Andrew Lentvorski wrote: > NO!--NO! NO! NO! NO! NO! > > More brain damage has been inflicted due to that belief than any other > in computer science. > > A filesystem is *NOT* a special case of a database. A filesystem has > very different semantics and performance goals than a database. A > database is strongly transactional and geared toward throughput. A > filesystem is weakly transactional and has some latency constraints. > There are lots of others--multidimensionality vs. limited > dimensionality; query restructurability vs. enforced constraint; etc. > >
No computer class that I ever took defined "database" as necessarily requiring transactions and being geared towards throughput. I had a "database" as part of Appleworks back on my Apple ][c and it had neither transactions nor throughput to speak of. Are you telling me it was not a database? > Whenever anybody says: "X is just a special case of a database" they are > almost always wrong. > In my opinion a database is just a store of information. Since most things in computer science deal with stores of information I can see how this idea would come up often. What I don't see is why you take such exception to it. You seem to have greatly overloaded the concept of "database" far beyond what it is traditionally ascribed. -- Tracy R Reed http://ultraviolet.org -- [email protected] http://www.kernel-panic.org/cgi-bin/mailman/listinfo/kplug-list
