Russ Tedrake

I'd prefer to not resort to sampling unless we need to. I've opened https://robotlocomotion.slack.com/archives/C0JH2GSUR/p1724321626756439 to discuss.

Just to be clear, you do not need to use `SceneGraph` to add the polytopes to a meshcat animation. `MeshcatVisualizer` definitely makes that workflow more convenient, but it's not necessary....

Tagging https://github.com/RobotLocomotion/drake/issues/15263 for the connection to the in-memory meshes request.

@jwnimmer-tri -- could I ask you to weigh in here about whether it is currently safe to parallelize convex optimization solves? In #21770, it seems like you were ok with...

I'll close this due to lack of activity. Please feel free to reopen if you do want to push it through.

I strongly support this. And it came up in response to [your own question in slack](https://drakedevelopers.slack.com/archives/C2WBPQDB7/p1738698187829379?thread_ts=1732652785.917409&cid=C2WBPQDB7), @amcastro-tri .

Even having the ability to change the constraint specs programmatically (not via ports) post-finalize would already be a huge improvement. Like the way that joint locking is being done in...

> Also, just because I am curious, what is the NonnegativeConeT(0) doing here? Thanks for pointing that out. It really shouldn't be there. I've opened https://github.com/RobotLocomotion/drake/pull/21646 to update our parser.

I'm also a little worried that we need to be careful with what APIs we offer here. Of course, one can always get the underlying GCS object and work with...

Perhaps this requires a whiteboard conversation. But I might prefer something more along the lines of: ``` GcsTrajectoryOptimization::Subgraph::AddPathPositionConstraint( const std::shared_ptr& constraint, const std::shared_ptr & convex_surrogate, double s) ``` or similar,...