> We had a human/computer interface a hundred years before we had computers. When computers came into being around the time of the Second World War, humans, quite naturally, communicated with them by simply grafting them on to the already-existing technologies for translating letters into bits and vice versa: teletypes and punch card machines.

Is this… right?

I thought some of the earliest mechanical computers (as opposed to human computers) that had much real uptake were “fire control computers,” for things like naval guns (for example). You move around dials and cranks to put the measurements in. I’d call this essentially graphical… it isn’t a series of text based commands that you issue, but a collection of intuitive UI elements, each of which is used to communicate a particular piece of data to the computer. Of course the GUI of the past was made of gages and levers instead of pixels, but that’s just an implementation detail.

I much prefer the command line to a gui, but I think we should call it what it is: an improvement. A much more precise and repeatable way of talking to the computer, in comparison to cranking cranks and poking dials. And a general, endlessly flexible channel that can represent basically any type of information, at the cost of not necessary being intuitive or glance-able.

▲

btilly a day ago | parent | next [-]

This is absolutely right.

The teletype system was invented shortly after 1900. It was in widespread commercial use by the 1920s, for sending text over telegraph wires.

The US government began using punch card machines to do the census in 1890. They were named Hollerith machines. Hollerith is one of the companies that later became IBM. When they entered into electronic computers, their prime market was their own customers who were already using their punch card machines for things like accounting and payroll. For backwards compatibility, they kept the format the same!

Punch cards themselves date back to the early very 1800s, where they were introduced for the Jacquard loom. With the cards providing programmable instructions for fabric design.

It is worth noting that Hollerith was not the first place to try to repurpose punch cards to computation. That honor goes to Babbage's analytical machine (which admittedly was not actually completed).

Basically everything in technology has a far longer and richer history than people realize. I could go on for a while about this...

▲

jazzyjackson a day ago | parent [-]

Still, one might distinguish from batch computing and interactive. Sending commands via teletype didn't really happen until well after interactive TUIs -- the text based system relating to punch cards was a separate machine that prepared cards for calculation in a separate batch process.

Gun Fire Control might be more interactive and predates ENIAC (which was, of course, initially used to calculate artillery trajectory). ENIAC's user interface was plugging wires from outputs to inputs same as telephone operators connecting calls. Hardly interactive.

I don't think we get to REPL/TUI like features until 1960s, you've got Sutherland's Sketchpad with a CRT and lightpen representing GUI and LISP REPL via Teletype just before it 1959ish (actually I'm trying to find an old video I saw demo'ing LISP or APL being used interactively by teletype, it's the earliest kind of terminal I've seen)

	▲	btilly a day ago \| parent [-]
		While the immediate interactivity of these old systems was technically limited, the way that people thought about them was more flexible than you might expect. Try reading https://www.theatlantic.com/magazine/archive/1945/07/as-we-m.... Written in the waning days of WW 2 (Germany had surrendered, Japan had not), it is the first recorded proposal of hypertext. The vision described is extremely interactive. Interestingly, it is also the original description of the science citation index. When this was later combined with hypertext, the result was Google's PageRank system... Now how could someone in 1945 be that visionary about how computers could be used some day? Well you see, he'd been in computers for nearly 20 years, and had been thinking about this system off and on for around a decade, in between real jobs like being in charge of R&D for the USA during WW 2... If you fast forward to the 1960s, everyone should watch the Mother of all Demos. That was possible in 1968. In some ways it was better integrated than what we put up with today...

▲

kragen a day ago | parent | prev | next [-]

This is a terminological confusion. The "computers" you're describing gave their name to the universal symbol manipulators we now call "computers" because, historically, the universal symbol manipulators were originally funded to perform mathematical calculations. They don't have much else in common. The "computers" that Stephenson is talking about are the universal symbol manipulators, the smallest and most limited of which can boot Linux: https://dmitry.gr/?r=05.Projects&proj=35.%20Linux4004

▲

mjevans a day ago | parent | prev [-]

In this context I believe 'computer' refers to only general purpose computing devices, not fixed function calculation machines.

In some sense, early player pianos (IIRC with holes in paper that controlled key presses) weren't computers, but were a related precursor technology / infrastructure.