[tuning analogies snipped]
>
> Likewise with self-proclaimed computer tuners.
You have no idea how much I agree with you there.
> > I really don't understand why people expect computers to do everything
> > for them, the burden of using tools properly belongs to the user.
>
> I of course agree in principle to this statement (and the whole tone of
> your reply) -- but, my statement doesn't reflect my opinion -- it
> reflects reality. Facts are stubborn things.
>
> Of course the defaults will never be perfect -- nor will all users RTM.
> But, therein lies the utility of 'defaults' benchmarking -- let's see
> what the trade offs really are so that we the 'experts' can
> intelligently recommend things -- as well as intelligently tweak the
> defaults.
Unfortunately, there are two ways you can load the defaults. You can strip
them down for maximum speed, or you can load them up for maximum
reliability. The latter will make the users complain about speed on the
support lists in a generally annoyed fashion.
Stripping things down for sppeed, OTOH, will work great - until someone
gets a powercut, or some other kind of wierd hardware failure that will
wipe out their data. Then they will come back again and complain.
And the answer is always to simply spend an hour or so reading the
documentation...
Some people, eh...
Regards.
Gordan