From my standpoint, easy is a plus, not a minus. I've taken to describing it this way. In computer science we're used to measuring algorithms and data structures in terms of the two basic tradeoff costs -- time and space. The third basic cost -- error -- takes some people a while to get used to. That is, your algorithm can be "wrong" (either probabilistically, or in that it gives an approximate answer) in some hopefully well-defined way in order to trade off against time and space needs -- many times you're willing to settle for a good answer instead of the optimal answer if it's much quicker. But there's also a 4th basic tradeoff cost -- programmer time -- that I find the theory community is happy to just ignore. Simple is good, because more people will use simple things, even if they're not the most efficient possible, because often the usual time/space efficiency isn't really the bottleneck. Coding up something that works is. This is why Bloom filters show up in most of my talks (Suresh, drink!); for me they provide an outstanding example of issues related to the 3rd and 4th tradeoff costs.
But I was inspired to go ahead and post something about this because of the following from Paul Krugman's blog today. (His title is "The Power of Two" -- if that's not a sign, what is?) I figured he says it (everything?) better than I could, though you'll have to substitute CS keywords for economics keywords in the below.
If this strikes you as too easy, and you think that real economics should involve harder math, well, I feel sorry for you — you just don’t get what it’s all about. (You’re what Rudi Dornbusch used to call a “fearful plumber”). And by the way, coming up with a really simple formulation of what seems at first like a very hard problem can take a lot of work. It sure did in the case of the MF lecture, where I actually did start with a really ugly saddle-path thingie until I realized that formulating the sudden stop the right way would make all of that go away.
Simple doesn’t mean stupid. Thinking that it does, does.