I can tell I’m dumb because of how much effort I’ve put into trying to understand this and I still haven’t gotten it. But for the life of me I can’t grasp how we got from punch cards to machine language to assembler language. It is almost magic.
In: Technology
Good question — it was critical mass. Via punch cards, we got enough code into computers to process keyboard input. The code ingested via punch cards allowed keyboards to work. And then keyboards worked and it snowballed from there.
Many many incremental developments.
First off, a punch card is a persistent representation of assembly, which itself is a mnemonic for machine code. There isn’t much disparity there. The main difference you’re referencing has to do with input device, but really the more relevant factor was storage and memory. With a punch card, you didn’t need persistent storage for code because the cards themselves contained the program. If you want to input code using a keyboard, you need to save it somewhere and retrieve it later. Floppy disks and, for very fancy systems, hard disks were the first on the scene which could serve this job well. Floppies remained pretty dominant until the price on hard disks came down enough that they could take on more and more of the OS and applications. By this point, floppies had shrunk to a smaller size with more space and CD-ROMs were starting to become much more common.
There were keyboards on some very early computers. You may really be wondering when we went to interactive instead of batch mode use. It really started around the time MIT let students poke at their machines in the late night hours, in the very early sixties.
Really mainly was the cost and density of memory- you need enough memory to run programs that take the forms of input that are convenient to humans.
No real breakthrough was needed, keyboards that generate an electrical signal that a computer could interpret predate the computer. [https://en.wikipedia.org/wiki/Teleprinter](https://en.wikipedia.org/wiki/Teleprinter) was used since 1887 in telegraphy. You could connect a keyboard on one end of a long wire and a printer on the other side. They could alos print to purchase tape.
You can store machine code (language) on punchcards, it is not separate concepts. You can store code in high-level language on punch cards to that need to be compiled to run.
There is a keyboard to enter information onto punchcards too. The common IBM 80-column punchcard standard was introduced in 1928 and used with computers on a large scale into the 1970s.
Early computers that had a keyboard and text output interface simply used existing teleprinter technology.
Puchard is primarily a storage medium, it was replaced by magnetic tape and floppy discs, not keyboards.
To remove the need for any storage media like a punch card to run programs from memory, the first computer that do that is the Manchester Baby from 1948. It would have been quite clear even earlier that it was an option, but memory was very expensive so running stuff directly from physical media made sense early on.
The early computer was extremely expensive to purchase and run. Connecting them to a keyboard and printer/monitor was simply a waste of resources. It was better to use the electromechanical device to input the data and then run the program with that data on a computer. That way as soon as the program has been completed you can run another. It is when computers get cheap and you can have a computer that multiple users can use at the same time it makes sense to have a keyboard and monitors as a way to input data like we do today. You could have done that before but it did not make financial sense
Look at [https://en.wikipedia.org/wiki/Semi-Automatic_Ground_Environment](https://en.wikipedia.org/wiki/Semi-Automatic_Ground_Environment) that is a US air defense computer system from the late 1950s. It use a light pen and consoles with lots of buttons, not exactly a traditional keyboard but the principle is the same. It runs a program that the user interacts with so you need input for real-time information to the computer, this means punchcards are not an option. The program might have been stored on a punch card and read into memory, I have no idea how that part works.
The first drive was made in 1956, it was very expensive back then. So what storage media was used was often a question of money. Compare it to mechanical hard drives today that are still used even if SSDs exist because they can store data cheaply.
You can store any data on punch cards — machine code, assembly language, compiled languages like FORTRAN, novels, recipes, poetry. (Source: still have some of mine somewhere.)
Data is data.
What changed was the economies of scale. The more people bought computers, the more R&D went into them, the cheaper and more useful it got to make them easier to use. A room full of card punch machines is pretty cheap to set up. A room full of teletypes that can talk directly to the mainframe is more expensive, and it takes more sophisticated software and infrastructure to let multiple people use the same computer rather than feeding in one person’s card deck at a time.
All the advances we take for granted: online connections, glass CRTs, mice, networks, the Web, were incremental improvements on what came before, and were made because the previous interation was useful and there was proof that it was worth improving.
Could we have built a smartphone in 1969? No. Were computers useful enough in 1969 to keep improving? Yes.
Steve Wozniak was a hobbyist programmer and hardware guy, who put what was in the computer (6502 microprocessor) onto a monitor. Then he met Steve Jobs at a geekfest. Then they made Apple Computer.
The transistor. The invention of it is the start of the Tech Age. Everything goes back to it. It made switching possible by electricity alone. They replaced the holes in the punch card.
Why is switching by electricity wanted? That’s because it is not manual (you don’t have to do it by hand or by help of physical machine), and it is scalable (you can make it as big or as small as you want depending on how big or small the currently available materials allow). It’s also fast, because it is not manual and you can take advantage of the nature of electricity.
Your typical light switch in rooms are manual switches for electricity. Your input is either turn it on or off then that information is converted into electrical signal. Another example is the push button where when you press it there’s light; and when you let go, the light turns off.
Keyboards are basically a lot of push button switches.
Latest Answers