All, I am very sorry to have kicked over what is apparently an old and hotly debated ant hill regarding memory and zoom and such.
I will repeat that as a newbie, I have no dog in this fight, other than seeking to understand how to use the scope I actually have (as opposed to the ones that I don't have which happen to handle things differently).
That's a very sane attitude
Learn to use whatever tool you have to its best advantage.
Learning any complex tool takes time and energy; being simpler is an advantage of analogue scopes.
The number of people that have
detailed experience of more than one
similar tool is very limited. Various people have allegedly noted something to the effect that "academic fights are so vicious because the stakes are so small".
A very primitive signal generator (basic sine, square, triangle waves), plus whatever I can generate from an Atmel ATTiny84 or Arduino Nano, i.e, I2C, SPI, UART, plus PWM at various frequencies. And for a "real" test, I am planning to test and experiment with the effectiveness of basic R/C de-bounce filters on various switches.
That's an excellent cross-section
The debouncing should show an advantage of digitising scopes over analogue scopes. While it
is possible to use an analogue scope, it is useful to be able to push the button many times and use infinite persistence to overlay multiple traces.
I'd regard I2C/SPI/UART as being more or less equivalent, and just use whatever's most convenient.
What you haven't explicitly mentioned is correlating two or more signals, e.g. time delays between signals, or triggering on the combination of several signals.
You might also like to explore signal integrity issues. Classic examples are poorly constructed (TTL) digital counters, or octal buffers driving a heavily loaded bus.
Have fun, whenever you can