[tsimionescu]

I would argue that 2 is the much more important reason. A dependency is worth it if it saves you a lot of time building the same thing. For example, if you wanted to ship a general computing device, you should probably run Linux on it rather than building your own OS from scratch, because you'll save literal years of work. Even if Linux had stopped being maintained, this would still be the right call compared to building your own (of course, it would be even better to choose some other OS still being actively maintained, if any is available).

[whstl]

I wish we could have this discussion without going into exaggerated extremes like "Linux". The OS rarely gets into discussions of rewriting dependencies.

I recently had to rewrite a table component and it took about an hour with extra performance and it became faster-to-use and tailored to my company's needs. One hour, give or take! With measurable performance improvements, and it's simpler to use for our use cases, and it doesn't prevent anyone from using the old one.

Comparing one-hour or even one-day work with rewriting Linux is ridiculous fear-mongering.

[tsimionescu]

But this is exactly my point. Some dependencies don't actually save you any time, they're trivial to write yourself, so there is no point in getting them (perhaps the most absurd and infamous example of this being left-pad). Other dependencies are extremely hard to write, so finding a version someone else built, even if they are not maintaining it anymore, is worth it.