Re: TRM - Morbidity has set in, or not?
From: Bob Badour <bbadour_at_pei.sympatico.ca>
Date: Thu, 18 May 2006 22:02:53 GMT
Message-ID: <ha6bg.9542$A26.237330_at_ursa-nb00s0.nbnet.nb.ca>
>
> For data representation, I agree completely. But I think of UIs and
> reports as forms of data extracts: they're restricted and (typically,
> but not always) tree-structured "views" of a subset of a database. The
> power of relational for shared databases is in part its ability to
> support arbitrary extracts of the data in an egalitarian fashion - no
> application bias. Any given UI or report includes a set of (typically
> key-related) restrictions and projections across different relations.
> It appears to me (and I could be wrong, as this is only a pet
> hypothesis at this point) that expression languages to derive these
> would be tree-structured, or at least would be different than a
> relational algebra or calculus.
>
> In fairness, it wasn't Dawn who introduced the term. I thought maybe it
> had a meaning, but suspect it doesn't. Agreed that the O-O folks don't
> have a consensus on this, nor is it critical in the discussion of data.
Date: Thu, 18 May 2006 22:02:53 GMT
Message-ID: <ha6bg.9542$A26.237330_at_ursa-nb00s0.nbnet.nb.ca>
erk wrote:
>>If one wants to use >>a different syntax, it would make much more sense to use a translator. >>However, none of the examples of alternate syntaxes she gives could >>possibly offer any useful benefit over predicate logic.
>
> For data representation, I agree completely. But I think of UIs and
> reports as forms of data extracts: they're restricted and (typically,
> but not always) tree-structured "views" of a subset of a database. The
> power of relational for shared databases is in part its ability to
> support arbitrary extracts of the data in an egalitarian fashion - no
> application bias. Any given UI or report includes a set of (typically
> key-related) restrictions and projections across different relations.
> It appears to me (and I could be wrong, as this is only a pet
> hypothesis at this point) that expression languages to derive these
> would be tree-structured, or at least would be different than a
> relational algebra or calculus.
Either the algebra or the calculus easily handles the simple restrictions required. Why would you want a different language?
>>Even if one goes the extra mile and tries to make sense of the gibberish >>by substituting "computational model" or "model of concurrency", the >>object oriented folks do not have a consensus on either so using the >>definite article and suggesting that provides a unique starting place is >>lunacy.
>
> In fairness, it wasn't Dawn who introduced the term. I thought maybe it
> had a meaning, but suspect it doesn't. Agreed that the O-O folks don't
> have a consensus on this, nor is it critical in the discussion of data.
In fairness, she embraced the concept when she stated agreement. Agreeing is easy if one's only goal is sophist rhetoric. Received on Fri May 19 2006 - 00:02:53 CEST