collisions aside, are there any indexing and/or query performance benefits of using a uuid
vs a string
as some entity identifier? for example
{:player/id #uuid"cb7afbf9-95ca-4c5b-af42-b096342bae61"}
vs
{:player/id "a4st390xrvskm9ecm452jbn"}
I asked a similar question, https://github.com/zelark/nano-id/issues/12#issuecomment-625522643 The answer appears to be that there's not much of a difference, https://clojurians-log.clojureverse.org/datomic/2020-05-08/1588924450.320800 (currently getting a gateway timeout error)
that's perfect, thanks @tvaughan. go figure, i prefer UUIDs but i have to incorporate some string ids coming from a non-Datomic system. :man-shrugging: