[general_re]

I agree, insofar as my own philosophy is "the right tool for the right job" - Linux definitely has a place in the world. But that's not the same as saying it's right everywhere - Linux as a replacement for the Windows desktops of the world is an awfully elusive goal, and on the other end, it's not ready for prime-time in the world of really big iron.

Notice I'm not talking supercomputing, but rather traditional mainframes. The advantage that mainframes have always had is not in raw computing power, but rather in their tremendous I/O throughput - to use another technical measurement, their ability to move a s***load of bits from 'A' to 'B', and do it really, really fast. ;)

Which Linux just can't really do well yet. Maybe soon, but not today.

[bvw]

As far as can tell right now what locks MS on the desktop is two things -- Word and Power Point. Those two are needed in so many unavoidable places -- school, work, g-ment -- and the work-alikes are not good enough.

The Word work-alikes always seem to be one or two MS versions behind and the variances are usually a big nuisance -- especially to the "basic level users" one usually has to sell to or deal with. I'm surprised that there is no good -- as far as I know -- Power Point substitute.

[bvw]

In my humble opinion, the strength of the mainframes is in a fossil-like stability, and a priesthood of mainframe developers. That means that aps on mainframes work reliably. It has zilch to do with data throughput.

Unix aps are always in development -- not talking the basic unix tools which are fairly stable -- just the aps. Everyone knows they can get radical changes from unix aps and there are zillions of programmers to do it. The Mainframe priesthood is well-learned in the techniques of rebutting anything but a little change here and a little change there.

The consequences are that unix aps look more quirky and unreliable in managemnet outage summary reports. Mainframes are low outage.