Science structure and balance are weird

[alex-dev]

Problems

Balance Issues

• Duplicate experiments: bd_pegasus and logmmImpacts are the same conceptual experiment;
  • Either we merge them or we create included experiments.
• Lot of light based experiments without proper allow_shrouded conditions;
• Compatibility between BDB and DMOS is spotty at best and not really working;

  Structure issues

• Kerbalism patch order is kinda spotty at best. #17 does solve quite a few of those but more should be done.
• Directory stucture itself is hard to understand and extend.
• Lot of features are not defined in KERBALISM_EXPERIMENT_VALUES which makes balancing hard to do.

  Solution

  I'd like to work on both of these issues one after the other.

• [ ] Refactoring the structure to bring it closer to RO-Kerbalism (which is much simpler than Kebalism)
• [ ] Setup proper default for all experiment
• [ ] Merge duplicated experiments (whether they should be real merges or includes is to be decided case by case)
  • This would also handle most compatibility between mods With a proper structure in place, it will be easy for anyone to just tweak values for a better balance.

I won't be able to work on this during the next week, but after that, I'd have some time.

[Gupyzer0]

Go ahead. In the case of mmimpacts and pegasus id like to include mmimpacts in pegasus, as pegasus is for larger surfaces.

As for the structure, well, yes, it was a bit of a pain to implement everything in kerbalism like structure. I might be able to do a ro-kerbalism structure, as it was the original layout a lot of commits ago hehehe 😅

Btw I might try to rename BDB experiments to SSS_xxxx like skyhawk did and create newer bd_camera derived experiments based on higher resolution cameras like I did with the film based cameras.

[alex-dev]

I would go with owning the experiment we want from BDB and converting everything to our own experiments. That way BDB becomes only a supported mod. Not a core. And we would also try to avoid adding experiments as much as possible. Unless we want the mods experiment, we would patch parts to use one of our own. This would solve most compatibility problems... And is pretty much inline with Kerbalism philosophy (AKA we do it better so we own it).