Jerome Kelleher

Let's just let this one sit for a bit - 3.11 packages aren't crucial for tsinfer here, right?

Good idea. We can add a metadata schema like we have for nodes and an enumeration of the types of mutation.

Best to wait until we also have node metadata sorted out with a proper schema. Will have to wait until 0.2.1.

A better solution might be to move intervals.py into tskit. We will probably be dealing with these tedious issues for getting ratemaps into tskit for LS matching soon, so that...

If it's a clean swap-in then yes, we should probably just move the code straight up into tskit. We can leave the file in msprime until the tskit version is...

You could imagine doing this with SSE instructions as well, but almost certainly not worth the hassle.

I'd rather punt this one to the next release after we've got basic interface with sgkit done (#728). We'll get a better sense of how attributes like this could be...

Let's avoid adding anything we don't *need* to the data file formats before the switch over.

See also #729 - will need to deal with this problem more generally. I'm hoping that we can get to a point ultimately where we can provide a list of...

Good idea - let's get it in for 0.2.1