At 02:44 PM 3/9/99 -0800, Greg Stein wrote:
>Justus Pendleton wrote:
>>
>> On Tue, Mar 09, 1999 at 02:19:26PM -0800, Greg Stein wrote:
>> > I'm more than willing to agree to disagree. I will continue work against
>> > a relational database. It provides much more flexibility, data
>> > integrity, reporting ability, and the ability to scale.
>>
>> You forgot to mention transactions :). I would think it would be
>> really nice to say "gee that jerk just submitted 25,000 bogus CDs but
>> I can just ignore the whole damn transaction rather than going through
>> each one by hand".
>
>The relational database is going to conduct transactions on a much
>smaller time scale than what you're referring to. It will open a
>transaction, the records will be inserted, then it will close the
>transaction. Done. An editor cannot come back along and unwind it
>without some serious effort (you'd need to go back to an old version and
>replay some (edited) update logs).
>
>In any case, I don't believe transactions are actually called for in the
>system. That is mostly necessary when data is being altered or deleted.
>For the most part, we're only looking at inserts.
>
>Cheers,
>-g
>
>--
>Greg Stein, http://www.lyra.org/
>
>_______________________________________________
>cdin maillist - cdin nospam at cdin.org
>http://mailman.cdin.org/mailman/listinfo/cdin
>
>
>
-----
Jason Dufair
funne nospam at iquest.net
http://www.iquest.net/~funne
http://www.iquest.net/~funne/jdufair.asc for PGP public key.
"A laugh for the newsprint nightmare, a world that never was
Where the questions are all 'why' and the answers are all 'because'"
-Bruce Cockburn