"Now, people who Get Things Done but are not Smart will do stupid things, seemingly without thinking about them, and somebody else will have to come clean up their mess later. "
Inexperienced (young) programmers don't know what's been tried, and what's available. I've been dealing with this a lot at work recently, where we ended up doing poor reimplementations of off-the-shelf stuff due to a mix of ignorance, and honestly, a bit of hubris.
It's a good attitude to have in academia/non-production critical work, but the GP is right, production demands a more conservative approach, especially when money/safety is at stake.
The type of ignorance Joel writes about (see parent) is different, it's more like "slavishly following design patterns" and "writing copy-and-pasted, improperly factored code". Both are ignorance, but the first is better called "NIH", whereas Joel's is, "writing bad code".
That sounds like a brilliant technical guy, capable of running with a daft idea to completion (unlike me, with my collection of at-best-half-built personal projects), who should have some layers of protection between him and Real World Production...