Re: Experimental features

[Stefan Monnier]

>     I know, but I think for example that the vc-bzr.el and vc-hg.el (and
>     potentially others, I hope we'll get vc-darcs.el soon) are somewhat
>     dangerous to add to Emacs-22.[23...] unless we disable them by
>     default: they're not 100% modular since they normally get involved
>     every time we open a file (in 99% of the cases they'll just bail
>     saying that they're not interested in this file, but still).

>     So just like we have obsolete features, we could have experimental
>     features, which are completely deactivated by default but can easily
>     be activated by the user.  This might allow us to safely include a few
>     more new features in minor releases.

> I see the point; maybe so.

>     I'll send a patch soon, to give a more concrete idea of my I'm
>     thinking of.  Basically, I currently picture it as adding something
>     like a function (activate-experimental-feature <feature>).

> What is the benefit of this, over having a simple variable to control it?

There are a few issues:
1 - sometimes a single `setq' is not enough to activate a feature.
2 - I like the idea of being able to list the experimental features.


        Stefan