I definitely understand the appeal of using SpiceDB as the source of truth for some of your business domain, as you'd potentially eliminate a set of problems surrounding state synchronization. You can certainly build client logic that would know how to walk the SpiceDB graph and issue a bulk delete (e.g.

ReadRelationships

+

DeleteRelationships

). The latter comes with preconditions https://buf.build/authzed/api/docs/main:authzed.api.v1#authzed.api.v1.DeleteRelationshipsRequest so I wonder if you build a statement that fails if a new element appears in the graph. If that was the case, you'd need to retry the operation again (similarly to how distributed databases like CockroachDB would do in case of overlapping transactions).

Hi guys a beginner in Spice DB here, I'm following some modeling examples, and I have one doubt: ()

// Write this rel to allow any org member to search for the group
group:test#searchers@organization:authzed#member

How does this type of relation translate to the Go cli for example ? Do you just nest the Relationship object ? Thanks in advance

when you write a relationship in the api you can specify an

optional_relation

for the subject: https://buf.build/authzed/api/docs/main:authzed.api.v1#authzed.api.v1.SubjectReference so when writing relationships like that, you would specify the optional_relation as

member

instead of leaving it blank

Thanks !!!! I'll give it a try 🙂

hey guys, new to SpiceDB here so not sure if this is a dumb question. As well as storing relationships, can I store additional information about my entities in SpiceDB? e.g. for example, if I'm storing a relationships about a document entity, can I use SpiceDB to store info about the document itself (e.g. date-created, contents, etc)

currently no, but you can see discussion about this requested feature right above this, as someone just also asked for it 🙂

Hej, the "go" version of https://docs.authzed.com/guides/first-app needs some updates, is this hosted on github too (so that I can send a PR)?

Obviouly I find it myself, right after asking 🙂

Sent PR for first fix.

is this option:

--datastore-revision-fuzzing-duration duration            amount of time to advertize stale revisions (default 5s)

the one that dictates how long a cache will hold onto a result for the purposes of

minimize_latency

requests?

no, that'd be revision quantization:

--datastore-revision-quantization-interval duration        boundary interval to which to round the quantized revision (default 5s)

but yeah revision fuzzing also influences it I believe

does either option provide a hard bound on how long a given revision would be available for a

minimize_latency

request?

minimize_latency

chooses the revision to use based on the quantization interval

do you mean how long

at_exact_snapshot

can be used for a revision, or...?

essentially trying to establish an upper bound on how long a revision will be used in the

minimize_latency

case

I know

at_exact_snapshot

has to do with the gc window

the quantization window +/- some padding

gotcha. where there aren't statements that can be made about that padding?

not a huge deal if so - just had someone asking me about it

the padding should be at most a second or two

but the point is: its not user defined nor exposed

if you use

minimize_latency

, you are saying "I don't mind being slightly out of date; I want you to minimize latency"

otherwise, use

at_least_as_fresh

with a ZedToken

yep, that makes sense. thanks!

is there a guarantee that every relationship applied to the database will show up in the WatchService?

at least once

Hi everyone! I just started messing around with SpiceDB, and even just simply joining discord answered the first question I had (with a little help from the trusty search function). So, thanks y'all 🙂

great!

I found "•" is not supported in resource id's like resource ID = ABC.XYZ any specific reason why,