>Nick Bastin - RBB Systems, Inc.
>Out hme0, through the Cat5K, Across the ATM backbone, through the
>firewall, past the provider, hit the router, down the fiber, off another
>router... Nothing but net.
>Nick,
>
>That's not necessarily true. While it is almost always true for
>consumer-off-the-shelf software, there is plenty of software that
>doesn't fit into that category. Quite a few software companies will
>sign support contracts (IBM is one) where they will take responsibility
>for damage that may arise from the use of the software. This is also
>the case for many industrial software packages. Granted, PostgreSQL
>doesn't really fall into any of these categories, but these types of
>warratees *do* exist.
>
Nick, I didn't say that those contracts don't exist. But
they are usually in conjunction with some specific application.
For example, Oracle's general license on there database
product's is written in such a way that they cannot be
held accountable for anything a customer may do with there
database. The reason's are simple, Oracle couldn't possibly
come up with all of the possible scenario's that their
general purpose software could fail in. But, on the flip side, Oracle has an Iron clad warranty
on the use of "Oracle Financials". And you can be assured
what a DBA can and cannot do are very strictly defined within
that contract. And if you violate that contract in any minor
way and "Oracle Financials" has any problem, the lawyer's
will use this as a way out.As for comparing commercial database software to PostgreSQL
reliability. I had to rebuild a Oracle 7.0 database engine
two weeks ago because it was corrupt. And last week I spent
two days exporting/importing a Oracle 7.3 two tablespaces
because of some corruption of some type. And I did this
after Oracle's tech staff suggested it. What does all of this mean. Well, Oracle (Commercial)
database packages have some of these same problems that plague
PostgreSQL. The only difference is they are less frequent and
are generally tied to the development cycle. I tend to think
of this as a evolution cycle and Postgres's cycle is on
steroids.
My two cents.
D.