You signed in with another tab or window. Reload to refresh your session.You signed out in another tab or window. Reload to refresh your session.You switched accounts on another tab or window. Reload to refresh your session.Dismiss alert
RFC 4122, Section 4.1.4 specifies the first 60 bits of a UUID to hold a time stamp, and Section 4.5 specifies the remaining bits in the absence of a network address.
Proposing an alternative (and non-standardized) ID system will only fragment usage.
The text was updated successfully, but these errors were encountered:
Well, for one, it uses 40 bits for the timestamp as opposed to 60 bits; you pointed that out yourself. Another difference is that it doesn't bother with a MAC address (even though a UUIDv1 doesn't require one either). Yet another difference is it's 'notation'; where UUID's use hex, ULID's use the (slightly more efficient) Base32 notation (with or without hyphens is not yet clear).
I agree with your 'yet another standard' sentiment, but on the other hand we shouldn't be afraid to introduce better / different standards if they improve upon other standards or improve something in general.
How is this proposal different from UUIDv1?
RFC 4122, Section 4.1.4 specifies the first 60 bits of a UUID to hold a time stamp, and Section 4.5 specifies the remaining bits in the absence of a network address.
Proposing an alternative (and non-standardized) ID system will only fragment usage.
The text was updated successfully, but these errors were encountered: