technical wet dreams and SQL fantasies

Guido Stepken <stepken_at_little-idiot.de> wrote in

> State of the art is MVCC, where all transactions can write into/read
> from table without blocking each other and without showing the symptom
> of dirty reads or phantom entries. This is well possible in PostgreSQL.
>
> So, IMHO oracle is not perfect.

The bottom line is that if rows are locked for change (update or delete), then it _will_ block any other change transactions attempting the _same_ rows.

Are you now saying that two people can change the same rows at the same time in PostgreSQL without locking out one another and without one transaction overwriting the other's changes?

> Why do you argue ? MS SQL Server programmers have admitted, that
> MVRC/MRCC ist still missing. This feature will be implemented in autumn
> 2004.

My argument is *not* about how technically kewl a database is, but how it best serves the business needs. Read my posting again. Slowly.

A MS-Access application can do pretty well as a single user POS for a video rental shop. Similarly, SQL-Server *is* the best choice under specific & unique circumstances as imposed by the business environment, requirements, budgets, etc.

Do I personally think that SQL-Server can hold a technical candle to Oracle? No ways in hell. But my opinion has NO relevance in choosing the *best* solution for a customer. If page locking will be a problem in that business environment, then yes, SQL-Server will be a problem. If the customer has only M amount to spend and Oracle licensing cost 2xM, then yes, Oracle will be a problem. If there are only Microsoft skills and no Linux/Postgre skills, then yes, Linux/Postgre will be a problem.

Horses for courses.

Fact. There is no single and definitive "best" database in the corporate and business world, irrespective of technical wet dreams and SQL fantasies. Period.

--
Billy