1960s IT department | Hacker News Reader

1960s IT department

luckham.org

228 pointsanigbrowl50 comments

A few stories behind the pictures:

- The programmer is probably playing guitar because he is waiting for his compile. It had to be batched and returned, sometimes hours later. Imagine getting things done with only one compile per day.

- The programmer is giving an obscene gesture to the terminal because he forgot a semicolon (more likely a period in his COBOL program) and will now have to wait another day for a clean compile. Move that deadline back to October.

- The terminal the programmer is giving an obscene gesture to is a state-of-the-art Datawriter, a PAPER-DRIVEN terminal. A moment of silence is needed for all the trees sacrificed for our future.

- The programmer was special because he had access to the Datawriter, which was probably in the computer room. Anyone else, including all users, had to use pencil and special forms (like taking their SATs) to enter data into the IBM Mainframe. Those forms went to the Data Control Unit where teams of mostly women keypunched the data onto 80 column cards which were fed into a hopper. The early keypunch machines put the holes in the cards as you typed, so that if you made a mistake, you had to start over with a new card. Later, these machines had memory, so you could finish the virtual card, and hit a key that did the whole card. Probably saved a lot of paper.

- Notice all the coats hanging in the break room. It may have been winter, but just as likely, they were needed in the computer room. It was cold in there!

- Note that the "Data Terminal" pictures are at the bottom of the page in the mini-computer section. This was a big deal back then. Mini-computers got CRT displays before IBM mainframes. That's how they competed with Big Blue. For programmers, this was as big an advance as we would ever get. Imagine building that web app today, buying cases of cards from Office Depot and getting one compile per day.

- The Demonstration Center was a big deal, too. Several places I worked had the computer room behind glass in the lobby. Companies wanted their customers to physically see how advanced their technology was.

Thanks for the memories. Now burn those pictures.

I'm old enough that my Mom was a keypunch operator and would bring me home bags of "square confetti" from the chads punched out of cards and also discarded keypunch cards to play with.

I may not have written my first program until I was 16-17, but I read my first IBM manuals around age 8 or so :-)

they were needed in the computer room. It was cold in there!

The other reason I liked going with her when she had to work Saturdays: it was always air-conditioned.

Me too. All my first drawings are crayon-on-punchcard. This would have been ‘83,‘84, guess my Dad’s computing department we somewhat behind the times!

Me too! Punch cards and punch tape.. rolls and rolls of it. And for me, it was 88-89, because Russia was (and still is) a backwards country.

So wait, you're bitching that programmers were forced to think and to proof-read?

Using a pen + paper to write programs isn't a huge deal and I'd do it more often if it weren't for the retarded amounts of API docs you have to wade through nowadays before writing a complete program.

So wait, you're bitching that programmers were forced to think and to proof-read?

1. On any given day, how many times does a compile (or interpretation) of your work fail?

2. On any given day, how many times does a test run of your program give anything other than the most desirable result?

If your answer to either question is less than 2, then you're either (a) not working enough or (b) the greatest programmer in the history of the world.

I was not "bitching". I was pointing out that in those days you usually got "one shot per day". Think about how this one little fact affected everything you did.

There was no such thing as "Release early and often," "Ready, Fire, Aim," "Iterate, iterate, iterate," "Prototype," or "Get User Feedback ASAP." The existing technology simply didn't permit it. When you got one compile per day, you damn well better spend 6 hours per day proofreading. How many hours per day to you spend proofreading? If it's more than .5, then you're wasting your time. It is your morale imperative to use technology to do that for you.

We laugh at those pictures and those days, but forget how hardcore you really had to be. We're all standing on the shoulders of giants. Respect.

I often go days or weeks between successive compilations of my programs. Not everyone programs in the "Ready, Fire, Aim" style, nor does everyone agree that it is the best.

Love how everything had color back then. Bright red Minicomputer, Aquamarine phones, blue typewriter. My cube consists of varying shades of grey...

Hey at least your Swingline stapler is red. ... Uh... Where did your stapler go?

Dark wood veneer desk, black keyboard and monitors (and computer under the desk), and taupe walls, here.

Not bad, but gear that's not grayscale would be pretty cool.

