Re: Code in the database or middle tier (the CLR controversy)

--CELKO-- wrote:
> LOL! Actually, it will be all of the barbarians, snow monsters, etc.
> in those "What's in your Wallet?" ads ALL hitting you at once.
>
> What do the CLR languages do with bits? Did you know that +1, +0, -0
> and -1 are all used for BOOLEANs, but not consistently? This varies in
> MS proprietary language like C# and VB. How do they define their
> functions? In the first edition SQL FOR SMARTIES, I had a list of the
> various vendor MOD() functions differences in the SQL products of the
> day. What about NULLs in the CLR languages?
>
> How do you maintain a mixed system that uses languages you probably do
> not know? COBOL is the most popular language on Earth and 70% of the
> commercial code is in COBOL. I would guess this newsgroup has less
> than a dozen COBOL programmers in it. How many DBAs know C#, C++, C,
> Java, VB, Pascal (Delphi), and whatever else a developer can sneak into
> the schema before they leave?
>
> What they should have done is implement the SQL/PSM and Standard
> trigger syntax as a replacement for T-SQL.

Good points but IMO missing the most essential.

When database programming is restricted to SQL, T-SQL, PL/SQL, etc. then database programming is "unofficially" restricted to people who at least have some familiarity with database concepts (yes we might agree on how much or little but at least some).

Open up the database to VB, C#, etc. and you are encouraging every incompetent, untrained, self-annointed programmer, to think they know something about relational algebra, set theory, constraints, and more: And we all know they don't.

My prediction: 2 years after VB is available in SQL Server ... the majority of SQL Server code will be VB. And it will be as solid as an overripe tomato.

-- 
Daniel A. Morgan
http://www.psoug.org
damorgan_at_x.washington.edu
(replace x with u to respond)