Re: Shared game-data

From: Bob Badour <bbadour_at_pei.sympatico.ca>
Date: Mon, 01 May 2006 13:40:55 GMT
Message-ID: <Hdo5g.1342$A26.37703@ursa-nb00s0.nbnet.nb.ca>

Dmitry A. Kazakov wrote:

> On 30 Apr 2006 15:56:39 -0700, JOG wrote:
>

>>.... Programmers need the tools - a relation template in the
>>STL, and support for RA in <algorithm> anyone? Given I still can't even
>>use a standard hashmap cross platform yet, I'm not holding my breath.

>
> Relations do not need templates. (Nobody needs them. (:-)) What is needed
> is generic programming. Templates represent a rather bleak example of. But
> that's aside.
>
> As for generic programming (=programming in terms of sets of types, as
> opposed to individual types). RM can be considered as one. It has some
> operations which can be viewed as ones producing new types (algebra of
> types) or defined over sets of types (OO classes are closures of types
> sets.) But to be fully supported, RM requires much better types systems
> than we have now. It is a long way to go. And, there is a thing (maybe
> unpleasant to some) at the end of the road, it will strip the glamour of
> being a "paradigm" off RM (OO as well).

Unless you mean "paradigm" in the sense of "an example", the RM is not a paradigm. Applied mathematics is a paradigm. The RM is a logical data model. Received on Mon May 01 2006 - 08:40:55 CDT