Open JakobBD opened 1 month ago
Adding @LaviniaBaumstark and @robertpietzcker to the discussion about the future of fixed_shares.
Pull request that are (hopefully) merged soon are not the best place for discussions that (unfortunately) will take a while. :point_right: https://github.com/remindmodel/development_issues/issues/290
Test runs are fine, can be merged from my point of view.
Will become obsolete soon with deleted fixed_shares, but that deletion will take a while if done thoroughly, checking for obsolete code in core and R packages, so maybe we can merge this in the meantime.
Purpose of this PR
A large chunk of code in the core is concerned with exogenous CCS shares for industry. The values coming out of it are overwritten (and therefore obsolete and unused) in the subsectors realization. This PR moves part of them to fixed_shares to declutter the core. Some other parts stay in core: It seems reasonable to keep the calculation of
pm_emineg_econometric
in one place.This branch has #1673 merged, so it should be merged after that.
I haven't run tests of this using the fixed_shares realization, only using subsectors!
This is part of a compromise: fixed_shares currently does not compile any more, since I broke it with the process-based steel implementation. According to @0UmfHxcvx5J7JoaOhFSs5mncnisTJJ6q, even with a fix to that, fixed_shares would take a while to get running again, since it would need new input data, and new calibrations. subsectors being as stable as it is right now, we could think about getting rid of fixed_shares altogether sooner or later. At the same time, subsectors is quite stable right now.
The current compromise is to keep it as close to a functional state as possible without much additional work.
My long-term proposal is to build a version of process-based industry where the process route shares are fixed, and to get rid of fixed_shares at the latest by then, maybe earlier.
Adding @LaviniaBaumstark and @robertpietzcker to the discussion about the future of fixed_shares.
Type of change
(Make sure to delete from the Type-of-change list the items not relevant to your PR)
Checklist:
remind2
where it was neededforbiddenColumnNames
in readCheckScenarioConfig.R in case the PR leads to deprecated switchesFAIL 0
in the output ofmake test
)CHANGELOG.md
has been updated correctlyFurther information (optional):
/p/tmp/jakobdu/remind_temp3/output
no changes