[armedbear-devel] Mixing Java/Lisp abstraction barrier in ABCL implementation

[Ville Voutilainen]

On 16 April 2011 17:19, Mark Evenson <evenson at panix.com> wrote:
>> At any rate, I'm all for having such an extension even if it initially
>> uses jcall/jnew, we can do the abstractions as it evolves. I don't think
>> we need to sit on our hands and not release such extensions just
>> because it's not philosophically "correct" from the get-go. :)
> In practical terms, we rarely have the time to go back to correct
> things, so if we let this "slide", it will probably remain this way for
> a while.

Well, if that's a valid concern, I'd prefer this one being done through
primitives rather than jcall. It fits the criteria of being non-java-specific.