Re: teaching relational basics to people, questions

Mr. Scott wrote:

>
> "Roy Hann" <specially_at_processed.almost.meat> wrote in message
> news:3eudnYx_I7CpiZXWnZ2dnUVZ7qSdnZ2d_at_pipex.net...
>> Sampo Syreeni wrote:
>>
>>> [snip] And since
>>> one would have to have a bona fide range datatype, building in
>>> handling for infinite ranges would also be easy; that'd get rid of one
>>> of the most persistent reasons why people incorporate nulls into
>>> designs.
>>
>> I think you are being excessively optimistic. The most persistent (and
>> most common) reason people incorporate nullable columns into designs is
>> because they have a misplaced desire to minimize the number of tables in
>> the design, and think that conflating multiple fact types in one table
>> is clever, efficient, and harmless.
>
> I think you are oversimplifying. The most persistent (and most common)
> reason people incorporate nullable columns into designs is the not so
> misplaced desire to provide for information that is relevant but not
> required, and that doesn't necessarily involve conflating multiple fact
> types in one table. Each table design represents a family of interdependent
> predicates.

That last sentence is the crux of your objection to my comment. I guess since you and I are implicitly talking about SQL and since SQL makes few if any claims to fidelity with the relational model, you are entitled to think whatever you like about tables that represent "a family of interdependent predicates". Personally I have no idea what those are supposed to behave like, but I am pretty sure that in practice they must suck. Certainly every example I've ever had to deal with did. At this point I invoke the Principle of Incoherence and drop out of the discussion.

I will however allow that you may be right that yours is an even more common reason people introduce nullable columns. It would be worth investigating which really is most common, so that appropriate remedial training can be devised.

[much erudite-looking stuff snipped]

-- 
Roy