In <[log in to unmask]> [log in to unmask] writes:

> > Well, I guess its war story time.
> >
> > My introduction to the HP3000 came in 1977 with the Series II (I knew
> > I could gray-hair you guys).

Ok, Nick's post got me reminiscing, and reminded me of a few stories that,
while they may not be "success stories" per se, they do fall into the "it's
a small world category".

I first worked on a 3000 in 1977; a series II at the high school I was at
(and later worked for). We had the 2-bay model (very state of the art at
the time) with 128Kb main memory, later upgraded to 256k I believe -
probably when we upgraded it to a Series III. A couple of 25Mb and 50Mb
disc drives and two VERY flakey 2617 printers (HP had OEM'd from Data-
Products). We had a bank of 110/300 baud modems for all the other high
schools in the southern half of Prince Georges County Md, who all did their
course scheduling, grades, and attendance on the system (about 12-19 schools
if I remember correctly; it was one of the largest school systems in the
country). The computer-science/math classes used a combination of 2640A
and 2640B terminals (with about 1/2 page of memory), and IBM keypunch
machines to feed programs on card decks into our HP punched-card reader.
We also had those ($#@!) optical mark-readers that HP also OEM'd that ate
about as many grade/attendance cards as they read. Of course, one of our
favorite toys was the old "peekaboo" program that allowed you to watch
a device and see what was typed in keystroke by keystroke. It also let you
type your own message and wrote it directly to the device (with no From:
info...) "heh heh heh" to quote Art Bahrs...

Getting hold of the source to Peekaboo at some point was one of our
highlights, and I spent long hours trying to decipher it (very little
comments, lots of assembly subroutines, and the only clue to the author
was the line "written by the ESR Phantom"). After acquiring lots of internals
info (SEs back then were our only links to things internal, and some of
them would allow us to "borrow" their holy grail of documents; the MPE/V
system tables manuals...which rapidly made the rounds among the other
programmers I knew) I finally did figure it out well enough to both make
some minor improvements, and use some of it's code as a basis for later
systems programming tasks.

> > I was assigned an HP SE, Bob Chafin.  I believe he still works handling
> > the
> > huge amount of equipment at the HP Atlanta facility. Our design involved

I met Bob Chafin at a job interview with HP in the early 80s. At the time
he was the manager of the datacenter at HP Rockville. I had "SPL" on my
resume as one of the languages I was fluent in, as well as HP assembler.
Intrigued, he gave me a "test" on the spot, asking me to write him an HP3000
assembly program. When I did, he promptly corrected a minor goof, and we
proceeded to "talk internals". After a while he asked me if I'd ever heard
of the "ESR Phantom"... I confessed that I had - at least seen his work.
Bob then confessed that he was, indeed, the ESR Phantom. He had written
peekaboo some time ago when he was doing some other internals work. I was
duly impressed.

I'll never forget when the school system upgraded the Series II to a Series
III. The upgrade was supposed to take a "couple" hours. Our CE was pretty
good (Dwight Springer - who later became a trainer at HP Rockville).
Unfortunately, we began the update during a rainstorm at about 4pm (just
after the schools had let out) which shortly turned into a tremendous
lightning storm. There was no UPS or power protection in the school's
computer center; so power started going on and off. The system update
that went along with the update had to proceed through uncounted power
brownouts and periodic blackouts; turning the upgrade into a 13 hour
ordeal. But it came out tickin.

Even more fun were the D.C. summers when the air conditioning system in
the school would go out. Many afternoons we'd have the windows open, the
bay doors on the cpu standing open, thermometer reading well over 100
degrees in the computer room, us with fans in front and behind the system
coaxing it along... It never failed though! Kept right on running. Only
once did we give up and shut it down...which was fine by me - I took the
rest of the afternoon off and went bowling. :-) You just don't get that
many days off around a 3000! ;-)

> > custom hardware.  Our HP rep introduced me to Bob and we told him what
> > the application involved - on line Point-of-sale terminals attached > (some
> > remotely) to the HP3000 through the custom hardware.  At the time the > only
> > inter-system communication was to use the HP1000 as what was called from > t
> > 3000 end a "programmable controller".  Our design involved making our
> > hardware look to the 3000 like a 1000 (the programmable controller).

In the mid-80s, the company I was working at had several HP1000 computers
with software from a company named Forest Computers that acted as interfaces
between our HP3000s and some Navy IBM mainframes. Neat package; it allowed
3000 users from an HP terminal to "log in" to the CICS system and work on
the mainframe, and likewise allowed IBMers to log into the 3000. The neat
part was that the 1000 translated all the terminal escape-sequences and
screen handling stuff so HPers looked like IBM workstations to the mainframe,
and vice versa. Also talked to Univacs and a couple other boxes. Not real
fast (and a few quirks), but the technology was pretty neat at the time.

I bought my first 3000 back in 87. A Series 37 that I got a very nice
deal on. I spent several years working on a package I called "Synch/3000";
an Image database shadowing package that automatically synchronized multi-
ple (any number actually) Image databases using no PM code. I developed
code that monitor Image logging in real time, then picked out updates and
applied them to any number of local or remote databases. About 100,000
lines of SPL I later archived after encountering too many "quirks" in
Image logging (back then). Variations of it still run today at one or two
sites though.

After the company I worked for got a connection to the "DDN" (aka Milnet
or Arpanet) back in about 1989, I became frustrated that our 3000s could
talk to each other over the 'net, but not to anyone (or anything) else.
I started reading the DDN handbooks (actually voluminous manuals that
filled most of a bookshelf) where the secrets of the 'net were documented
(the now famous, then barely known, RFCs). I'll never forget the article
Alfredo Rego wrote some time ago about the proper way to read a manual;
I can tell that he's another man that can truly appreciate a good reference
manual! ;-)

Anyway, I started designing and testing an SMTP based e-mail system for
the HP3000, that later evolved into NetMail/3000. Then with Jeff Kell's
encouragement started on DeskLink - the SMTP/MIME gateway for HPDesk.
With the help of several others since, we've gone on to add Pop2 and Pop3
servers, gopher servers, and have several other Internet-related goodies
in the pipe for 3000s.

It's a small world. And getting smaller all the time.

         -Chris Bartram