Well, the demo is not everything. According to this article, you must also be able to change ideas and trains of thought. What good will a demo do if you are ultimately going to change your train of thought and move on to something else?

Demos are great to crstallise thoughts and really build the platform for your application. Don't underestimate the value of the experience gained by putting put a demo. Afterall, you want to build something people use - you can't do that with an idea.

We've got a number of applications around the same core bunch of ideas. prototyping one application and getting people using it couldn't be a better validation of one's ability to execute when it comes to Y combinator. Lying back on an idea and waiting doesn't show anyone you want to make it happen.

If your idea changes - well it just means you've proved you can execute before. Get a demo out even if its very rough (anything is). Just start!