Google is, to a first approximation, the web's only search engine. As such I think there's an argument to be made that declaring your bot's User-Agent as Googlebot is a technical necessity, similar to how every browser declares itself as Mozilla.

This is fascinating. Seems like at least Bing openly ignores robots.txt?