Modernizing the OWL implementation stack

Would there be any interest in building a modern OWL stack, based around
Rust and DuckDB/parquet?

The OWLAPI has been a fantastic boon for the community for the last 20
years, but I think it's time to look beyond this. It's slow for even
mid-size ontologies. Java/JVM made a lot of sense at the time but it's
obviously not what we'd pick now.

There is already a mature reference implementation in Rust:
https://github.com/phillord/horned-owl

It would be great to see more community uptake of this.

Rust makes sense for obvious reasons, and there are already python
bindings: https://github.com/ontology-tools/py-horned-owl/, and reasoners
with the most useful OWL profiles <https://github.com/INCATools/whelk-rs>
in Rust.

For CRUD operations, SQL makes more sense than SPARQL. SPARQL is too low
level for OWL, and all embedded SPARQL engines are too slow. Having a
standard SQL/Parquet mapping and Rust implementation makes a lot of sense
for performant declarative operations, and integration into modern data
lakehouse systems.

In the past, the complexity of OWL was a barrier to implementation: you
would need to allocate many hours of an expert software engineer for even
small applications. If anyone has seriously used Claude Code + Opus 4.5
with a Max plan in the last 2 months, you'll realize the implementation
costs have dropped to close to zero, provided specs are good and there are
good implementations to draw from (they are for OWL).

If this resonates and you can contribute requirements/vibe coding, respond
on the list or DM me. I'll try and coordinate activities. We also have an
OBO slack with relevant channels we can use (
https://join.slack.com/t/obo-communitygroup/shared_invite/zt-3mz8ccwps-4UAr6R_obUvWSiGqs_z41A
)