> -----Original Message-----
> From: [EMAIL PROTECTED] On Behalf Of Michael A. Stone
> Sent: Friday, October 30, 1998 9:44 PM
> ... the actual amount of code which would have to be rewritten
> (to fix Y2K problems) would be very, very small,
> and that he already knows exactly where it is.
Changing code from 2-digit year values to 4-digit is one of the
smaller parts of the Y2K problem. Larger parts include handling
existing data with 2-digit year fields, finding the damn source
code, and figuring out how to deal with that foot-thick deck of
cards containing hex patches that were applied to the executable
file fifteen years ago.
> ... the programmers who created those systems in the first
> place were *not* idiots. ...
Thank you. We're also not all dead or retired yet.
I once did a back-of-the-envelope analysis of a system that I
wrote that was seriously Y2K-deficient. Leaving out the bits
'00011001' (19 in packed decimal) everywhere they would have
occurred in dates saved three times as much in storage costs
as the entire development cost of the system. We're talking
here about a state-of-the-art IBM 2314 disk storage system,
approximately the size of five large refrigerators placed
side-by-side, costing about $1.25 million in today's dollars,
and storing 240MB. Not GB, MB. It never occurred to me that
the program would still be in use 32 years later, and it
wasn't, but it made it to 25 years.
> these guys knew that the rollover problem would occur before they ever
> wrote the first line of code, and most of them built their systems
> accordingly.
Huh?
> according to the gentleman in the interview, the majority of old systems
> have the solution built into their bones, you just have to know
> how to find it.
I'm surprised that your leg didn't come off in his hand.
> ... in the meantime, they've gotten all their
> insurance, retirement benefits, and back pay which somehow disappeared
> during the age of downsizing.
More age discrimination than downsizing. It's all these young punks who
don't have two smart kids at Brown and a less-bright one at MIT, a big
mortgage, and retirement fast approaching, so they'll work for beans.
> the same thing applies, in general, to Microsoft. they don't have to
> write anything *into* the operating system to have a major advantage, it's
> much more effective to leave out one or two small things which are hard to
> find.
>
> it's almost trivially simple to write an OS which is stable for programs
> that do A, B, and C, and crashes anything else immediately. it's jsut as
> easy to crash anything which does K, L, and M, and avoid using that
> specific set of operations in your own code.
Nah, no cigar here. Microsoft is too large to have that kind of internal
coordination. Too much turnover, too much publicly-available documentation.
You can't simultaneously assume that they're so good that they can do
this and so bad that they write unusable software.
Bob Munck
____________________________________________________________________
--------------------------------------------------------------------
Join The Web Consultants Association : Register on our web site Now
Web Consultants Web Site : http://just4u.com/webconsultants
If you lose the instructions All subscription/unsubscribing can be done
directly from our website for all our lists.
---------------------------------------------------------------------
