What about implementing some signals dillo would react on. I.e. SIGHUP to make dillo reload the page.
Additional code must be justified by satisfying a user need. I don't think very many people are going to send signals to Dillo to reload the page -- they'll just click the Reload button.
No code needs justification. Only the inclusion to release version does. One person thinks he needs this feature that is trivial to implement. Even if it never goes to the main trunk, let's give it a try: it might prove usefull to some people. Or not, who cares? Dillo is great for this because the code is simple enough to understand and experiment with. Good luck trying this kind of stunt with mozilla! The Sylpheed project experimented embedding dillo, and eventually dillo included the feature. Bottom line: I'll always favor experimental contribution that prove useless to sterile discussion about the value of an hypothetical feature. Eric.