The latter would be the crucial point, because if you're going to make the effort of programming all the contenders anyway, why would we actively remove one of them again at a later point?
You can think of it as that I'm going to go to the effort of making a
draft implementation of each skill, and share that draft. Only if they show promise, it'll proceed to a refined "good copy" of the skill - otherwise the draft gets thrown out.
The experimental builds will likely have bugs, and I will only fix those bugs (within the context of the experimentals, I mean; obviously they'd be fixed for any skill making it into stable) if they interfere with being able to properly analyze the skill's usefulness. EDIT: Or, maybe, if they're really really easy to fix.
If what im hearing is correct then I'm glad to see that its more than one skill getting in.
It's not guaranteed. If only one skill proves itself worthwhile after testing / in-depth discussion, only that one skill will make it in. If none of them do, we might retain the status quo. All I'm saying is that I'm removing the
limit on how many can make it in, and taking each skill at least as far as the "draft implementation" phase.