I'm tempted just to say "what BrowserUk said". But, I have something of my own to add:

Good code is not "cool", because ego is not involved. The code is correct, and works. How the programmer *feels* about the code is irrelevant. Good code and good practice stands on it's own merits; "cool", and other subjective, emotional metrics do not.

I simply wish to call BS. I have my reasons:

1. Good code can be cool code too. "Cool" can mean a lot of things: innovative and interesting among them. Innovative approches to solving problems can fall well within best practices -- and even when they don't, sometimes they are such an improvement that they become best practices. Object-oriented methodology was novel "cool" code at one time -- and it broke the rules of procedural programming that were generally accepted. Yet it's become so useful that it's become a best practice in many organizations.
2. Ego -- that is, pride in one's work -- is incredibly important to writing good software. Developing software certainly has disciplined aspects to it, but it's also a creative endeavor. So is designing a suspension bridge. Any master of any discipline will tell you that experience develops intuition. Intuition is nothing more than seeing something and knowing it's got potential (or not) because it "feels right" (or wrong). Granted, you should always back up such feelings with evidence before you go production, but that emotional process is essential to creative problem solving and efficient work.

Yes, there are times when "creative" or "cool" solutions are inappropriate, but it is not (as you suggest) always inappropriate. It only becomes a problem when the niftiness of the solution begins to outweigh practical considerations -- when "cool" becomes more imporatant than "good".

See, "cool" and "good" are not mutually exclusive.