Grumpy_Mike:
I WAS a kid in the 70s, you could not build anything from a computer manual. You had to get the data sheets, by post, study them and make your own design. I know because that is what I did.
For the most part, that was true, Mike - and maybe it was a different thing for you guys across the pond, as well.
I wasn't a kid messing with computers in the 70's - I got my first real computer (TRS-80 Color Computer 2 - 16K, etc) in 1984. My knowledge of the time was coding in BASIC, and typing in programs from magazines (and converting those that didn't have an example for my machine). More "standing on shoulders" I suppose - but it did lead me to an eventual career in software development.
That said - it was possible (barely) that - if you had access to the Apple II/IIe (and maybe even the Apple I) manuals - you could use them to build your own machine. The manual had a complete schematic in the back (folded out and everything), as well as a text listing of the ROM and monitor assembler code. So - in theory, and if you had the equipment, knowledge, time, and money - you could "clone" the Apple. I'm sure more than a few people did so. I know the Apple was cloned in the Soviet Union, too.
With the Color Computer 1 and 2 - you could also get the schematics (I don't recall if they were included with the machine in the manual or not - maybe - or you may have had to get a copy of the service manual from a friendly Radio Shack bench tech). The system was mainly a bog-standard Motorola 68xx design - all the chips and such were all from Motorola, including the VDG and such. What you would have trouble getting would be the ROM - you would have had to get a dump of it somewhere, then had some way to burn it to an EPROM (not cheap equipment back then). But I bet people did it.
Side note: You couldn't do the same with the CoCo 3, though - the GIME chip, which was a custom ASIC of the day, was not available to the public - to this day, it's lack of availability has been a thorn in the side of us who run these machines, as there isn't any chip-level documentation on how it worked, nor any real datasheets available. While it has been emulated to a great extent, that emulation isn't anywhere near perfect. Nor has anyone been able to yet recreate it in hardware - though there does exist a very large wire-wrapped CoCo 3 prototype that is in the process of being reverse-engineered - it does not have a GIME on it - but it was used for OS-9 development at Microware, so there is a suspicion that it's GIME was implemented using a more discrete method.
Otherwise - I know there was available more than a few books on how to build your own computer from ICs and such (CPU, glue 74xx TTL stuff, RAM, etc) - most of them from TAB Books. Byte Magazine - among others - also published many articles on building your own computers. Most of these were pretty rudimentary - even compared to contemporary equipment - but they worked, and gave people a lower cost way to experience computers and computing - and for someone skilled enough, most of the time they could be easily expanded in some pretty extreme ways. There were also several robotics books for hobbyists of that era (I have copies of most of them on my bookshelf) - and more than a few of those described how to build a computer (or something akin to a hardware state machine controller - at the minimum) to control the robot - microcontrollers were definitely not a thing then, as you know.
So - here in the States at least - it was possible (and yeah - you probably still needed the datasheets - not arguing that) to build a computer "from the manual" back in the 1970s; ok - late 1970s - really, homemade computers weren't a thing until around 1975-76, with the advent of the 8080 and other low-cost cpu's and glue logic that came out from Intel and Motorola. Prior to that, I do know there were people playing around with everything from the 4004 (at best), to old telephone relays, custom analog computing machines, discrete logic, switches and lightbulbs and other things - but most of that stuff was mainly very basic logic circuits - maybe an adder or counter here and there, or basic flip-flop style "memory" built from relays. I know there were a few very ambitious people who built custom electro-mechanical machines that used custom paper tape or other methods for I/O, memory - relay-based registers - and other custom pieces - and the machines were very large, very slow, and had little capability - but they taught their builders a ton. Even those were typically informed by more than a few books, though such information wasn't easy to obtain in a form that showed exactly how it worked (I have collected a few university-level textbooks from the 1950s and 60s on computer design and programming - some of them go into enough depth for you to construct a whole tube or transistor machine - if you had the money, space, and time to do so).
Well - enough of the nostalgia trip I guess - I'm not trying to argue with you here, Mike. Overall, yes, we all stand on the shoulders of giants, and we should use that leverage when and where possible. Ideally, someone will do the research and realize that the pre-constructed parts and PCBs are a great value, and free up your time to do more interesting stuff. But - some people do have old L297 and L298 chips lying around or got them for a great deal (I have a few tubes myself laying around) - so, if they can fit the application, and efficiency isn't a great concern - and the extra functionality isn't needed - people shouldn't berate somebody on their choice, but should instead help them to understand whatever their problem is. They would get their problem (hopefully) solved - and they'll have the knowledge perhaps to next time look into the more modern devices that are available.
