Hi!
I checked out the code yesterday, and the first thing I noticed is how... Monolithic it all is. No offense to you, but Claude clearly has no reflexe to actually create untangled code.
It's quite a mess, some parts of the program know way, way too much about other parts without it being needed. It is difficult to maintain such an untangled codebase, both for AI and for humans, so I am proposing to work on a pull request to untangle everything, to have a clean separation of concerns and no god objects.
Of course, before I actually submit this work, I'd do a comprehensive bunch of testing, manual/runtime as well as automated testing if possible, to ensure that absolutely nothing breaks.
Is this something that would be of interest?
It will also make it much easier in the future to add features and fix bugs in existing features without accidentally affecting whole building blocks of the code that are unrelated.
Hi!
I checked out the code yesterday, and the first thing I noticed is how... Monolithic it all is. No offense to you, but Claude clearly has no reflexe to actually create untangled code.
It's quite a mess, some parts of the program know way, way too much about other parts without it being needed. It is difficult to maintain such an untangled codebase, both for AI and for humans, so I am proposing to work on a pull request to untangle everything, to have a clean separation of concerns and no god objects.
Of course, before I actually submit this work, I'd do a comprehensive bunch of testing, manual/runtime as well as automated testing if possible, to ensure that absolutely nothing breaks.
Is this something that would be of interest?
It will also make it much easier in the future to add features and fix bugs in existing features without accidentally affecting whole building blocks of the code that are unrelated.