jay

> Is there a reason why this can't be as simple as https://github.com/dream3d-ai/lance/pull/1/files @tonyf If there is only one add_column, your implementation seems to be fine. However, for other tasks...

> > > Is there a reason why this can't be as simple as https://github.com/dream3d-ai/lance/pull/1/files > > > > > > @tonyf If there is only one add_column, your implementation...

@westonpace The comment has been fixed. Please merge it.

Well, I am interested in participating in this development. I want to confirm if there is a design document already.

> FWIW, since I first brought this up, I have actually implemented a version of this within an internal system. Unfortunately this means I cannot open source it. However I...

> [@Jay-ju](https://github.com/Jay-ju) Yes, it's a performance thing, none of this is necessary if the lance dataset exists locally, but when the lance dataset is remote and you must make network...

> Thanks [@Xuanwo](https://github.com/Xuanwo)! I believe [@plotor](https://github.com/plotor) is putting together some initial benchmarks for us to work off of for prioritization of these features @Xuanwo @jaychia I'm still dealing with benchmarking....

update： https://github.com/Eventual-Inc/Daft/pull/4874 I've added some benchmarks here. Currently, they are all very simple test cases, and more will be added later. moreover, the pushdown logic hasn't been merged yet, so...

### More Roadmap Tracking | Status | Issue ID | Title | Technical Scope | |----------------------|----------|---------------------------------------------------------------------------------------------------|----------------------------------| | :hammer: In Progress | #4842 | **Refactor Lance Integration: Migrate from Python SDK...

> Sounds interesting, would def be something we would be looking help for! @srilman I really want to hear your suggestions here. This is also what I've been doing recently.