Mind, then we'd have to worry about whether our keyboards and monitors matched...

Boy, that brings back fond memories of IBM 360 mainframes in big, noisy, cold rooms.

I remember the transition from cards to online terminals, when hacking (as a high-school junior) at the Naval Electronic Labs computer center in Point Loma (San Diego) in the early 70's as a part-time systems programmer. (Don't remember how I got a job there, but I do remember adding some small features to their WATFOR Fortran compiler over one summer.)

TSO (time-sharing option) was a huge upgrade at the time, running as a permanent "job" under OS/360, and virtualizing the job partition to run separate sub-jobs per terminal user. After using punched cards for a while, the thrill of actually typing in programs directly (mostly assembler and PL/I in my case), running, and DEBUGGING them interactively was nirvana.

The predominance of women in the pictures is very striking. Was that common at the time? If so, I wonder what happened to change the demographics so heavily.

Maybe “computer operator” was considered a secretarial role?

Note that once upon a time a “computer” was a person who did computations, often a woman with a BA in math. ENIAC’s programmers were women (http://www.columbia.edu/acis/history/eniac.html).

    I wonder what happened to change the demographics so heavily.

Fran Allen says in her interview for Coders at Work that IBM's gender balance slowly shifted in the 60s as a side-effect of new IT hiring standards. Recruiting preference was given to candidates with advanced degrees in engineering, and most of these new graduates were male [1].

Through this lens the gender shift can be seen as an artifact of the higher education system: Fewer women were getting into these schools and finishing with advanced degrees in math and computing. Even today women can have a tougher road putting together the 7-10 consecutive years of full-time academics required for a terminal degree given that the 20s are also prime child-bearing years.

pg explains the corporate hiring mindset better than I can with a "nobody ever got fired for buying IBM" analogy in News from the Front:

A recruiter at a big company is in much the same position as someone buying technology for one. If someone went to Stanford and is not obviously insane, they're probably a safe bet. And a safe bet is enough. No one ever measures recruiters by the later performance of people they turn down [2].

[1] http://www.codersatwork.com

[2] http://www.paulgraham.com/colleges.html

Depended on the institution, but in general males were predominant in management and programmer positions. Operators were split about 50/50 at the places I worked, and I never met a male keypuncher.

I/O control staff (who gathered input data, scheduled batch jobs to run, and distributed printed output) were mostly women.

If you were male, you wore a tie. Those of us kids who didn't know how to tie one (or who operated drum printers and high-speed decollators) were forgiven for wearing clip-ons. I worked one place where men were required to wear ties, and the women all wore a uniform dress.

I seem to recall reading somewhere that at the time, it was mostly women who could touch type. This was the days of secretarial pools, most of those women were probably at around 100 wpm.

In pictures 01 & 02 you can see a copy of the book 2001 - A Space Odyssey on his desk.

Also this impressed me:

"The computer room was in the basement of a building for security and other reasons. There was no natural light and I had a slim budget for decorations. I also had staff with artistic talents so I bought the materials and they made their own decorations."

With the guitars and the art projects it sounds like he fostered a very progressive work environment.

The server is in agony.

Maybe a cache can help:

http://www.luckham.org.nyud.net/LHL.Bell%20Labs%20Days.html

Out of curiosity, how were those photos preserved? As someone who was born in the 80's, I'm used to anything pre-1990 looking washed out. Were you using an especially powerful camera, or was it purely good preservation technique? Are these restored somehow? Some combination of the above?

slides and negatives preserve rather well.

Not to mention the fact that chemicals degrade in certain ways, so an experienced photo tech can correct off-balance scans almost by muscle memory.

Dan Margulis' book on this subject is definitive.

I took one look at the sideburns and thought this can't possibly be real. It had to be a well done (and amusing) hoax.

I was about 60% sure someone got access to a preserved data center that's now a museum, raided the Mad Men wardrobe and took a bunch of pictures, but then I saw the personal page of the guy and it's an older version of him:

http://www.luckham.org/LHL_Index.html

The Internet has made me such a cynic. But still... someone should take my idea of faking old timey things for a different theme and set of pic ;)

On a side note, I remember growing up in the early 80ies, travelling by plane with my Dad who worked for IBM and he would sometimes bring these giant tape spools with him in his luggage. I guess flying across the country with them was faster than uploading the data across the Net ;)

Either that, or he was committing corporate espionage. In which case, I wasn't getting nearly enough presents, Dad!

