Re: Transactions: good or bad?

From: (wrong string) öm <robert.sundstrom_at_upright.se>
Date: Tue, 29 Apr 2003 09:29:00 GMT
Message-ID: <b8lgfd$r7t$1_at_yggdrasil.utfors.se>


In article <pyora.664367$L1.189041_at_sccrnsc02>, "Marshall Spight" <mspight_at_dnai.com> wrote:
>I've always thought of transactions as being a good thing.
>It seems to me that there are various synchronization mechanisms
>out there: mutexes, semaphores, Java synchronization, etc., but
>that transactions are actually the highest level synchronization
>mechanism I've encountered.
>
>But recently I read that "transactions ... are not compatible
>with the 'arrow of time'. They let you freeze time and that
>is not a good model of reality."

I don't really see the problem. From a abstract point of view, a transaction merely allows an application to perform multiple operations logically at a single point in time. The database compensates for the fact that the computer physically can't do several things simultaneously.

Many database products offers different levels of isolation so it is up to the programmer or user to decide what is most appropriate for his problem.

I think this "arrow of time" statement just was to stir up some dust in this newsgroup. :-)

-
Robert Sundström, Mimer SQL Development
Upright Database Technology AB, http://www.mimer.com Validate your SQL statements at http://developer.mimer.com/parser Received on Tue Apr 29 2003 - 11:29:00 CEST

Original text of this message