2 ms·
this is not a new problem; peter deutsch was very annoyed in the 90s about linux distributions distributing outdated, and often modified, copies of ghostscript,
by kragen 2y ago
this is not a new problem; peter deutsch was very annoyed in the 90s about linux distributions distributing outdated, and often modified, copies of ghostscript, whose users would then complain to him about their bugs. compounding the problem in his case: virtually nobody else was capable of contributing third-party code that was up to his quality standards; for context, he'd written i think the second or third implementation of lisp, in assembly language, when he was 15, some 30 years earlier, and hadn't stopped honing his craft since then, for example inventing jit compilers
ultimately i think the answer is to limit your interactions with the takers; some of them may become makers later, often of different free software than yours†, but most of them won't. they may provide useful feedback (bug reports, feature requests, etc.) but most of them will not. what's important is preventing them from overrunning spaces where the makers are collaborating (and especially harassing makers into quitting), and maybe to give them a path toward growing into makers, if they are so inclined. dries's system seems like a gentle, probably sufficient way to do that
______
† the contributors to vim mostly don't contribute code or bug reports to gcc, and the gcc maintainers mostly don't contribute code or bug reports to vim, but they each benefit from the others' work. similarly, many linux distributors eventually became important indirect contributors to ghostscript development, even though at first peter wasn't using linux, and i think even today very little of the code is contributed by outsiders