Re: [DR202] RPC and data encodings

On Wed, 15 Nov 2000, Martin Gudgin wrote:

> I'm strongly against this. There is significant value in having a standard
> data rep that everyone agrees on. Allowing multiple data reps leads to
> incompatible implementations. Let's not open the door for that in XP. If
> people want to use other data reps, they will anyway, but if we mandate a
> data rep they won't be able to claim XP compliance.
> 
> Gudge

Gudge, you are absolutelly right ... but in 3 years we come up with a
better data representation and what do we do? Nothing??!! Let's try to
make this as flexible as possible so the standard could evolve over time.
If we try to fix everything to something definite the protocol would not
be able to use anything else. It is true that we would have problems of
user disagreeing about data represention but we make a mandatory
implementaion of a data rep. and than say how you can use any other
one.... In this way you can claim XP compliance....

- Octav

Received on Thursday, 16 November 2000 00:42:55 UTC