25th Machine Records Unit, HDQ FECOM.
Duties
On Sun, 12 Jul 1998
Robert Humble Email address wrote:
25 Machine Records Unit
datein: 1949 dateout: 1970
comments:
Arrived at Yokohama on October 3, 1950. Original destination was Pusan,
but the Inchon invasion changed the Army's plans.
The unit was split between a DS group with the 96th MRU in Tokyo and a
group sent on to Seoul to prepare for later location there. After the
Chinese intervention the unit finally was brought together in Taegu.
The unit was engaged in strength accounting using card files and IBM
tabulating equipment. Doesn't sound combative, and it wasn't, but it
was a valuable service to the Army and to all us soldiers who hoped to
go home someday. I was with the unit from Ft. MacPherson to DS in Tokyo
(Finance Building) to Taegu and home in July 1952.
Activities
The IBM machines involved were:
The Keypunch.....operated from a typewriter like keyboard, punched the
little rectangular holes in the cards.
The Interpreter.....read the holes and printed out the information
across the top of the cards.
The Sorter.....sorted decks of cards into alpha or numerical order.
The Reproducer.....reproduced cards in whole or in part and also could
gang punch data into entire decks of cards.
The Collator.....merged 2 decks of cards to form one larger deck in the
same sequence.
The Tabulator.....did the calculating and printed out reports.
The machines we had were electro mechanical in their workings. They were
of WW II or early post war vintage. The first electronic machines I saw
were sorters that showed up in Japan in late 1950 or early 1951. They
used vacuum tubes. They were blazingly fast :-). Or so it seemed then.
Graduates of the AGC Tabulating Machine School at Ft. Benjamin Harrison
or later at Camp Lee were assigned to the machine records units. There
they became either machine operators, MOS 0400 later 3400, or analysts
0402 later 3402. I was an analyst.
The input document for the files was the Morning Report. The MRU is
where the first copy of the Morning Report went. Every unit every day
submitted a Morning Report detailing whatever changes had taken place
in the previous 24 hours in whatever elements of information the Army
specified at a given time. Gains, losses, promotions, demotions,
casualties, changes in MOS, etc.
On receipt at the MRU the Morning Reports were coded to reflect whatever
change was of interest at the moment and sent to keypunch to be typed
into what were called change cards.
Coding was done in red pencil, it being said that red did not photograph
when the Morning Reports were finally put on microfilm.
The change cards then went to the analyst who matched the change card to
the individual's card in the unit file. The matched pair were sent to
the machine room where the change was incorporated into a new individual
card which was sent back to the analyst who refiled it in the unit file.
At the end of each month and as of the last day of the month a roster
was prepared, by machine, and sent to the unit. The unit was supposed
to review the roster for accuracy, make whatever corrections were
necessary and return it to the MRU. Any changes were then incorporated
into the individual cards.
The information in the cards included name, rank, serial number, MOS,
arm or service, perhaps ddzi (date departed zi) or deros (dated expected
return from OS), type of last change and date of last change.
Not much, compared to what you would expect today where I expect
everything down to shoe size is in the computer. But, you can see the
value of the information to the higher commands. Like, how many riflemen
in Company K of a given regiment, what rank are they, and how long
before the go home.
There were many other specialized reports on information of interest to
the commands. Like casualty reports. It should be understood that what
we were dealing with were statistics. The soldiers actual records were,
of course, still maintained by his unit or another echelon of his
command.
The Army as far as our operation was concerned did not care what a
soldiers name was, but about where he was, what he did and so forth.
Some of the things that analysts did could have been done by machine.
However, it turned out that the analyst, while much slower, was much
more accurate. For example, if the name Smyth was misspelled as Smith
the analyst would find it where the machine wouldn't. Or if one or two
digits of a serial number were incorrect the analyst would still find
the individual in the file.
An analyst would handle files of several thousand cards. I, for example,
in Korea had the file for the 25th Infantry Division, the 5th Regimental
Combat Team, and the associated units. Probably over 20000 cards.
It was considered vital to get the month end rosters out at the end of
the month. When we arrived in Japan and began working in the 96th MRU
they were more than a month behind in processing changes. We worked on
the files that would go to Korea with the 25th MRU about 12 hours a day,
7 days a week, for months to catch up. The accuracy of the files was
sadly deficient initially. I'd say that, as from October 1950, we were
pretty well up to speed as to processing changes and accuracy by the
fall 1951.
Of course, some units were worse than others. Some had to hold physical
musters to find out who was there. Some units disappeared in the Chinese
intervention and had to have the records reconstituted later. Very
difficult.
I am still in touch with several other members of the 25th MRU. It was
my first permanent military organization and I have a soft spot in my
heart for it and for the guys in it. There have been a couple of
reunions which I could not attend. I've received pictures from the
reunions, though.
Wouldn't recognize anyone.