Still is, though for larger amounts of data.

the site sure loads like it's on a 60's internet connection... :)

lova to see a zoom enhance on the cartoon on the corkboard here http://www.luckham.org/images/Bell%20Labs%20Days/Bell%20Labs... I know it can't be Dilbert (can it?) but it sure looks like it

It wasn't around in the 1960s, but Crunchly was definitely one of the things I had to go back in time and appreciate, despite having not been born when most of them were drawn. http://xivilization.net/~marek/imgs/crunchly/

IMHO Roxanne & Helen were very beautiful.

Anyone else notice that he seems to have forgotten the names of the ones that weren't quite as attractive?

And the author's sideburns were pretty impressive as well. ;)

Umm, why is this sort of comment relevant or appropriate on HN?

all users are human, and beauty exists.

..and both cases come to surface only in certain cases. And this is one of those.

Can't disagree with you there.

In the early 80's I use to go into a place my dad did hardware contract work every month to help him vacuum out filters in all their CAD machines. Their server room didn't look much different from this (except for more crowded, and probably a smaller footprint) and they even had one teletype box for their tape library backups. I still have a huge plywood box the size of a toy chest (actually, it was my toy chest!) that a 20MB hard drive and enclosure came in.

Anyone who hasn't seen Desk Set (1957) with Katharine Hepburn needs to. (Reference department at a television network that gets its first 'computer')

Nowadays all that work [and more] can be accomplished by a single person.

On their mobile phone.

That's an overstretch. Although modern mobile phone has much more power than mainframe from 1960-s, mobile phone does not have convenient interface that allows to input and output massive amounts of data.

If we ever get an optic nerve tap and thought-writer worked out, we'll scarcely have any need for physical interfaces that laptops and desktops offer.

The thought-writers are halfway there already, they're working on tech for disabled people that lets them manage somewhat crude gestures by thinking in directions.

The photo of the decor in the "Demonstration Center" is priceless. Almost straight out of Mad Men.

648 meg of hard drive! geezz. Would not even be able to view those pics on that computer back then.

Pics as displayed are about 8k each, or 128 to the MB, so the drive would have been capable of holding over 82,000 of them. Of course jpg format didn't come about until 30 years later...

Full size images are 1600 x 1061, so with 16 bit colour and no compression you'd be looking at about 25MB per image, or about 25 per drive. As a jpg they're about 600k each

Leaving aside the fact that decompressing a decent-sized JPEG was a very CPU-intensive task as recently as the 486 era, taking seconds.

My father worked in the engineering department at the DuPont Corporation in Delaware in the '60's, and I remember going there on "take your son to work day" and seeing my first Cray. This really brings back some old, old memories.

This is cool. Can't say I'm sorry to see the sidburns gone though...:)

Clearly you're not based out of Portland. Those sideburns are alive and well.

thanks for the pictures, it is fascinating to see open colored pictures. loving it.

Back in the late 70's I once knew someone who knew someone who knew someone, who shall remain nameless, who was a fairly senior programmer for a large bank. On certain occasions, this person was able to bring home thick printouts of the actual COBOL code that ran in the bank's mainframes. Probably one of my first exposures to source code. The lines would alternate between a green and white paper color. Along both the left and right edges were rows of holes, needed by the printer mechanism. The code itself was highly highly verbose plus was probably >50% comments, both inline and in section delimiting blocks. I later got into BASIC programming myself as my first language, then 6502 Assembly, then C. Though I appreciate Java's strengths, I totally understand why some folks call Java the New COBOL: it's pretty verbose, full of ritual, and it's used everywhere now in large corporations and government. (Not sure if used in banks heavily, but would not be surprised. Though I know COBOL is still running in many banks as well.)

People apparently still use that:

http://www.staples.com/Staples-Green-Bar-Computer-Paper-14-7...

Reply on news.ycombinator.com