Mail Archives: geda-user/2015/08/24/20:52:49
On Monday 24 August 2015, John Doty wrote:
> Yes. That’s constructive. Bare-bones simplicity out of the
> box, opt-in complexity. That’s the way a lot of excellent
> software (Python, LaTeX, …) works.
That's one of the reasons gnucap does everything in plugins.
"Bare-bones simplicity out of the box, opt-in complexity".
It's important here that gnucap does EVERYTHING in plugins. All
device models, even the resistor, are plugins. All commands
except one are plugins. Which one is the exception? The
command to load plugins. But really even that one is designed
as a plugin, and can be replaced by a plugin. Even the input
file format is a plugin.
On Monday 24 August 2015, Dave McGuire (mcguire AT neurotica DOT com)
[via geda-user AT delorie DOT com] wrote:
> You know, I hate (HATE!) all of the forking that happens in
> the FOSS world.
That's another reason gnucap does everything in plugins. It
greatly reduces the need to fork.
Without plugins, forks are necessary, as the only way to share
experimental code without turning the whole project into a big
mess.
On Monday 24 August 2015, Dave McGuire (mcguire AT neurotica DOT com)
[via geda-user AT delorie DOT com] wrote:
> But, as opposed as I am to willy-nilly forking of FOSS
> projects, it sounds like you need a fork. Gschem is not
> going to sit still just because you are opposed to any
> change.
Gschem and PCB both need to "fork" to rearchitect to use plugins
like gnucap does.
- Raw text -