Hey, my Ignore button is gone!? Maybe I'm still on the old SI here at home?
twister,
you appear to be falling into the same trap as your detractors, i.e.: trying to defend a position based on historical numbers. If that worked, then I could quickly cobble up a program (perhaps I would run it on my 1991 all-Intel 386 system, including Intel System V.4 Unix) to guarantee myself handsome returns.
but, alas or not, value is much more psychological than that. we are gambling, here, and we must place our bets according to the individually unique information which each of us has gathered over the past n years.
Fundamentals do, indeed, matter. But the market is a fickle thing. Today's BetaMax can be tomorrow's VHS.
We have to go with what we know, estimate what we don't know, and then shoot from the hip.
I have been using Sun products for 10 years, primarily for development, and they have served me, IMO, remarkably well. For much of that time they were not the glamorous system-of-the-month, but their utility was clear to those who had to churn out the product (using GNU compilers, of course, lately, which I download from Sun.com's web site). My experiences porting and supporting my products on a couple dozen flavors of *NIX for many years lead me to rate Sun as the best on average, hardware + software, integrated solutions, standard systems.
I have also used Microsoft's products for a long time, from pre-DOS to NT5. I think I bought their Basic Compiler for my CP/M system in 1982. Back in those days, and even thru the 3.1 era, they were a company that provided enabling technologies to aspiring software developers. Now, however, they want to tell everyone how to do everything and, if it starts to get lucrative, they want a piece (or all) of it. They've really gotten away from what got them to where they are.
That's a reason to bet on Sun, who is still putting out the tools for us developers to do our thing. And they've got Bill Joy, who has had a nose for the right stuff even if his code stinks.
-tck |
