4 ms·

Re [5]'s asssertion under "blunders" of the diminish usecases post sql/pgq, what do you think of [something] like Quine? https://github.com/thatdot/quine https

Re [5]'s asssertion under "blunders" of the diminish usecases post sql/pgq, what do you think of [something] like Quine? https://github.com/thatdot/quine https://github.com/thatdot/quine Their claim to fame is progressive incremental computation - each node is an actor responding to events -- and I'm not sure how a relational db could do that and match the latencies. That usecase is pretty much pattern matching and forensics and stuff like that. https://docs.quine.io/core-concepts/architecture.html https://docs.quine.io/core-concepts/architecture.html

szarnyasg 4y ago
Quine seems to be an interesting project but I'm not too familiar with it. Its main feature, evaluating complex multi-way join queries on incoming streaming data, exists in the relational world as the Materialize database which leverages differential dataflow for computation. Quine uses Cypher so expressing path queries can be done with the concise Kleene-star syntax, e.g. (p1:Person)-[:knows*]-(p2:Person). Materalize is getting support for WITH RECURSIVE and WITH MUTUALLY RECURSIVE (their own SQL extension that fixes some issues of WITH RECURSIVE): https://github.com/MaterializeInc/materialize/issues/11176 https://github.com/MaterializeInc/materialize/issues/11176
szarnyasg 4y ago
It's also worth to mention the idea of WITH ITERATIVE. This is a proposed SQL syntax extension that allows formulating iterative computations. For many problems (including graph traversals), this makes queries easier to write and execute. > We derive new iterative CTE variants from the simple loop-based, operational semantics of SQL:1999’s WITH RECURSIVE. https://www.cidrdb.org/cidr2023/papers/p14-hirn.pdf https://www.cidrdb.org/cidr2023/papers/p14-hirn.pdf https://www.cidrdb.org/cidr2023/slides/p14-hirn-slides.pdf https://www.cidrdb.org/cidr2023/slides/p14-hirn-slides.pdf https://www.youtube.com/watch?v=ZDtSGAeYKU0 https://www.youtube.com/watch?v=ZDtSGAeYKU0

szarnyasg 4y ago

Quine seems to be an interesting project but I'm not too familiar with it. Its main feature, evaluating complex multi-way join queries on incoming streaming data, exists in the relational world as the Materialize database which leverages differential dataflow for computation. Quine uses Cypher so expressing path queries can be done with the concise Kleene-star syntax, e.g. (p1:Person)-[:knows*]-(p2:Person). Materalize is getting support for WITH RECURSIVE and WITH MUTUALLY RECURSIVE (their own SQL extension that fixes some issues of WITH RECURSIVE): https://github.com/MaterializeInc/materialize/issues/11176 https://github.com/MaterializeInc/materialize/issues/11176

szarnyasg 4y ago

It's also worth to mention the idea of WITH ITERATIVE. This is a proposed SQL syntax extension that allows formulating iterative computations. For many problems (including graph traversals), this makes queries easier to write and execute. > We derive new iterative CTE variants from the simple loop-based, operational semantics of SQL:1999’s WITH RECURSIVE. https://www.cidrdb.org/cidr2023/papers/p14-hirn.pdf https://www.cidrdb.org/cidr2023/papers/p14-hirn.pdf https://www.cidrdb.org/cidr2023/slides/p14-hirn-slides.pdf https://www.cidrdb.org/cidr2023/slides/p14-hirn-slides.pdf https://www.youtube.com/watch?v=ZDtSGAeYKU0 https://www.youtube.com/watch?v=ZDtSGAeYKU0