Hidden Figures

[Beliavsky]

One of the most popular current movies, "Hidden Figures", has a Fortran angle:

http://www.newyorker.com/culture/richard-brody/hidden-figures-is-a-subtle-and-powerful-work-of-counter-history
HIDDEN FIGURES” IS A SUBTLE AND POWERFUL WORK OF COUNTER-HISTORY
By Richard Brody
The New Yorker
December 23, 2016

The basic virtue of “Hidden Figures” (which opens on December 25th), and it’s a formidable one, is to proclaim with a clarion vibrancy that, were it not for the devoted, unique, and indispensable efforts of three black women scientists, the United States might not have successfully sent people into space or to the moon and back. The movie is set mainly in 1961 and 1962, in Virginia, where a key nasa research center was (and is) based, and the movie is aptly and thoroughly derisive toward the discriminatory laws and practices that prevailed at the time.

[...]

Dorothy’s pursuit of a formal promotion to supervisor also takes place against the backdrop of the civil-rights movement. She learns that her entire department of human “computers” will soon be replaced by an electronic computer—an enormous I.B.M. mainframe that’s being installed. A gifted technician, Dorothy seeks out a book from the local library (a segregated library from which she’s thrown out), in which she’ll learn the programming language Fortran; she soon becomes nasa’s resident expert. On that trip to the library, in the company of her two sons on the cusp of adolescence, they witness a protest by civil-rights activists chanting “segregation must go” and see police officers, with police dogs, approaching the protesters.

[Damian Rouson]

It's a great movie for the whole family. I hope everyone sees it. It's also especially nice to know that Katherine Gobel Johnson, who inspired the main character, is still alive at 98 and NASA Langley is building a new computational sciences facility that will be named after her. Fortran still plays a major role at NASA Langley. I taught a modern Fortran class there in 2014 and Walt Brainerd has taught several modern Fortran classes there.

Several of the people who attended my course work on the FUN3D computational fluid dynamics Fortran package. Here's a quote from a FUN3D description at the URL below:

"FUN3D is widely used to support major national research and engineering efforts, both within NASA and among groups across U.S. industry, the Department of Defense, and academia." The page goes on to mention "A past FUN3D collaboration with the Department of Energy received the prestigious Gordon Bell Prize for outstanding achievements in high-performance computing."

https://fun3d.larc.nasa.gov/papers/Nielsen_SC14.pdf

It's great to see Fortran continuing to play a significant role at NASA Langley more than five decades after the scenes in the movie.

Damian

[herrman...@gmail.com]

On Saturday, January 21, 2017 at 4:41:49 PM UTC-8, Beliavsky wrote:
> One of the most popular current movies, "Hidden Figures", has a Fortran angle:

Great movie! I recommend it to everyone.

(snip)

> Dorothy’s pursuit of a formal promotion to supervisor also takes place
> against the backdrop of the civil-rights movement. She learns that her
> entire department of human “computers” will soon be replaced by an
> electronic computer—an enormous I.B.M. mainframe that’s being installed.
> A gifted technician, Dorothy seeks out a book from the local library
> (a segregated library from which she’s thrown out), in which she’ll learn
> the programming language Fortran; she soon becomes nasa’s
> resident expert.

I did wonder how easy it was to find Fortran books in the public
library in 1961. By the time I was learning Fortran, in 1972, I don't
remember finding much in the library. (In the middle of silicon valley!)

I remember a McCracken Fortran book, though I don't remember
learning much from it. I remember books for Jovial, and other
language that I didn't have a compiler for. I also remember the
book for IBM's ECAP (Electronic Circuit Analysis Program), but
again didn't have the program to use with the book.

[WILSON]

-- Interesting subject. I wrote an introductory Fortran book in 1967 that
was used by some universities. It was one of several Fortran books
written in the late 1960s. (At one time I had a few boxes of Fortran IV
and V tesxbooks, so they were comming out at a pretty fast cliip.) But,
McCraken's book pretty much put all the others, including mine, out of
business. Not sure why; they were all about the same.





Using Opera's revolutionary e-mail client: http://www.opera.com/mail/

[Terence]

On Saturday, January 21, 2017 at 4:41:49 PM UTC-8, Beliavsky wrote:

>I did wonder how easy it was to find Fortran books in the public
>library in 1961. By the time I was learning Fortran, in 1972, I don't
>remember finding much in the library. (In the middle of silicon valley!)

I Joined IBM in 1960 In London, and went straight into the education process
there (on punches, relay-driven machines and room-filling electronic
computers). But after about three weeks of this I was given an internal IBM
manual for Fortran on the 704 and requested to produce a working program on
anything I wanted as a final test.

So at least one manual WAS available in 1960. Actually the Fortran I manual
predates this.

