Skip to content

Comment on Ask HN: What kind of information do you look for before learning/using new tech?

Comments

Basically, I'm interested in knowing that adopting this tech will make my life happier and less complicated.

* Look at the open issues -- Is it full of buggy edge cases and "3d chess"?

* Quality and maturity -- Stuff works, as documented, under all possible conditions. Maintaining existing features / "tech debt" is prioritized over new features.

* What problems does it solve? What problems might it introduce? Does it actually solve the problems it's claimed to solve?

* Is it designed in a way that will totally break a possible future use case?

* How much of a "custom ecosystem" is there (bad), and how much of it is "open standards based" interfaces (good)

* Is it getting more or less complicated to use it over time? Can you learn it as a "simple cognitive model" or do you learn it as a "huge collection of random facts" e.g. "a cognitive edge case LUT"

* Documentation must be okay (no elasticsearch)

* History and sustainability -- How long has it been around, is it updated regularly (regularly != often), do updates break stuff (no pipenv)

AboutSource Built by g1lg1l

Hackerly is an independent reader for Hacker News, built on the public HN API. Not affiliated with Y Combinator.