Except that exactly what most people will want and get mad if it's not easy to run. The backlash from that isn't worth it.
> I just feel that this is not a barrier for open-sourcing their product.
IP alone can be a bitch and that coupled with the fact that a number of companies have secrets (key/tokens/passwords/etc) in their code and cleaning those out (AND removing the git/svn history) is no small task. Lastly there are number of open source RSS readers out there, I seriously doubt Digg was doing anything particularly innovative in this space and I don't imagine many people care to build on something they made 5 years ago and probably didn't update much since then. Especially if it's not built with the "New hotness".
Also: "Nothing in the world is worth having or worth doing unless it means effort, pain, difficulty… " -Theodore Roosevelt
That said, you are absolutely right. Software at scale is complicated and often difficult to deploy and maintain. I just feel that this is not a barrier for open-sourcing their product.