keeping it off rather than anything that could be fixed by a concerted developer effort on a single distro.

(Some serious backing to Wine development would possibly be useful here. But again, the forces aren't susceptible to pure development.)

Furthermore:

* there are different distros because people can.

* there are different distros because there are different use cases.

* Red Hat and Ubuntu are already huge and neither is going away any time soon.