My efforts to track down source code and documents from the original IBM 704 FORTRAN project have been one of the pilot projects of the Software Collection Committee at the Computer History Museum. I’m starting to assemble a web site at the Museum to organize and present the materials I’ve collected so far. I’d appreciate your comments regarding both the form and the content.
[Edited 10 May 2014: community.computerhistory.org/scc => softwarepreservation.org. Note the Software Collection Committee was renamed the Software Preservation Group.]
I was one of the original IBM Customer Engineers that installed 704 serial 13 at Lockheed Aircraft in Marietta, GA in May 1956.
When Fortran was released in April 1957 we were one of the first customers to run it. We soon ran into troubles where compiler would print out the message “Machine error” on the 716 printer. To solve this problem we needed a source listing of the compiler. It took the Branch Manager several days to get one for us and it was on a roll of 35 mm microfilm. This had to be under IBM control at all times and when we finished the bug it was sent back to the Branch Office. I don’t know what happend to that roll but I just wanted to point out that a listing did exist in that form at that time.
If you are interested send me an email address and I will send you a two page Word document from my memoirs describing this period.
Van Gardner
Smyrna, Ga.
I first learned FORTRAN in the Summer of 1961 while working at Pratt & Whitney in West Palm Beach, FL working on the RL-10 liquid fueled rocket engine.
I now teach a course called Scientific Computing at the University of Maryland Baltimore County (UMBC).
The course is for those students majoring in Physics and Chemistry.
I have a Bachelors degree in Aeronautical Engineering and a Master of Science in Physics (minor in Math) from the University of Florida in Gainesville, FL.
I will be leaving UMBC at the end of June 2005. I hope to be taking a position at NorthropGrumman as a Supervisor of their Fortran group.
Waiting to hear from them as their HR people met today (May 17) so I should hear from them soon.
The memoir that Van mentioned in the first comment above is now available online — see http://community.computerhistory.org/scc/projects/FORTRAN/index.html#Memoirs_and_user_stories .
Paul, you might check with the University of Toronto. I remember running a Fortran II compiler “UTO Fortran” (I think it was) on the paper tape driven IBM 1620 at the University of the South, Sewanee, TN, in the fall 1968 through fall 1970.
As a summer intern, I remember using UTO Fortran II on a CARD ORIENTED 1620 system at the New England Electric System office in downtown Boston. On that system the UTO compiler was loaded from 80 column cards. Following the compiler, the user’s program (also on cards, of course) was loaded. The 1620 then processed the source cards and (hopefully) punched out a deck of object code. Finally, the object code was reloaded along with a deck of cards containing the UTO library. After this rather time-consuming process, results from the user’s program (if any) were punched onto yet another stack of 80-column cards. THESE cards were then printed via a 407(?) “accounting machine”! If I remember correctly, this whole process could easily take 1/2 for a short program. BTW, this 1620 only had 20K words of memory…
Everyone using the 1620 was extremely happy when it was replaced with an IBM 1130, although cards were still used (of course!) for input.
Wayne
Probably 1963/1964 I used UTO Fortran on an IBM1620 Model 1. Paper tape I/O and teletypewriter. This 1620 was a 4k BCD machine. Each BCD contained a single character.
System was installed at the University of the Witwatersrand, Johannesburg. Department was headed by the late Dr Derek Henderson.
Debugging involved splicing paper type. Seems such a