Twenty years ago, they would have done that, but these days, someone would just write a replacement from scratch (quite possibly, proprietary), or everyone would abandon it, along with their own projects that depend on it.
Most open-source (and closed-source) APIs and SDKs are “Swiss Army knife” projects, with many functions that serve many purposes. Most API consumers use only a subset of the interface, so “rolling their own” is not that intimidating.
I think with all the security problems that we’re seeing, these days, we may be headed for some agency/consortium that validates dependencies (with all the myriad problems, therein).
We may be seeing more “semi-proprietary” stuff, soon.
Not necessarily a bad thing, as I think that anyone that makes money on software should probably pay for it.
As I have made my point before, if all I want is money I would spend my time working a paying job. But money is not everything isn't it?
But making things that other people depend on is an obligation; not just a vocation.
I liken writing an SDK/API/library to having children. Once they are there, they aren't really my "property" anymore, and I am under an obligation to maintain and support them, so I need to keep that in mind, when I publish them. I need to play the long game.
I've written a system that is used by thousands, around the world, and is the backbone infrastructure for a particular demographic. It is not hyperbole to say that lives depend on it. I worked on it for ten years, before transferring it.
The best thing that I ever did for that system was toss the keys to a new team, and walk away. Nowadays, I'm just the dorky old man that chips in his two cents' worth, from time to time.
I'm not particularly interested in getting into religious battles, which is fairly common in the [F]OSS community. I mostly code for the love of the craft. Delivering and supporting open-source projects helps me to have a purpose, but I also take my obligations quite seriously.
One of those obligations is to go to great lengths to deliver very high-quality software. I'm quite aware that it is not commercially feasible to write software that meets my personal Quality bar, so I give it away for free,
People have no idea that's it's mainly charity. And even then people demand, get angry, don't read any docs, etc, etc.
Bigtech sometimes helps by employing these guys. What is really needed is a service, which does the moderation and filtering. Kind of like oss communication management as a service. For example by github.
It also means supporting other people and stepping up. It’s so few and far that’ll pay value for value.
Outsourced moderation would probably end up something like Stack Overflow — useful things will be closed because non-subject-experts can't tell the difference between invalid questions and valid questions that look like invalid questions.
