It's worth considering whether that's really a problem. With AI, It's easier to reinvent the wheel 1000 times then get 1000 people together and agree on requirements (which most people would then modify).

▲

acdha 2 days ago | parent | next [-]

There’s some truth to that but I think it’s a huge problem for anything which isn’t isolated to a single team because you’re giving up a shared consensus reality. The article mentions things like copies of specs or datasets becoming stale or missing updates, and I’d also expect problems for many analytic processes — e.g. if we don’t calculate daily active users the same way, we might make conflicting decisions and both be convinced that the other team is wrong.

▲

andriy_koval 2 days ago | parent | prev | next [-]

The real problem is maintaining, troubleshooting, debugging, extending 1000 wheels instead of one.

	▲	belval 2 days ago \| parent [-]
		Bah the other wheels usually die of natural causes if you wait a bit. I am a big believer in Amazon's "1 > 2 > 0" for "one perfect solution is better than two, but both are much better than no solutions". It's also misunderstanding how Amazon works. If you want to move fast, you can't wait for an SVP 6 levels above you to approve every effort. Instead you build something and then you run it up the chain to have it be adopted at team/org-level.

▲

cdrnsf 2 days ago | parent | prev [-]

I'd imagine there's a wide gap in the quality, security and reliability of the two approaches.