E. Belfer

I'm picking up on the integration process here. I've migrated the changes from this PR into a new branch / draft PR, which can be found [here](https://github.com/catalyst-cooperative/pudl/pull/2281).

> I think it probably makes sense to keep these tables separate post harvesting just because there is so much information that would be duplicated from the annual 860 table...

@jdangerx In addition to resolving this "where should our fields live" question, can you add the `dagster instance migrate` instructions (for minor version updates) somewhere to our docs? Otherwise I...

@cmgosnell: Out of 205 unique entities in the raw XBRL instant data, there are only 4 utilities that report data in the leafy "accumulated provision for depreciation of X" columns...

@davidmudrauskas Happy to chat about the design process here for the transformation, or look over a draft PR. Just nudge me if you need anything!

That's exactly what's happening. When implementing, I considered this and didn't wind up choosing the window option, partially because it's slightly more challenging to implement but primarily because it raises...

Thanks for wading into our extractors! `process_raw`, `process_renamed` and `process_final` are functions that we tend to use when something is strange in the original data that prevents us from effectively...

Looks like there's a tiny docs issue but otherwise this direction is looking great so far! I'm going to ping another infrastructurally-minded person to take a look this week and...

This looks great to me! I think after the last few little changes, we're going to want to run first the fast ETL and then the full ETL + validations...

@davidmudrauskas I'm going to kick off a full build on this branch to see if I can reproduce the calc check failures you're seeing. The code changes you made _shouldn't_...