Sunday, December 16, 2007

IT Memory Lane



Another snowy day in Boston. Luckily it is Sunday, and this provided a perfect opportunity to go up and clean out the attic.

What should I find but my old class notes from the spring of 1969, a course entitled "Information Systems" -- "1.00," in MIT parlance -- taught by a wonderful professor named Dan Roos who, by coincidence, I happened to run into recently.

We used an IBM 360 computer in those days. It took up an entire room and probably had less memory than your cell phone does today. You would write your program in Fortran -- an example from my final project is shown in the bottom picture above. Then, you would punch Hollerith cards to encode the program on a medium that could be read by the computer. I show several above, along with a diagram they gave us to show what fields were punched by which key on the punch machine. The machine used a hexadecimal numbering system, so we were expected to be conversant in base 16 -- delineated as 0-9 and A-F.

After keypunching, you would hand your batch of cards to the system operator, who would feed them into the computer. Depending on your priority in the queue, you would wait five minutes or an hour for the program to be compiled by the computer, only to discover that you had made a programming error, a keypunch error, or had a hanging chad on one of your punch cards. The problem, of course, is that you didn't know which of the three problems had occurred!

Class 1.00 is still taught at MIT. Here is the description of the 2005 class from MIT OpenCourseWare. Every one of those desktop computers you see in the picture has a gazillion times (whether in base 10 or base 16!) more capacity than the old IBM 360. By the way, Fortran still exists, but these kids get to use Excel and Java instead. Seems a bit too easy to me . . .

24 comments:

Patient Dave said...

Okay, now tell the nice people what we used to do with chad from those Hollerith cards. Oops, I didn't mean "we" - YOU were far too good a citizen at 19 to have anything to do with that mischief, but you lived precariously near to people who did.

JP said...

I always thought the endless cycle chasing character counts, re-punching, waiting on computer ops for the re-run, and finding out you still had a decimel placement off was the best example of "make-work" that existed in academia. Hope the current offering of 1.00 provides error checking for Fortran programs.

Steve said...

It was pretty entertaining writing online business apps in assembler language with only 3k of RAM to work with.

Steve
Modulator

Paul Levy said...

Patient Dave "might" be referring to the apocryphal case of a plastic garbage bag full of chads that was swung around someone's head in someone else's dorm room, only to be let go and spray chads throughout the room -- each chad charged with static electricity and therefore clinging to the walls, furniture, ceiling, and floor of said room. Removal, reportedly, was very time-consuming.

On the other hand, he might have something else in mind.

YogiBear_ said...

1969 was a good year. Neil Armstrong walked on the moon. The conversation between him and NASA was 'aired live' across the world. It certainly was a great leap for humanity.!!

Shoes had to be removed before entering the computer room. Data Entry was a specialised operation requiring skilled people to punch cards and operate the computer. Huge magnetic tapes were used for storage Memory & back-up. It was awe inspiring to see the machine in operation through a glass wall with the tapes whirring and stopping at sudden intervals. Multiple users were kept busy at numerous terminals behind a clear glass wall. Card punching activity was done in a separate room. Feeding the Monster with data in the the correct sequence was an achievement & lot of hard work. It felt good to be involved in such activities which could carry out a million calculations in a fraction of a second!! Wow !!

Sometimes it feels like ,It is a Pity we have evolved, to become more busy but less happy.

What 'something' else(?) could 'Patient Dave' be possibly refering to...??

Nasov said...

C'mon, patientDave -- is that what you meant?

Anonymous said...

Patient Dave, I am also curious, considering I was not even born, I would like to know what else could be done with those chads.

Sarah said...

Ha ha! Did you also find your abacus, I mean, slide rule, in the attic?

Paul Levy said...

I didn't have to. It is in my desk.

Patient Dave said...

Well.

The fact that Paul thinks the flinging / slinging / swinging chad bag is apocryphal... well, that just shows what a non-vandal he was.

