It’s a good thing I referenced the year! Nice find, thanks!
2 Likes
Certik,
I think I replied to Arjen Markus, concerning a book or pamphlet he is creating concerning old FORTRAN conventions. I couldn’t remember at that moment what I was using back in 1975, I think Fortran IV with WATFIV complier. I get mixed up with the 4 and 5 of those acronyms. WATFIV was the University of Waterloo compiler, I guess software version 4, we had smaller numbers to remember back then. I wish I still had the user manual. Memory tells me it was likely printed on fan-fold paper with holes on either side for the sprocket drive to advance the paper, with perforations to tear off the hole strips. The “book” was a cheap production with big, exposed staple binding for the two post-board outer covers. As you mention, this would likely have been FORTRAN 66 language.
If you think I might be of help to you, I have a program example from Fall 1975. I and a colleague of mine wrote a Finite Element program to solve a linear stiffness 2D structural frame. I have the full source, 10 pages printed of course in caps on fanfold paper, hole strips torn off. The input/output might be about another 10 pages. The pages have pin stripes alternating white and light green, each stripe containing three printable lines. I also have with this an input example and its corresponding output. The source code is understandably left-justified to column 7 and we used “/” in column 6 for our continued lines. Columns 1-5 of course held our statement numbers. Hard to believe this was 50 years ago.
If you express interest, I would be happy to scan this example (source, input/output) and send it to you, I assume by some means of secure communication on this site (drop box?). Also, based upon your interest I can do further research in my files for additional nuggets of FORTRAN examples, some of which contain date stamps from the printer and mention of the WATIV which I notice on this example is labeled: “WATEIV”. Maybe I am wrongly calling it WATFIV, the printout is pointing to IV, not V.
Hi @TomBeam thanks for the offer. I think it would be great if you could post the program online. Especially if you could open source it, then you can just post it online, and we can try to get it working.
Ondrej (is that like Andre’ ?)
I assume that you mean by “Post” to post the scan somewhere on the “Fortran Discourse” site?
By open source, do you mean that I should make the scan file be editable? I don’t have the original source file, just a listing which I would be able to scan to a file of the image of source code lines and post that for you, even as a part of my own post on Fortran Discourse.
Please bear with me, I am not as savvy with posting on the site.
Thanks
Tom Holderread
I’m considering using an OCR (image to text converter) to render the FORTRAN source code that we discussed as editable text.
These historically have not been very accurate, mixing digit 1 with lower case “l” and digit 0 with upper case “O” and the “rn” pair with “m” and so on, and requiring a lot of manual corrections. I wonder however if modern AI can step in nowadays to do that last step?
Ron,
Thanks for your input! Historically, my own experimenting with 1980’s era OCR showed it to be very finicky with artistic or “custom” fonts. It was also notoriously dumbfounded by printouts done on the old dot-matrix printers which created a font with poor resolution that resembled the font of the old “ticker-tape”. I wouldn’t be surprised if someone has not invented AI-assisted character recognition of late. It sounds crazy, but I used to enjoy doing manual post-processing of the results so I could pretend to be a cryptologist. I’ll soon see what luck I have with OCR on the code that I am preparing for Certik. It has historic plus sentimental value to me.
Hi @TomBeam, yes, my name is the same as Andrew. Yes, I meant if you could post it anywhere online and post a link here.
Yes, by opensource, I meant if you could license your code under MIT or BSD licenses (ideally).
Yes, scanning it is fine.
I am a big fan of archiving old code and making it available. It would be cool if we can compile it and get it working again.
Let me know if you have any other questions.
1 Like
I placed (4) files on dropbox.com but I don’t know how to run their software. I am on the “basic” plan since I don’t intend to put huge amounts of data up there and it is free to the best of my knowledge. My source code is in the file with the extension .f which I think identifies it as FORTRAN language of fixed length “rules”. The files are marked as “only you” which suggests that only I can get access to them. I don’t know how to make them generally available so that you and anyone else on this site can freely access them which I am OK with.
The Fortran source is about 500 plus lines of code. I have included a sketch of a sample 2D structural building frame for testing the program. Also included on Dropbox is a sample input file and corresponding output file, both referring to the sketch of the “building frame”. I am not sure how to proceed.
My source code includes the MIT Copyright statement at the top.
I was not successful with my OCR program and ended up re-typing about 500 lines of source code. But I’m a hyper-typer so the project went fast and is done.
1 Like