#2 incomplete daemonization

[Jeff McAdams]

Around line 392 of version 0.9, there is some comments
concerning the proper daemonization of egd, and that
its not completely done, and some of the rational
behind that choice. While I can certainly understand
the reasoning in the commentary, I wanted to point out
an issue that arises thanks to the incomplete
daemonization.

If you start up egd fairly normally...it forks and
exits, returning control to the
terminal...unfortunately, since it doesn't do the rest
of the daemonization process (closing stdin, stdout,
setting TIOCNOTTY, etc.) if you hit ctrl-c (or whatever
the terminal interrupt key is), egd will exit(!)
Suffice it to say, this wasn't expected, and was rather
difficult to track down because you don't expect a
piece of software that has forked and returned control
back to the terminal to respond to input from that
terminal like that.

Anyway...might want to go ahead and implement the rest
of the daemonization to avoid weird issues like this.

Thanks!
Jeff