On Fri, 12 Feb 2016 09:22:40 -0800, Califbill billnews wrote:

wrote:
On Thu, 11 Feb 2016 19:26:58 -0800, Califbill billnews wrote:

wrote:
On Thu, 11 Feb 2016 20:14:49 -0500, Keyser Söze
wrote:

On 2/11/16 8:08 PM, wrote:
On Thu, 11 Feb 2016 19:35:56 -0500,
wrote:

On Thu, 11 Feb 2016 14:18:34 -0500, Keyser Söze
wrote:

I was writing the user manuals for
minicomputers being sold to the Peoples Republic of China to aid in
weather forecasting for agricultural programs. Well, at least that was
their ostensible purpose.

===

Whoa, there's another Harry Tale we haven't hear before. Did the
Chicoms use the computers to develop nuclear weapons? That would be a
nice embellishment.

I am having a little trouble with the time line. 1965 in Bethesda?
When was he at that Kansas college and the paper? How old is this guy?


Where did 1965 come from? This was a decade and a half, nearly, later.

By 1980, we had pretty much run all of our competition out of town.
Certainly Burroughs and Control Data.





Between the name and great salesmen, IBM did very well. We lost Macy's
California POS business to you guys. Our salesman said they would have to
run the data cables in conduit over the fluorescent fixtures to the
terminals. IBM guy told them did not need to. He knew the fire department
required that, and would be done, wether needed or not. Smart guy.

I never understood that. It is not required by the NEC if you use the
right cable (CLP if it is a plenum) and there is certainly no
interference problem. (in spite of the urban legends). We tested it
six ways from Sunday and I had customers all over town who had cables
sitting to troffers.
I had a coil of wire sitting on the 277v light over my desk, connected
to the token ring I was using. (IBM type 1 cable) We also tried it
with ethernet, store loop, broadband and 3270. That is really easy to
do if you have the IBM structured wiring network because you just need
to swap baluns on each end and plug in the right machines.


We blew away the enterprise competition simply with price and the
response time we used to provide. IBM dropped the "off shift" price
hike and if you had a lease on an MA, it was all included in your
basic contract. That blew away the MAIs and other maintenance
competitors. We were very close to their prime shift price, giving
them 24/7 service. When I moved down here I killed off the only 3d
party maintainer we had here in about a month.

We also came out with very competitive pricing on the hardware. They
really worked on cost reduction in the design and it actually improved
reliability. The last water cooled (3090) processor complex I had, ran
5 years without a single call. The log read "Install" ...
"Discontinue" with a few PM entries, mostly replacing filters.
That was when I knew I needed to look for a different job and I got
into installation planning, connectivity, availability, contract
services and service delivery. The service delivery deal was the most
interesting, basically database analysis. That was pretty much all
self taught.
IBM still spent a lot of money on schools for me the last 5 years I
was there and I was doing my inspector training on the side.
In the end the decision was whether I wanted to work in Ft
Lauderdale/Miami and commute or retire. I retired.



This was mid 1970's, and I seem to remember was a 3 wire connection. I
figured the noisy old style fluorescent would generate enough noise to
induce errors. As to noisy electronics, in the 1980's working for System
Industries, we had to ensure our disk and tape controllers met FCC noise
requirements. One of the testing engineers brought in a touch control lamp
his wife had just bought. Touch the lamp to turn on or control brightness.
Since it was not commercial computer equipment, I guess it did not have
meet standards. Was about a 1000x noise of the requirements we had to
meet,

The trick was to use shielded twisted pair. That was very noise
tolerant. In fact regular old STP (Cat3) is pretty good against most
stuff.

On Fri, 12 Feb 2016 13:59:31 -0500, Keyser Söze
wrote:

Sorry, I don't keep your CV handy.

Your ignorance of the facts never seems to keep you from talking about
lots of things.


Send me your C.V., and I'll tack it to my wallboard.

