Re: relational reasoning -- why two tables and not one?

From: paul c <toledobythesea_at_oohay.ac>
Date: Thu, 15 Oct 2009 22:15:47 GMT
Message-ID: <naNBm.48184$Db2.29866_at_edtnps83>


Keith H Duggar wrote:
> On Oct 15, 11:37 am, Bob Badour <bbad..._at_pei.sympatico.ca> wrote:

>> paul c wrote:
>>> Roy Hann wrote:
>>>> ...
>>>> Can one have a donor who has not donated (yet)? ...
>>> That is a great question because it indicates the rampant database
>>> mysticism in the semi-literate so-called developed world.  I'm sure
>>> there are db's where prospective donors are called donors and a donor
>>> isn't required to actually donate!  Managers without budgets are similar.
>> I am not sure what you mean by mysticism. What part is mystical? What
>> makes it mystical?

>
> I wonder too. For example, if one is a registered organ
> donor but has not yet died and donated their organs, are
> they mystical?
>
> KHD
Okay, let me take another stab at it: making up a place-holder to stand for something we aren't sure of (okay), then pretending we are sure without recasting the place-holder (not okay). It always seemed to me that the most valuable result of requirements meetings (maybe the only result!) was discovering what we were really talking about. I think it's fair to expect a reasonable common reading of such terms. Donor doesn't mean volunteer!

Every so often somebody posts a link here to a new "technology" (a very over-used word, the discovery of fire certainly meant a new technology but I doubt if, thousands of years later, the Swedish "Match-King", forget his name, would have made that claim about his matches because the result, what people could do with his matches, wasn't novel at all).   Usually the only thing that is new is somebody's pet re-phrasing of an existing computer language. Nine times out of ten the write-up is full of grammatical, spelling and definition errors, making it at least confusing and more likely incoherent, so it only serves to demonstrate the author doesn't understand what he saying and probably doesn't understand what he's writing about. I once complained to a bright support programmer for a database compiler about a feature description and he replied "Jeez, we have CS degrees, not English degrees."

I don't think db design is as difficult as language design but it does have its own subtleties (to be fair, I think the biggest problem in db design lies not within the subject itself but comes from the inadequacies of the typical dbms product). Like language design, db design is likely to go awry without precise language.

Russell made the same point as Bob B when he said of pure math, something like it is the subject in which we don't know what we are talking about, or whether what we are saying is true. He was talking about the excision of meaning to allow application of logical rules. That is part of the programmer's province. The programmer is tested when users receive his result. I don't mean to sound as if I'm lecturing the regulars here but vocabulary is important for user understanding and agreement. Received on Fri Oct 16 2009 - 00:15:47 CEST

Original text of this message