I know BIDMC is all about minimally invasive treatments. Could this have been embedded in Paul's priorities as a result of certain *very* invasive treatments he heard about, involving sprinkling that staticy hard-to-remove chad into someone's underwear drawer?? Hm??? Well, *I'll* never tell. And whatever it was, I had nothing to do with it.

A key factor in the optimization of aggravation in this method (so I'm told) was that when an MIT student goes to the underwear drawer, it's typically when time is very short - it worked particularly well on frosh who were rushing for their 8 a.m. physics lecture, in the course where they teach freshman physics in the fall term and sophomore physics in the spring term. (There's real pressure to attend those classes - I personally tried the approach of not attending them, and it didn't go so well. And just to confirm that it wasn't a statistical fluke, I kept trying it, with delightfully consistent results. p<.0001, as I recall.) So when the harried and fearful frosh frantically opened the drawer and found sharp little pointy things all over everything, loud bitching could sometimes be heard through the door - followed by guffaws from elsewhere down the hall.

A substantially stickier situation arose when Certain People Who Shall Not Be Named (not on OUR floor of the dorm, of course) mixed a bag of chad with maple syrup (genuine Vermont, as I recall) and did things with it that I never witnessed. But I hoid about it, so it must be true.

By the way, this sort of aggravating inventiveness among 19-year-olds built real creative skills. One denizen of the next floor up, notorious for his shower-fighting skills (don't ask), may or may not have had anything to do with chad hacks, but he went on to be one of the inventors of the SMTP email protocol. "And," as the man says at the end of Fever Pitch, "that's a fact."

Patient Dave said...

Sarah,

Now let's see if Paul remembers the Tech Cheer that mentions the slide rule. (It's very, very, very nerdy.)

Paul Levy said...

Something like this:

e to the x du dx,
e to the x dx.
Cosine, secant, tangent, sine,
Three point one four one five nine,
integral, radical, u dv, slipstick, slide rule, MIT!

Patient Dave said...

Paul,

Well, I guess that's close enough. Google shows there are way too many variants for me to prove you wrong.

I must add, though, that my mother (who was a rowdy child, by all accounts) came all the way from Vassar to party with some MIT guys. She taught me another cheer, which I saw proven on too many occasions:

Rooty toot toot, rooty toot toot,
We're the boys from the Institute.
We don't smoke and we don't [chew*],
Don't go out with girls that do.

Rooty toot toot, rooty toot toot,
We're the boys from the Institute.
We're not rough and we're not tough,
But boy are we smart!

*For bracketed words, you may substitute a rhyming word that might seem appropriate for a vandalous Vassarette in my ancestry.

Patient Dave said...

btw, the slide rule (a decidedly analog device) brilliantly illustrates that there's little need for more than 2-3 decimal places in most practical calculations. Today's science and engineering students would do well to ponder how much successful calculation and building took place using slide rules and other analog tools, before the concept of a bit even existed.

AND NOW I'M GOING BACK TO WORK! (Distant echoes of "You're hacking me ... you're hacking me right out of the institute. Go away!" Or, the physics major who posted a note on his door: "You do not exist. Neither does my thesis. It has priority.")

Sarah said...

Dave,

I have had the distinct pleasure of hearing him chant the Tech Cheer in person. It's not quite as spirited in text, is it?

I'm just glad that it was genuine maple syrup. That Aunt Jemima stuff is appalling.

I've never heard of this type of thing happening at Harvard. I'm not sure if that makes you MIT boys more or less cool (groovy? swell? neat-o? oh, it's before my time)...

n, md said...

It wasn't 1969, it was 1979 when I graduated from college AND WE WERE STILL USING THE IBM 360 AND THOSE CURSED CARDS!!! Batch processing, hours to find out you forgot a semicolon at the end of one line of Fortran code. . .it goes on and on. And I still have my slide rule, too! Thanks for the memories!

