Nice discussion.
Minor nit unless I'm confused. On slide 49 you say the SN75188/SN75189 are TI version of MAX232. The big thing about the introduction of the MAX232 was it no longer needed +-12V to generate proper RS-232 level. The SN75188 requires +-12V. It is the chip the MAX232 is intended to replace. Yea I'm old enough to remember when it came out.
Going off on a tangent, as a hobbyist I used to hate the MAX chips. Well, hate is perhaps exaggerating it a bit, but still. They were really expensive, so expensive that it usually was cheaper to buy a random AC wall wart at a thrift store to create the missing +/-12V.
If there would had been a way to differentiate hobbyist and small series commercial customers I think that the MAX chips would had been way cheaper for hobbyists.
The high price for the MAX chips led me to find out that many but far from all 1489 receivers will accept 0V/+5V. In the case the 1489 didn't you could use the 1488 output signals to power your DIY circuit that sent signals to a 1489 input. I remember doing that with discrete transistors for various projects back in the day, like for example level converters for, ehh, extended satellite TV viewing but also for my DIY caller ID receiver.
(Anecdote: here in Sweden DTMF tones were used, so a MT8870 DTMF decoder chip, a 4040 counter that divided the oscillator that already existed in the MT8870 to a suitable baud rate clock, and two 74xx165 chips and an opto coupler resulted in a circuit that would send printable characters to a serial port, albeit only digits 1-9 would have the correct character representation directly. Still good enough for testing with terminal software, and easy to DIY software. I remember having an old 8088 PC with an amber monitor with software that would automatically add unknown numbers with the current date/time as it's "name" making it easy to see if an unknown number had called previously or not. I didn't have any driver for the RTC card I had for that 8088, but the chip was available from an electronics component seller that offered data sheets sent as fax, so I received the data sheet with my fax modem and wrote my own software. IIRC that chip was weird, it obviously kept time, but also date, month and day-of-week but not year, so the driver would have to look at the combination of date, month and day-of-week, and maybe also look at a command line parameter and/or date of file to tell which year it is).
Also I think that electronics components sellers would likely had sold more MAX chips to hobbyists would had sold more of them if they had sold a package with the chip and the required capacitors. Never underestimate peoples laziness.
(There were no electronics component shop local where I used to live, so I had to order every part that I couldn't salvage from old junk or that I already had laying around from older projects. If there had been a shop nearby where I could had bought MAX chips I would probably had used MAX chips a lot more).
Btw it's obvious that MAX tried to milk as much money out of the customers as possible. It would had made more sense if they only made their own transmitter chip, with as many transmitters that would fit in their chip size, and then just make their own clone of the 1489/75189 chip and let the customers buy either that chip or buy it from any other manufacturer.
===========================
Sorry if I'm writing something that's already been said in the video (that I haven't yet watched), but: IIRC there were a shortage of RS232 driver/receiver chips for a while in the 80's which is why you in some cases can se OP amps used as RS232 interface chips.
github.com
(my prior discussions are also on github for archive)
Perhaps that'll be my next X16 project.
