Apparently you just interconnect functional modules rather than designing the low-level hardware. I hope I'm wrong, of course, as that's a quick way to do things, but not very cost-effective in production quantity.

Yes, I "interconnect functional modules" isn't that what design is regardless of whether you refer to function calls or the UART in the uC or a Bluetooth unit.

If what you mean is "do you go buy ready-made units instead of designing boards" the answer is yes for very small runs, no for the larger runs with the following exception. In some instances I do buy "functional modules" regardless of volume. That is in the areas where know-how really count. I (actually our customers) need the GPS/dead recogning unit to be totally precise at all times and buying it from a company with the know-how to make that gets me a happy customer. I see very little reason to reinvent the wheel and instead concentrate on what makes our products unique. One, often overlooked fact, is that the NIH syndrome often create severe "birth pains" which makes the first customers totally lose faith in you. As to the precision of the positioning, there has been lawsuits won against my customers won beacuse the positioning was not accurate enough (I am happy to say this was my competitors product)

The 805x had 99.999% of its time available to do useful work. During the very few milliseconds per day that it had to interact with either the console or the CRT display ... just a question of waiting for a blanking pulse.
One thing that you are bypassing with examples like the above is that not every application can "take a pause". If an operation takes 2 seconds and occur once a day the "time available" approaches 100% but that does not mean that the unit necessarily can stop for 2 seconds. "Time available" is about the most abused criteria I know of, "largest hunk stolen" is the valid criteria.

Erik