One more point.  I am a proponent of dynamic languages.  I want to use
them where they make give most benefit and not let the complexity of that
flexibility creep out to where it is harmful to flexibility (at the
scalability of interoption layer).  I will explain how...

> And I am also proposing that NekoVM should add support for interface
> declarations on builtins.  Then HLLs can scalably interopt, by exposing
> their interfaces in common data type layer.  I do not see how you can
> possibly get scalable interoption without?  Otherwise, either interfaces
> are not used (and thus runtime exceptions complexity profileration and
> interoption failure) or HLL need to write N^2 translation layers to each
> others' higher level interface/types.

http://lists.motion-twin.com/pipermail/neko/2010-January/002709.html

I wrote:
"Dynamic lovers might argue that enforced interfaces are a barbaric relic,
but I don't see how one can possibly get scalable code without them?  You
could revert to check for interfaces dynamically, but then you proliferate
exceptions.  That eventually deadlocks composeability at some exponential
rate of mashups."

HLLs can expose the above proposed universal interfaces only for the
external interfaces they wish to expose.  One could continue to use
dynamic code internally (i.e. internal to a closed projects, internal to
external interface implementations).

My point is that for large scalability, we must expose interoperable
interfaces.  And those interfaces should be founded on a common data
types, and the sub-classing types should be done at the interface layer. 
Internal to our HLLs we may use all the sugar we want.  So we don't lose
anything.  Isn't it all win-win?

--
Neko : One VM to run them all
(http://nekovm.org)

Reply via email to