"Rhythm Section Want Ad" comes from TMBG's first album, but TMBG broke a weird kind of new ground with their first album: they skipped the rhythm section, and worked against a backdrop of drum machines and DAT tapes. They toured like this, and didn't have the funds to hire a rhythm section until at least their third album. I've seen them live at least four times and only seen a rhythm section twice.
Combine this fact with the incredible number of songs they write, and you'll see the logic in my "Rhythm Section Want Ad" conspiracy theory: They Might Be Giants doesn't have a rhythm section because they were so into songwriting that when they realized they needed a rhythm section, they wrote a song about finding one, and then when they were done, they wrote a song about something else.
TMBG have never been flashy, but a big part of their success is very solid songwriting. Their demo tape got in the national mainstream media. This is the "Speak softly, drive a Sherman tank" school of productivity: one good way to get good at something is to do it over and over again.
If you want to talk about how to use
lambdain Ruby, I understand it better than a lot of people. But what about simple shit? What about putting an app on a server? I'm currently pursuing the Sherman tank approach with my miniapps side project and my "Twitter a new beat every day" side project as well. I'm skipping that whole Pragmatic Programmer "learn a new language every year" thing because I spent a lot of time learning
lambdaand all the rest of it and I figure that's already my strong suit. There's a very clear pattern in my skills: I suck at anything that bores me. That's fun to a point, but overall, it's a weakness. "Learn a new language every year" is good up to a point, but you want to make sure it doesn't trap you in a useless mastery of esoteric voodoo. If you want to get good at something, you should work on every part of it, not just the parts you already do well.