> 
> True. I think that's why CTM'd need some sort of promise from the user to
> actually use the diffs. Like you get a copy of the source for the day when

Encrypt the diffs using a checksum of the file the diff is to be applied
against. Or, have the protocol use a checksum of the file to be diff'd to
request the diff in the first place.

							David
							david@vex.net