Re: Operation "Fat Emacs"

Philip Kaludercic <[email protected]> writes:
> No, I am not aware of any plans in that direction.  That has the
> inherent issue that packages would suddenly stop working if they depend
> on some part that a more minimal Emacs would throw out.  That being
> said, the point has been raised that having a tighter binding between
> the core and GNU ELPA would not only make it easier for packages to be
> added to the core, but also to move to ELPA if the opinion is that they
> would be better maintained there.



Thanks. That makes sense: there is indeed a risk of breaking packages that depend on parts removed from a more minimal Emacs.

At the same time, I wonder if that could also be useful as a way to make Emacs' internal dependencies clearer, and perhaps make its architecture more modular, understandable, and predictable over time.

It could maybe start in a very incremental way, by trying to remove obviously self-contained pieces first and seeing what breaks. Even if that only reveals tightly entangled dependency clusters, that would still be useful: at least those could be identified, documented, and perhaps disentangled later.