On Fri, Feb 11, 2011 at 3:37 AM, Simon Wilkinson <[email protected]> wrote:
>
>
> On 11 Feb 2011, at 03:53, "Matt W. Benjamin" <[email protected]> wrote:
>
> yes.  at this point, an rpc-l primitive and corresponding rxgen support seem
> like something we would need to enable manageable and consistent elaboration
> of different union types, if I at all understand the issues.  I'm happy to
> make xcb conformant with this, for example.  I am not interested in seeing
> this wheel invented several times.
>
> Coming at this from a slightly different angle, I think we have an immediate
> need for a new address primitive. We can create, and implement, one
> relatively easily. I don't see doing so as reinventing the wheel.
> Separately, I agree that we need a new method of encoding unions on the
> wire. However, I think this is going to require more standardisation and
> implementation work. I'm definitely interested in seeing that work done, but
> I don't think we should delay moving forward with addresses on waiting for
> that to happen.

I agree with that summary, and had went to bed with the same idea in my craw.

-- 
Derrick
_______________________________________________
AFS3-standardization mailing list
[email protected]
http://lists.openafs.org/mailman/listinfo/afs3-standardization

Reply via email to