On Sat, 17 Jun 2000, Cris Jacobin wrote:
[n.b.: I never replied to this. I know it's old, but I kept it around
with the intent of replying, and only just now got to it.]
> Yes, but have you ever seen a database running under some form of
> comitment control actually perform a rollback?
Yes, but computational cost is not an issue. Ideally we are not needing
to rollback unless there's been a commit of corrupted data, which should
be a rare enough case in itself. It's a failsafe so that we don't need to
abandon world state if we have a fault.
> In any case, SQL itself isn't all that efficient.
Again, efficiency isn't a big issue. SQL servers may not have optimum
performance for our particular needs, but for the majority of our work
they'll perform more than admirably. The slowdown won't be noticable to
users in 99% of the circumstances, and the flexibility, ease of
implementation, and well-known query syntax make it the best option.
> I'd sure like to see how the no-code mud imps ,who have trouble
> figuring out what the +/- mean during hand patches, handle being
> confronted with things like the difference between inner/outer joins,
> etc! Are you normalized? Huh?
Obviously a model which employs complete world persistence has a long way
to go before its accessable to the masses. And even then there will
always be things which are beyond their immediate comprehension. I was
suggesting it as the most economical answer to the question of
persistence.
-dak
+------------------------------------------------------------+
| Ensure that you have read the CircleMUD Mailing List FAQ: |
| http://qsilver.queensu.ca/~fletchra/Circle/list-faq.html |
+------------------------------------------------------------+
This archive was generated by hypermail 2b30 : 04/10/01 PDT