Going to be One Of Those Days

[Nick Christenson]

> >> But PCs are still crap.

They are, indeed.

> > Bang-for-buck-wise, they ain't bad.

This also seems to be true.  Economics unfailingly moves towards 
commodities.

> I'm not so sure.  I'd _love_ to see a real study done that takes into 
> account all the relevant factors [...]

I would too, but I'm coming to the opinion that there will never be one,
and that any circumstances under which one would be done in a truly fair
manner would be too narrowly defined to be useful to a general audience.

Generally, I've stopped reading TCO studies (except for those my boss
specifically asks me to analyze) 'cuz I haven't read one in the last 
decade that wasn't written by someone with an agenda that becomes
manifestly clear in the study itself.  When I have to read one of these,
I generally do it at home now lest I disturb my work neighbors with 
my swearing.  Summary:  They're all worse than crap.

> [...] so decided that "you get what you pay for" is somewhat 
> close to correct.  I have this uncomfortable feeling that that adage is 
> deeper than typically interpreted, and that "cheap" HW ends up costing 
> the same as or more than the "expensive" stuff in the long run.

There is a lot of truth to this, but there are many cases where the costs
are compelling.  A later poster does a pretty good analysis of cheap 
vs. rock-solid scenarios that I largely agree with.  In the business world
of the 21st century, so far inexpensive generally is winning out over 
high-quality in the IT space by a wide margin.  Is this foolish?  In
many cases, perhaps.  Is the tide reversible?  I guess I don't see how.

Consequently, the argument in the long term might be moot.  Here's a 
question for the list.  Does anyone here think Solaris has a long-term 
viable future?  You could also ask the question about HP-UX, I suppose, 
but I expect there to be consensus that the other proprietary Unices 
definitely have no long-term.  This is not to say that Solaris doesn't
have a lot of years left in it, I believe it does.  Heck, look how 
long it is taking IRIX to completely die.  Moreover, does Solaris on 
Sparc (arguably less-crappy hardware than the PC world) have any real 
future?

My predictions:  I believe that Solaris on Sparc has about five years 
before it "feels" like, say, IRIX or AIX of today, merely in maintenance 
mode.  I believe that Solaris on Sparc has ten years before it's dead for 
all intents and purposes, although some vestige of it may limp along.  As 
we all know, Sun plans to Open Source Solaris, starting with Solaris 10.  
Good for them.  Does Open Source Solaris (on x86) have a future, or is it 
just a matter of time before the PC hardware world is running Windows, Linux, 
or a hobby?  I'm guessing Solaris on x86 dies at the same rate at Solaris 
on Sparc does, but I'm less sure of this.

Other opinions?

-- 
Nick Christenson
npc at gangofone.com