[MichaelGG]

Basically the whole "version" UUIDs is a stupid fever dream. Any sane implementation should return 16 random bytes and be done. The fact that there's a spec for this and it's longer than 2 sentences is just wrong.

[mikeash]

I think they were just afraid of the birthday paradox. But I agree, just grabbing some random bytes really is the way to go.

Although, I'm a bit afraid of the birthday paradox, maybe we should use 20 bytes instead....

[MichaelGG]

The idea that a UUID should have a "version" and the random one uses 122 bits instead of 128 bits sorta makes ya wonder how concerned they were.

[jasonwatkinspdx]

Yeah, spot on. At least now in 2015 using 16 bytes for id's and hashes no longer seems like a lot of memory or storage.

It pains me to see people inventing weird id schemes for every app and api.