[bitc-dev] Are type classes just abstract, generic classes?

[Jonathan S. Shapiro]

On Wed, Mar 24, 2010 at 1:39 PM, Michal Suchanek <hramrach at centrum.cz> wrote:
>
> > Now if all of this is the case, then we must ask whether we shouldn't
> > just allow fields in [type] class definitions....
> I guess you don't really want fields but a possibility to substitute
> fields for methods with no arguments creating accessors on he fly.

That is sufficient ONLY if we can statically compile out all of the
resulting overhead.

> ... even normal functions are called through lookup tables these
> days to allow for relocatable executables and shared libraries...

It is true that *some* functions are called this way. However, both
shared library linkage schemes and relocatable code generation schemes
go to great lengths to eliminate this indirection in the usual case.
The vast majority of calls do not go through any lookup table.

> The other alternative is including abstract enough representation of
> the code in the objects so that specializations can be instantiated at
> a later time.

Yes. This is something I was looking at before joining Microsoft, and
something I plan to pick up again.


Jonathan