Etymology of “Foo” (2001)
ietf.org
ietf.org
on another note, i have always disliked the usage of these terms. they stand only for internal lingo and serve no useful purpose. in every case, there are better names that can be used, and i feel foo and bar do nothing but distract me from the example and use case at hand. also, as a student in my first assembly programming course, the professor just started using foo and bar indiscriminately and frequently. i was thoroughly confused at least the entire duration of that lecture and maybe the next until i realized these were seemingly arbitrary words he had picked and not, as i had originally thought, meaningful words or even instructions. i'll take a and b any day.
MUMBLE MACRO
I get where you're coming from, but an advantage to foo and bar is that once we understand they are arbitrary words, that's a reliable rule. I do a little teaching and new students are incredibly literal. (Which makes sense, the kid that answers "what time is it" with the exact hour/minute/second is more likely to grow into a programmer - I certainly did).
As a result I can't use "better" words, because any meaningful word will be taken to be the requirement. "a" and "b", as you suggest, have issues with being TOO small, too similar, and already have a meaning in sort() uses, not to mention that learning "a" and "b" being meaningless terms is just as hard as learning foo and bar. Pretty much any terms you use will face the same issues, so often the route of least pain is to use the same set of arbitrary terms, and a set that we have a general common agreement as being arbitrary so when a student reads a book, blog post, or watches a video each one of these don't have to explain that foo and bar are arbitrary - we each have to learn that once.
Which isn't to say that foo/bar/baz aren't overused in places, but I doubt it'd be a good idea to replace them entirely or even mostly.
https://en.m.wikipedia.org/wiki/IP_over_Avian_Carriers
and the QoS update with ASCII diagram included
https://tools.ietf.org/html/rfc2549
Category: Experimental
On 28 April 2001, IPoAC was actually implemented by the Bergen Linux user group, under the name CPIP (for "Carrier Pigeon Internet Protocol").
[4] They sent nine packets over a distance of approximately five kilometers (three miles), each carried by an individual pigeon and containing one ping (ICMP Echo Request), and received four responses.
https://tools.ietf.org/html/rfc527
The archaic jargon and playful imagery harken back to the culture and experience of a time before the Internet even had it's name. Or maybe I romanticize too much :).