Over at O'Reilly there is a article written by Derek Sivers where he details how he rewrote an entire PHP app in Rails and how at the end, he found that newer isn't always better. Derek lists a few reason why it didn't pan out so well but I find #7 probably the best... plus it has a great title. Derek makes a great point that you as a developer progress, so its a bit unfair to judge the code you wrote years ago.
I know there are a few apps I've written in the past that if I looked at them now i'd think, god, what the hell was I thinking! At the time though, with what I knew, it was probably the best that I could do. What I think that signals though is that I've evolved as a programmer since then. I'd be more afraid if I looked at past apps i've written in the past and couldn't see anything wrong, that'd mean that I haven't progressed at all.
Check out the article here
, I think it's worth a read and hopefully sparks some thoughts of your own.