• Please review our updated Terms and Rules here

Reading punched cards the modern way

Modern, feh. I'll take 50 year old (1963) technology any day:

vs-cdc-405-card-reader.jpg


1200 cards per minute (that's 20 cards per second); 4,000 card hopper and stacker.

Let's see you do that with Legos.
 
many years after the human race die out, most magnetic media will have been damaged. i wonder if paper punch cards will last any longer. we need to transfer our best code to something indestructible. Z80 processors will even survive a nuclear winter. i want the aliens that find and reconstruct a TRS-80 system to be able to laugh at the game SUPER NOVA.
 
Right now I can't find it, but I read an article about how paper storage is the best long term option for saving data. It said almost the exact words you did, teigan, that magnetic (and burned disks) won't survive very long, and that paper will given the right conditions. So yep, paper storage wins the game in the long haul.
 
This subject comes up from time to time on classiccmp. Granite is a pretty good storage medium as far as permanence goes, but the density sucks. Paper is inexpensive and, if properly made, is relatively durable. Again, the density sucks, but you can still read 25+ year old Cauzin Softstrip programs. The equipment needed to read them is simple.

Older than softstrips, we still have quite serviceable player piano rolls from more than a century ago. They're remarkably durable.

Wet clay might be a good medium as well, but again, there's the density problem.

What I'm certain of is that the Web is a lousy way to preserve material. I can't find things that I bookmarked 5 years ago--I can't imagine what things will be like after the Web has been around for 50 years. Probably mostly social networking and political blogs, stores and porn.

It's worthy of note that the Dead Media Project couldn't even make a go of it.

The best advice for preservation is to copy everything every couple of years. Those stringy floppies won't last forever!
 
Yup, same article is worried about the preservation of history, photos, music, etc due to the fast paced failures and replacements of technology. Can you still watch that wedding video or interview of a grandparents stories on that betamax tape? The history of X as we thought it to be burned on to that interactive laserdisc? What OS will care to read the fat32 partition on that probably ESD'd usb 1 drive in 60 years? In a thousand?

Paper isn't perfect either (weather, knowledge eating insects) but still, we discover new knowledge all the time from our past that's long forgotten.

My cynical answer: The media might survive but we probably forgot the password.
 
1200 cards per minute (that's 20 cards per second); 4,000 card hopper and stacker.

Just out of curiosity, and since you probably know off the top of your head, what's the actual data rate of that?

Regarding paper, it's hard to beat the historically proven permanence of that. Although only half as old as a few other examples, Gutenberg's 42 line Bible is apparently as good as new. IOW, 500 years is a piece of cake.
 
Just out of curiosity, and since you probably know off the top of your head, what's the actual data rate of that?

That's a little hard to quantify in modern times. The 405 reads and transfers 12-bit words, one per column (it actually reads each column twice and compares the two reads). There are 80 columns on a card, so 12 × 80 = 960 bits are transferred per card, at the rate of 20 cards per second. So the raw data rate would be 19,200 bits/second or 1600 characters/second if reading in character mode. The conversion from column binary to characters would normally be handled by whatever processor was handling the I/O.
 
There's also mylar punched tape, not the greatest bit density, but VERY durable
GREAT IDEA! I actually used it in 'the old days' to back up important programs and even still have a couple of 1000' rolls.

So, let's see; to back up my 3TB drive at 10cpi would take, umm... 633,600 cpM (characters/mile), 3 trillion (at least in N America) divided by 633,600...

Uh, oh; that's 20 times the distance to the moon; I'd definitely have to buy more tape...

Now, let's see; at 100 cps that's 52,560,000 cpy (characters/year), so divided into 3 trillion... hmm... 57,078 years to do that backup...

Let me think about this some more...
 
Last edited:
Okay, Mike--but how much of that 3TB actually means something? I suspect that the largest part of it is redundant noise or inefficient encoding. In other words, how much of that 3TB could be dropped before the content became totally meaningless?
 
Okay, Mike--but how much of that 3TB actually means something? I suspect that the largest part of it is redundant noise or inefficient encoding. In other words, how much of that 3TB could be dropped before the content became totally meaningless?
You're right as always of course; I bet if we passed it through 7-ZIP we could get it down to a length of no more than once to the moon and back and a much more reasonable 10,000 years or so to do the backup... I'd still better start it soon though...
 
Hey, I am new here. I saw the DIY card reader link and said to myself, this better involve an Arduino, and it so happened it did:eek:. I agree that the professional machine can chug a lot of those things I used as note cards in grad school every minute but they are bound to get broken eventually and you can only make your own machine to read the cards when you can no longer fix the professional machine. It is not about doing it quickly but about being able to do it, like the "what's best storage media" question. Not for speed but for "it still works" in the distant future.

I am not sure about Z-80 surviving nukes though. All electronics are suspectible to radiation hardening so what is special about Z-80?

To really not lose information, kind of, I think, enough redundancy is needed so digital media is not the solution. You need the redundancy like a holograph. You lose half the holograph (broken off), you still see the same picture, slightly less quality, remember analog, folks?. Holograph exposed with photographic plate may survive many years in a shock resistance case. Anything that is a direc representation of the information, like a bible or whatever, is bound to loss if a pice goes missing. You need every single part of a medium to represent the whole information so it can be broken in two pieces and that only creates two less quality copies :)
 
I kept almost no cards for the mountains of code I wrote during the punched card era (I had two of those 5-foot high card filing cabinets in my office and they were pretty full). The reason is simple--tape. All you really needed to do was to get your cards onto a reel of tape and you were set. You could take a 600' reel, write your cards to the tape, then add another load point to the tape and repeat for as many copies as you wanted to back up.

One of my old tapes turned up a couple of years ago--but it was 7 track 556 BPI. Who the heck has 7 track drives any more? :(

Paper probably isn't a good long-term solution, however. Remember the library at Alexandria?
 
So if you wrote programs in the 60' or 70' on those cards, were you called programmers or those who load these cards to the machines called programmers? I am not old enough to witness any of these actions, just word of mouth, like, "they will get back to you in several days when they have your results for pickup." The invention of terminals and modems must have put an end to the cards history. Oh, I am with the impression that only FORTRAN cards existed. So was FORTRAN the only language written to cards or were other langueges not invented at the time?
 
So if you wrote programs in the 60' or 70' on those cards, were you called programmers

Heck, in the 60's they would probably be called "code pigs" or 'coder pukes". They were very under-appreciated by the engineering community. Eventually, they earned the title 'Software Engineer'.
 
Last edited:
So if you wrote programs in the 60' or 70' on those cards, were you called programmers or those who load these cards to the machines called programmers? I am not old enough to witness any of these actions, just word of mouth, like, "they will get back to you in several days when they have your results for pickup." The invention of terminals and modems must have put an end to the cards history. Oh, I am with the impression that only FORTRAN cards existed. So was FORTRAN the only language written to cards or were other langueges not invented at the time?

Good grief, young 'un! My first Pascal program was on cards. Most of what I wrote was assembly, FORTRAN, or specialized implementation language (e.g. IMPL). But Algol, FORTRAN, COBOL, RPG, GPSS, PL/I, SIMSCRIPT--even APL was available on cards. I suspect that even some C was done on cards at some point, but I can't say for certain. My first 8080 program was on punched cards.

Go to http://99-bottles-of-beer.net/ and be entertained by the variety.
 
Back
Top