How much detail do you want? I suppose the number of classroom hours
alone would make your degree(s) pale by comparison. It is well over
15,000.

wrote:
On Fri, 12 Feb 2016 09:22:40 -0800, Califbill billnews wrote:

wrote:
On Thu, 11 Feb 2016 19:26:58 -0800, Califbill billnews wrote:

wrote:
On Thu, 11 Feb 2016 20:14:49 -0500, Keyser Söze
wrote:

On 2/11/16 8:08 PM, wrote:
On Thu, 11 Feb 2016 19:35:56 -0500,
wrote:

On Thu, 11 Feb 2016 14:18:34 -0500, Keyser Söze
wrote:

I was writing the user manuals for
minicomputers being sold to the Peoples Republic of China to aid in
weather forecasting for agricultural programs. Well, at least that was
their ostensible purpose.

===

Whoa, there's another Harry Tale we haven't hear before. Did the
Chicoms use the computers to develop nuclear weapons? That would be a
nice embellishment.

I am having a little trouble with the time line. 1965 in Bethesda?
When was he at that Kansas college and the paper? How old is this guy?


Where did 1965 come from? This was a decade and a half, nearly, later.

By 1980, we had pretty much run all of our competition out of town.
Certainly Burroughs and Control Data.





Between the name and great salesmen, IBM did very well. We lost Macy's
California POS business to you guys. Our salesman said they would have to
run the data cables in conduit over the fluorescent fixtures to the
terminals. IBM guy told them did not need to. He knew the fire department
required that, and would be done, wether needed or not. Smart guy.

I never understood that. It is not required by the NEC if you use the
right cable (CLP if it is a plenum) and there is certainly no
interference problem. (in spite of the urban legends). We tested it
six ways from Sunday and I had customers all over town who had cables
sitting to troffers.
I had a coil of wire sitting on the 277v light over my desk, connected
to the token ring I was using. (IBM type 1 cable) We also tried it
with ethernet, store loop, broadband and 3270. That is really easy to
do if you have the IBM structured wiring network because you just need
to swap baluns on each end and plug in the right machines.


We blew away the enterprise competition simply with price and the
response time we used to provide. IBM dropped the "off shift" price
hike and if you had a lease on an MA, it was all included in your
basic contract. That blew away the MAIs and other maintenance
competitors. We were very close to their prime shift price, giving
them 24/7 service. When I moved down here I killed off the only 3d
party maintainer we had here in about a month.

We also came out with very competitive pricing on the hardware. They
really worked on cost reduction in the design and it actually improved
reliability. The last water cooled (3090) processor complex I had, ran
5 years without a single call. The log read "Install" ...
"Discontinue" with a few PM entries, mostly replacing filters.
That was when I knew I needed to look for a different job and I got
into installation planning, connectivity, availability, contract
services and service delivery. The service delivery deal was the most
interesting, basically database analysis. That was pretty much all
self taught.
IBM still spent a lot of money on schools for me the last 5 years I
was there and I was doing my inspector training on the side.
In the end the decision was whether I wanted to work in Ft
Lauderdale/Miami and commute or retire. I retired.



This was mid 1970's, and I seem to remember was a 3 wire connection. I
figured the noisy old style fluorescent would generate enough noise to
induce errors. As to noisy electronics, in the 1980's working for System
Industries, we had to ensure our disk and tape controllers met FCC noise
requirements. One of the testing engineers brought in a touch control lamp
his wife had just bought. Touch the lamp to turn on or control brightness.
Since it was not commercial computer equipment, I guess it did not have
meet standards. Was about a 1000x noise of the requirements we had to
meet,

The trick was to use shielded twisted pair. That was very noise
tolerant. In fact regular old STP (Cat3) is pretty good against most
stuff.


Yup, and we had opto-isolator. But those were stated requirements. The
IBM sales dude knew the law, and knew it would be in conduit per fire regs.