And REAL MANLY PROGRAMMERS WRITE IN BINARY!!!
(been there … long time ago … glad the world has moved forward…)
OSHPark takes .kicad_pcb files, PCBWay takes gerbers. But generating the gerbers is not painful. It also gives me a drill report, which I use to edit the .kicad_pcb to change the hole sizes (sed script). Don’t really care which way to do it - both ways are easy.
(Yes, I know I could copy/edit thhe footprint files, but doing it at the end is easy.)
And REAL WOMANLY PROGRAMMERS WRITE USING TOGGLE SWITCHES. (Yes, early program entry using the toggle switches was often done by women because the men felt it was akin to secretarial work and therefore beneath them. Really glad those days are long over.)
I had to toggle in the bootloader for my first system.
Theses women were better than me… gunga din…
If you go far back enough they used plug wires. ENIAC probably isn’t what anyone would consider a programmable computer by todays standards tho.
Punched cards were used in the second generation of computers starting in the early '50s. But the first generation of computers in the '40s used switch input on so-called “plug boards” (reference). An example of a computer like this is the Colossus. (Note the gender of the operators in the first picture.) Apparently there was a paper tape mechanism, but the Wikipedia article (I didn’t go further) isn’t clear what the paper tape is used for. But the first paragraph states that the computer was “programmed by switches and plugs and not by a stored program.”
I actually was just remembering that most of the people feeding in these punched cards were women. At least that’s what my grand ma told me. 
I remember computers with toggle switches, PDP-8s ere in use when I was at university.
We were not allowed to touch them and had to write in Fortran and input using punch cards for an ICL mainframe, it made you a careful typist
Me to.
It was MERA-300 (Polish construction).
I remember that using switches we (students) entered a simple tape reading program. If remember well that program had something about 7…10 machine words only (not sure 14…20 bites each). Then the tape into the reader and run that program.
At tape there were one after another:
We were doing some hardware to be connected to that computer. Connection (to its bus) with wires wrapped (using special tool for it) around I think about 2cm long square pins in a big bookcase (i didn’t found the right word - case seems me being too small, as this was about 1,5m high, and about 2m width).
Writing programs in Algol and punch them yourself at punch cards was the other lessons.
I recall having a shoebox full of punch cards and tape for an IBM/370 in the mid '70s. Online access was via a teleprinter which had a handy punch tape reader & writer. Fortran programming with plot jobs spooled to magnetic tape and then plotted on a Calcomp plotter or off the lineprinter. I still remember getting a stiff note (NOT an email!) from the SysOp saying my job had filled an entire reel of magnetic tape - oops…
I thought they were much more advanced, like this… https://www.youtube.com/watch?v=ljGH07Unfe8
I don’t know that movie, but I am surprised 30uA being so DANGER.
Oh the old, venerated art of wire wrapping. At my first job in electronics I was in the engineering documentation department and in our files were stored wire-wrap lists for much older products just in case we needed to provide support. Here are a couple links for those interested:
I haven’t seen any modern, commercial application of wire wrap in decades, but I’m sure there are some out there who still practice the art for simple (or slow clock-rate) projects. I would imagine crosstalk and RFI would be a major headache for analog or high clock-rate applications…
Wire wrapping was becoming common when I started work. I think that it was a popular due to the high price of developing and manufacturing multilayer PCBs in those days. The first mult-layer I ever saw was in about 1984
If my memory is accurate . . . . wire wrapping appeared in commercial equipment around 1960. The IBM System-360 family of mainframe computers, introduced in 1963 or '64, had a lot of wire-wrapped backplanes. By that time there were automated machines that produced a wire-wrapped chassis or backplane under program control. IBM had already been using wirewrap in many of its peripheral units, and was already tooled up to manufacture wirewrap assemblies.
I believe you are correct about wire-wrap filling a niche when the up-front costs, and schedule delays, of creating a PCB meant that PCB’s were not economically viable for small production runs or urgent requirements. The System-360 hardware used PCB backplanes for the main subsystems common to many models in the family but custom capabilities and limited demand features were implemented with wirewrap.
The low-end models of the initial System-360 family had clock speeds around 1 MHz or so. You could tune an AM radio to the clock frequency and get an idea of what the machine was doing - number crunching, memory-intensive operations, I/O with fast or slow devices, etc - by listening to the clock signal. My SWAG is that those signals can tolerate wirewrap construction. A decade later, when the product line was discontinued, I think the highest performing models had processors that clocked around 30 - 40 MHz. I can’t imagine those signals passing through a wirewrap backplane and emerging with even mediocre waveform integrity.
Dale
Ah, been there too. But nostalgia just ain’t what it used to be 
I remember this, toggle switches to load the noddy that loaded the paper tape reader. The paper tape had the boot loader that read from magnetic tape. The system was loaded from the mag tape into the massive hard disk, when I say massive I mean the size of a beer keg. 30 to 45 minutes to get the system up and running.
Them were the days, when small furry creatures from Alpha Centauri were real small furry creatures from Alpha Centauri.
My dad started at GE computer systems division in 1959 in Phoenix. Later, he became a consultant. He had a gig where, every morning, he would load the paper tape for the , start it, and made/ate breakfast. He then loaded the tape for the application he was working on, cleaned up breakfast, & read the paper. Finally ready to goto work. He did not reboot the system very often…
I worked at Floating Point Systems in Portland, OR in the early 80’s. They wire-wrapped the backplanes.
One of the women who did the work had been a weaver. She loved the colors and wove the backplane. Needless to say, they did not work && made her rip them up. I thought they missed a great opportunity for her to make art && sell it for her.
I had a gig at Boeing in Albuquerque 2003-2005. They wire-wrapped prototypes and many were used in flight. I suggested they get a copy of microcode’s circuitmaker/traxmaker for $1K and use a PCB house like AP Circuits. Nope. Would have saved them hundreds of hours.
Them were the days, when small furry creatures from Alpha Centauri were real small furry creatures from Alpha Centauri
I remember them days… Some of my best friends were real small furry creatures from Alpha Centauri. Best mates you could ask for. And cheap drunks - apple vinegar - but the hangovers lasted for days…