As a Physicist (and Mathematician), I used the 704 Fortran compiler to
calculate all the masses of all the stable and some unstable nuclei (from
the liquid drop model) and created a printer-output character based graph of
calculated atomic weight against atomic number (which I still have, and
probably the program somewhere). I was then appointed the Fortran guy for
the London offices. Next program was auto-designing transmission towers for
IBM client Balfour Beatty and producing a parts list and order of work, from
foundations to tower top and then cable arms.
Terence

[WILSON]

Interesting. I wrote my first computer program in 1957 in some restricted
vesion of Fortran that ran on the IBM 650. Can't rememer its name now.
but it did have a manual. (It first generated a punched card deck of
your program in assembler which then had to be fed back into the assembler
to get a working program.) Tedious.

A year or two later I wrote Fortran programs for the IBM 704 and 709.
Both Fortans had manuals.

So both Terence and I had early Fortran manuals, but I can't imagine IBM
not having a Fortran manual for the 704 almost as soon as the language was
avaialble.

Lee

--

[Norm Clerman]

Mr. Walter Spector and I were able to locate an early IBM report on FORTRAN, dated November 10, 1954; and we used the title sheet as the frontispiece of our book on Fortran, "Modern Fortran: Style and Usage," published a few years ago by Cambridge University Press. Interested persons can see the report at this website:

http://archive.computerhistory.org/resources/text/Fortran/102679231.05.01.acc.pdf

There's not enough detail in it to be considered a manual, but it is one of the first documents about the language.

I have a small, but far from complete, collection of old FORTRAN books; not one dates prior to 1961.

Norm

[herrman...@gmail.com]

On Sunday, February 12, 2017 at 9:58:40 PM UTC-8, WILSON wrote:
> On Sun, 12 Feb 2017 22:48:19 -0500, Terence <tbwri..t@cantv.net> wrote:

(snip, I wrote)

> >> I did wonder how easy it was to find Fortran books in the public
> >> library in 1961. By the time I was learning Fortran, in 1972, I don't
> >> remember finding much in the library. (In the middle of silicon
> >> valley!)

> > I Joined IBM in 1960 In London, and went straight into the education
> > process there (on punches, relay-driven machines and room-filling
> > electronic computers). But after about three weeks of this I was given
> > an internal IBM manual for Fortran on the 704 and requested to produce a
> > working program on anything I wanted as a final test.

(snip)

> Interesting. I wrote my first computer program in 1957 in some restricted
> vesion of Fortran that ran on the IBM 650. Can't rememer its name now.
> but it did have a manual. (It first generated a punched card deck of
> your program in assembler which then had to be fed back into the assembler
> to get a working program.) Tedious.

I was wondering about books not published by a computer
manufacturer. As an example, the McCracken books.
It does seem that McCracken goes back to 1961, but were
they in public libraries that far back?

I remember McCracken books from the library in 1972, but
I already had the IBM manual by then, and found it much easier
to learn from. (Seems that I am unusual in that one.)

From: https://en.wikipedia.org/wiki/Hidden_Figures#Historical_accuracy

it seems that some of the timeline is off.

[herrman...@gmail.com]

On Monday, February 13, 2017 at 6:43:09 AM UTC-8, Norm Clerman wrote:

(snip)

> Mr. Walter Spector and I were able to locate an early IBM
> report on FORTRAN, dated November 10, 1954; and we
> used the title sheet as the frontispiece of our book on Fortran,
> "Modern Fortran: Style and Usage," published a few years ago
> by Cambridge University Press.

The manual usually considered the description of Fortran I
(before it was numbered I) dates to 1956.

I presume that the 1954 manual is what Knuth calls Fortran 0.
That is, the language as defined before they started writing the
compiler for it.

[Beliavsky]

On Sunday, February 12, 2017 at 10:48:27 PM UTC-5, Terence wrote:
> On Saturday, January 21, 2017 at 4:41:49 PM UTC-8, Beliavsky wrote:
>
> >I did wonder how easy it was to find Fortran books in the public
> >library in 1961. By the time I was learning Fortran, in 1972, I don't
> >remember finding much in the library. (In the middle of silicon valley!)
>
> I Joined IBM in 1960 In London, and went straight into the education process
> there (on punches, relay-driven machines and room-filling electronic
> computers). But after about three weeks of this I was given an internal IBM
> manual for Fortran on the 704 and requested to produce a working program on
> anything I wanted as a final test.
>
> So at least one manual WAS available in 1960. Actually the Fortran I manual
> predates this.
>
> As a Physicist (and Mathematician), I used the 704 Fortran compiler to
> calculate all the masses of all the stable and some unstable nuclei (from
> the liquid drop model) and created a printer-output character based graph of
> calculated atomic weight against atomic number (which I still have, and
> probably the program somewhere).

This is a cool newsgroup. Long live Fortran!

