Hmm, darn.

It'll be 3.6a (3.5 is now gamma), but if some VM builders are willing to
do this (considering that the whole thing is still a stretch away from
being accepted into the image), it sounds like it would indeed help the
testing progress.

Daniel

Andreas Raab <andreas.raab at gmx.de> wrote:
> > I'll throw in another question (for Anthony/VM implementors). 
> > Is there a reason the additional primitives needed for Anthony
> > current BC code to work well can't be released as a plugin? 
> 
> Probably not. Two reasons: One is that it's about block activation which is
> at the heart of the VMs behavior. The second one being that they are some of
> the most time critical primitives (it defines how fast Collection>>do: will
> work!) and therefore those are some of the (few) primitives which would
> rightfully deserve an indexed entry.
> 
> Here's a proposal which might help to relax the situation a little: Consider
> adding just the primitives required to 3.5alpha. This allows people to chose
> what they want to use for the time being until a decision has been made
> either way.
> 
> Cheers,
>   - Andreas