Dan Foreman-Mackey

> In your latest build it seems to be updating the cache: https://github.com/dfm/araa-gps/actions/runs/3053508686/jobs/4924221075#step:3:146 No. The problem is in the _restore_ step: https://github.com/dfm/araa-gps/actions/runs/3053508686/jobs/4924221075#step:3:131 It still (silently) fails.

Awesome - I'm not clear on the best way to handle it, so I trust you! Hopefully we can do something more sensible with the new overleaf interface.

As a short term fix, I've added a `git push` step to my workflow, which I think should fix things if this truly is the issue...

I think I might actually prefer to default to having the workflow push. If not, the Overleaf and syw versions of the manuscript will continue to diverge, until you run...

I can confirm that the push fixed the issue on our project. I probably won't be able to try this implementation out until later this week, but I like the...

Sorry about the delay! This seems sensible to me. It's hard to test this in detail though since we mainly hit this when actively editing the Overleaf while also pushing...

I haven't looked in detail, but here are some initial impressions: > A. Switch over to fully constrained environment specifications. ... For this approach, the [conda-lock](https://github.com/conda-incubator/conda-lock) project is designed to...

Oh - Snakemake actually does have support for this via conda "pins" https://snakemake.readthedocs.io/en/stable/snakefiles/deployment.html#freezing-environments-to-exactly-pinned-packages We could probably generate the pins using `conda-lock`.

Agreed that this is a good interface to provide. One point I would add here is that while this is straightforward for a single transiting body, you can end up...

You might be able to use it via [schwimmbad](https://schwimmbad.readthedocs.io/en/latest/examples/index.html#using-joblibpool), but I haven't really tried it.