Excellent article, Alexander & co. Nicely sums up what we'd been talking about, and gives me an
idea of why you worry so much when we suggest that we use pessimistic locking (it's often a sign of
people coming from other db's, feeling the need for it and not thinking about firebird in context.)
It also contains a clean answer to Sugi's earlier question -- it's possible to do an update
regardless of the lock (by retrying) but if you do so, it's safest to use an update statement of
the form (a+=b) rather than (a=c) ... read-committed, or restart transaction with isolation ...
although I don't think we ever got an answer about what the actual update was that was being
attempted.

In any case, excellent work. I'm recommending this article to my fellow programmers, as a reminder.

-philip

> http://www.interbase-world.com/modules.php?name=News&file=article&sid=423
>
> We must express great gratitude to Helen Borrie and Dmitry Yemanov.
> Without their help this article perhaps could'nt be translated. Hope
> it will help somebody, last time here appeared a few questions on it's
> subject.
>
> Best regards, Alexanders.