[Clive Page]

On 13/02/2017 14:43, Norm Clerman wrote:
> Mr. Walter Spector and I were able to locate an early IBM report on FORTRAN, dated November 10, 1954

[snip]
> http://archive.computerhistory.org/resources/text/Fortran/102679231.05.01.acc.pdf

That is a wonderful document. I particularly like the start of the
paragraph on page 2:

"Since Fortran should virtually eliminate coding and debugging, it
should be possible to solve problems for less than half the cost that
would be required without such a system".

I guess by "coding" they meant assember-coding so it's true, but if only
their hopes over debugging had been realised...


--
Clive Page

[Dan Nagle]

Hi,

On 2017-02-13 14:43:08 +0000, Norm Clerman said:
>
> http://archive.computerhistory.org/resources/text/Fortran/102679231.05.01.acc.pdf
>

> Norm

Thanks for posting the link.

I find it interesting the Jean Sammet's name is hand written on the cover.
There's some real history here.

--

Cheers!
Dan Nagle

[baf]

Was the notation

do 10, 14, 50 i=4, 20, 2

used in the "final" form of FORTRAN I? I had never seen that before,
where the "formula" numbers that make up the body of the loop are
specified in the do statement.

[Terence]

I've never seen anything like that syntax, not even when I studied the
Fortran I manual with its sense switches and sense lights.
I was then comparing what I was used to in Fortran II in case IBM customers
asked for help in updating earlier programs.
That structure reads like the use of an interpretive language like an
early-version B.Gates' Basic.
Terence

[chris...@gmail.com]

awesome thread, thanks computer pionners

[brettc...@gmail.com]

Why doesn't anyone comment on the size of that FORTRAN book. It looks like it is 100 pages tops. I would read it. I joke with my wife (also a CS engineer) that it is the most unrealistic part of the whole movie. Most text books now-a-days are 600 pages. lol.

[herrman...@gmail.com]

On Monday, July 10, 2017 at 7:20:53 PM UTC-7, brettc...@gmail.com wrote:

(snip)

> Why doesn't anyone comment on the size of that FORTRAN book.
> It looks like it is 100 pages tops. I would read it.

I was, above, wondering what was available in libraries at
that time.

Though the movie gets some of the time line wrong, compared
to the book. For example, the restrooms were integrated long
before that part of the story in the movie.

> I joke with my wife (also a CS engineer) that it is the most
> unrealistic part of the whole movie.

No, the most unrealistic was moving the oscilloscope probe
to the right place, without at least looking at the diagrams
which have the pin numbers on them. Then careful counting out
the rows and columns of the pins to find the right one.

> Most text books now-a-days are 600 pages. lol.

The Fortran standard isn't all that long. The McCracken
books are larger pages, but not so many pages. Fortran
has gotten a lot more complicated in the years since.

[herrman...@gmail.com]

https://www.amazon.com/Guide-Fortran-Programming-Daniel-McCracken/dp/0471582816

says 304 pages.

[jfh]

On Tuesday, July 11, 2017 at 5:51:25 PM UTC+12, herrman...@gmail.com wrote:
> https://www.amazon.com/Guide-Fortran-Programming-Daniel-McCracken/dp/0471582816
>
> says 304 pages.

That was McCracken's Fortran IV book. His earlier book, on dialects available in 1961, was "a guide to FORTRAN programming", viii+88 pp.,Wiley. Lower and upper case as on the title page and front cover.

[herrman...@gmail.com]

On Tuesday, July 11, 2017 at 3:07:18 PM UTC-7, jfh wrote:

(snip, I wrote)

> > https://www.amazon.com/Guide-Fortran-Programming-Daniel-McCracken/dp/0471582816

> > says 304 pages.

> That was McCracken's Fortran IV book. His earlier book,
> on dialects available in 1961, was "a guide to FORTRAN
> programming", viii+88 pp.,Wiley. Lower and upper case as on
> the title page and front cover.

And 8.1 by 10.9 inches, so a little bigger than I remember from
the movie. But yes, I did want to know what was available
in 1961.

https://www.amazon.com/Guide-Fortran-Programming-Daniel-McCracken/dp/0471582123

[Terence]

jfh wrote:

>And 8.1 by 10.9 inches, so a little bigger than I remember from
>the movie. But yes, I did want to know what was available
>in 1961.

Yes, I saw that 'movie' and on my first visit to the 'states, mid-sixties,
had the experience of the city bus driver slamming on the brakes after I
boarded and walked down the isle to a vacant seat, to be told "you can't sit
there! Come up here or I'll have to drive to the police station!" .

After two degrees I had already worked in UK 'rocketry' myself and after two
more for IBM for some years and travelled a lot, but had NO IDEA then of
that society at that time! It didn't filter through.

I loaned a friend my first yellow IBM FORTRAN Manual (unfortunately), but my
small compact Burroughs BTOS FORTRAN Manual (almost F77) has only 66 sheets
on the language; a few more on the compiler and error code lists.

Terence