No one made me back out the ClientDelete command. I did it because most of the rest of the group didn't like it. Since I didn't discuss it on the list before implementing it, I didn't have any right to check it in in the first place. I now regret using the overly rhetorical phrase "under protest". All I meant by that is that I didn't concede the validity of the arguments against it.
While I remain unconvinced by Ian't technical arguments, the legal one reposted on CofE yesterday is compelling. There are other ways to do insertion validation that would be just as good and wouldn't have the legal risks. If there is support for a "nonlocal" flag on ClientGet that causes keys to be removed from the local store before retrieving them, I'll implement it next week. Whatever I do, I will make sure there is consensus on the list *before* checking in the code. --gj -- Freesite (0.4) freenet:SSK at npfV5XQijFkF6sXZvuO0o~kG4wEPAgM/homepage// _______________________________________________ Devl mailing list Devl at freenetproject.org http://lists.freenetproject.org/mailman/listinfo/devl
