Re: Does Codd's view of a relational database differ from that of Date & Darwin? [M.Gittens]

Ged Byrne wrote:

> And this sums up the real problem, which so often gets lost in the
> discussions.
>
> In practice Nulls are the only solution available, and although
> problematic they are getting the job done.
>
> Although unanimous in their criticism of NULLs, the DB-Debunks are yet
> to find agreement on the better approach. Pascal favours the use of
> meta data, while Darwen suggests the use of vertical decomposition.
> They do acknowledge, however, that they still don't have the answers.
>
> It is easy to scoff because they are not entire unified on the subject,
> but they do seem to be the only people to be thinking about the problem
> The rest of the industry appear to be satisfied with things as they
> are.

Many are quietly just using blank values and flags. They are no more aware of it as a concept than they are aware of air. For many this arises from early difficulties which they solve by just deciding not to use NULLs, since their code is so much simpler that way.

-- 
Kenneth Downs
Secure Data Software, Inc.
(Ken)nneth_at_(Sec)ure(Dat)a(.com)