Hear, hear, Bruce!

There are lessons to be learned from * some * of the old hands at this,
such as the disciplines that come from (gasp) a mainframe environment.
Go get the 20th anniversary edition of "The Mythical Man-Month", and
read the essays on No Silver Bullet. Things have changed quantitatively
more than qualitatively. The 'new ideas' largely grew out of the old
ideas. They have added a little, sometimes something valuable, and
almost always change the language radically, so that someone else cannot
understand it. Read Object Orientation: An Introduction for COBOL
Programmers. We old COBOLers already know most of the concepts, we just
use different names. But then, at 34, and self-taught, I bet I know more
lingo than the 20 year old, and guess what? I speak both, and can even
translate. What do you know? Mainframes? UNIX? NT? Macintosh? I can
explain the 3000 in terms you can understand. Oh, and before I had a
family and could afford to live on what they pay teachers, I taught
classes, too.

I won't even ask about familiarity with algorithms... But if you are
British Telecom, processing 60 billion transactions per day, and can
save one second per million transactions for a single process, that's 16
minutes, 40 seconds. This can add up to substantially improved
processing in a given business day.

I am afraid of systems written by three college kids in a garage or a
lab, that cannot tolerate last year's PC because it is "too slow", and
does unexpected things when 200 users sign-on all on the same day, and
cannot be maintained by the now 25 year old or the latest crop of 20
year olds. I would rather be unemployed because my employer moved away
from COBOL than because the system I designed in some $99.95 (CUE price
$94.95) 5GL suffered a complete meltdown, and all the King's horses and
all the King's men could not put Humpty Dumpty back together again.

> -----Original Message-----
> From: toback2 [SMTP:[log in to unmask]]
> Sent: Friday, April 10, 1998 7:53 PM
> To:   [log in to unmask]
> Subject:      Re: [HP3000-L] Cort and the HP 3000
>
> That's an interesting idea, but it represents a very naive point of
> view.
        <snip>