4 ms·

Isn’t it simpler to use sequence keys then?

Cthulhu_ 3y ago
Yes, but only on single machines, UUID and co are intended for distributed systems. Although now I wonder if / how UUID v7 can do sequential keys on distributed systems. Mind you, on those systems "close enough" will probably be good enough, and sorting will be done by date instead of incremental ID.
- moffkalast 3y ago
  You could also keep a global incremental index for that, assuming there's some authoritative server that has to be polled sequentially to get them. Probably too much overhead when conflicting UUIDs are a few orders of magnitude less likely than the clients crashing from some random bug though.
- EGreg 3y ago
  So just prefix the node ID which is assigned when the node joins a swarm
  - Timon3 3y ago
    How is that easier or better? Now you're letting your infrastructure details bleed into your database.
    - EGreg 3y ago
      They should bleed into your database. The database is part of your infrastructure and it contains details like where to find the replicas and so on
      - Timon3 3y ago
        Why? Tight coupling doesn't make things easier.
        EGreg 3y ago
        You’re complaining that the infrastructure details are bleeding into the intrastructure implementation?
        Timon3 3y ago
        Infrastructure details are bleeding into non-infrastructure implementation. Your database content is not an infrastructure implementation.
        EGreg 3y ago
        the content is not, but it can use a database function, such as CURRENT_TIMESTAMP or RANDOM() can it not?
        Timon3 3y ago
        I'm not sure what you're getting at. How are those connected to the topic? Those functions aren't infrastructure implementations either.
michaelt 3y ago
Ah but you see, UUIDs are web scale. And by web scale, I mean they're too big to exchange by any offline channel.
- Tade0 3y ago
  It's 128 bits vs 64 bits - not really that much of a difference.
- SV_BubbleTime 3y ago
  FWIW… we’re using UUIDv7s over BLE.
  - _flux 3y ago
    Should you, though? BLE itself uses UUIDs, but the ones it uses are in a shortened format just because the BLE frames are small and the protocol is not too fast. Granted I've even used JSON over GATT so maybe I should keep my mouth shut :).
    - SV_BubbleTime 3y ago
      It’s for an auth use. And at least we’re using CBOR. But I get your point.

Yes, but only on single machines, UUID and co are intended for distributed systems. Although now I wonder if / how UUID v7 can do sequential keys on distributed systems. Mind you, on those systems "close enough" will probably be good enough, and sorting will be done by date instead of incremental ID.

moffkalast 3y ago
You could also keep a global incremental index for that, assuming there's some authoritative server that has to be polled sequentially to get them. Probably too much overhead when conflicting UUIDs are a few orders of magnitude less likely than the clients crashing from some random bug though.
EGreg 3y ago
So just prefix the node ID which is assigned when the node joins a swarm
- Timon3 3y ago
  How is that easier or better? Now you're letting your infrastructure details bleed into your database.
  - EGreg 3y ago
    They should bleed into your database. The database is part of your infrastructure and it contains details like where to find the replicas and so on
    - Timon3 3y ago
      Why? Tight coupling doesn't make things easier.
      - EGreg 3y ago
        You’re complaining that the infrastructure details are bleeding into the intrastructure implementation?
        Timon3 3y ago
        Infrastructure details are bleeding into non-infrastructure implementation. Your database content is not an infrastructure implementation.
        EGreg 3y ago
        the content is not, but it can use a database function, such as CURRENT_TIMESTAMP or RANDOM() can it not?
        Timon3 3y ago
        I'm not sure what you're getting at. How are those connected to the topic? Those functions aren't infrastructure implementations either.

michaelt 3y ago

Ah but you see, UUIDs are web scale. And by web scale, I mean they're too big to exchange by any offline channel.

Tade0 3y ago
It's 128 bits vs 64 bits - not really that much of a difference.
SV_BubbleTime 3y ago
FWIW… we’re using UUIDv7s over BLE.
- _flux 3y ago
  Should you, though? BLE itself uses UUIDs, but the ones it uses are in a shortened format just because the BLE frames are small and the protocol is not too fast. Granted I've even used JSON over GATT so maybe I should keep my mouth shut :).
  - SV_BubbleTime 3y ago
    It’s for an auth use. And at least we’re using CBOR. But I get your point.