I met a medical student last week who is very techno-savvy and programs in Java; he was IMPRESSED that I programmed in Fortran (he didn't know about PL/C) but we still were able to commiserate about assembly language! (some things never change)

But we had the time to wait for those programs to compile and run because we didn't have to check email or (in a totally different vein) go to the gym or watch the shows we "TIVO'd" or the movies from Netflix or wait in line at Starbucks or keep up with the latest news 24/7 (because the "24/7" life had not been invented yet)

sigh. . .

Patient Dave said...

Hi, "n, md" -

> the "24/7" life had not been invented yet

Guffaw. It is to laugh.

To pick just one little data point, I give you the case of R, who lived across the hall from me. Week after week after week, at n in the morning (where n varies wildly), he'd emerge into the hall and ask "Don't you people ever sleep?"

It's not by chance that the MIT mascot is the beaver. Not much as an aggressive football icon, but is well known as an engineer that does its best work at night.

Bill said...

n may have varied widely, but it was quite predictable. R***** had the misfortune to live close to the lounge and to be a light sleeper. He would appear at 12:30 AM and regularly every 45 minutes thereafter if things were too raucous. Consistency was his hallmark: “Don’t you people know what time it is? Don’t you people ever sleep?” Eventually he didn’t have to say anything—his silent appearance would trigger the response, “Yes, R*****, it’s _______. No, we never sleep.” after which things would quiet down for a while.

Chris H said...

Wow, those punch cards and the mention of Fortran certainly bring back memories for me. The most painful memory being of the day I learned the extreme importance of "multiple banding". I think that's when I decided that it wouldn't be a bad idea to buy stock in a rubber band company...

Paul Levy said...

Oh yes, Chris. I well remember the day my batch decided to play "52 card pick-up" and went flying across the floor. Since the cards were not necessarily numbered consecutively, putting them back in order was no fun.

Anonymous said...

I remember spending hours trying to figure out why my stack of cards wouldn't generate any results other than error messages. It is amazing that enough computer engineers survived those days to bring us where we are now.

Kristin said...

When I took 1.00 in 1997 it was as equally mind numbing for this course V'er. My not yet husband (course VI) got me out of too many jams at 2am the mornings the problem set was due. This same not yet husband also took your course while we were undergrads and often touts his minor is course XI as if it's more impressive than my MD.

Keep up the entertainment,
KR, MD
MIT '00, BIDMC Resident

Paul Levy said...

Let's translate that, Kristin, for the rest of the world. At MIT, "course" with a Roman numeral means "major" or "department". V = chemistry. VI = electical engineering and computer science. XI = urban studies and planning. Class 1.00 is and was offered in course I, civil and environmental engineering.

But how can you question whether your boy's minor is more impressive than your MD? Almost anyone can solve medical problems: Solving the problems of cities is much more difficult! :)

Anonymous said...

Paul, I enjoy and value your blog.

I have fond memories of learning how to create drum cards on the 029. This was at Ohio State, where I got my B.S. in Comp Sci many years ago.

If memory serves (and more and more, it doesn't) I used to be able to have the 029 punch a sequence number in columns 73-80. Useful if you dropped your deck and knew how to run the sorter.

I also remember that over time, one could learn to discern the major of a person by the kind of jobs the ran.

If someone came in with boxes of cards, they were clearly a social sciences major -- no threat. Just let them read the cards, run their SAS or SPSS job, and they'll be gone soon enough.

If someone was looking tired and beat-up, and had a deck of about 300 cards, they were often comp science majors -- again, no threat.
Most runs were flushed pretty quickly (syntax errors) or were canceled (infinite loop, canceled as soon as the student's account balance was drained.)

The people you really had to fear, however, where the folks with a deck of say five to ten cards -- physics major. After submitting their deck, you could go to the job monitor and see all the tape and disk drives suddenly go busy and you knew the machine was going to be running that job for a few hours at least. Arrgh!

I finally took a job in Radiobiology so I could use the PI's TSO account.

John Sambrook
Common Sense Systems, Inc.