> > > Well, that will be the fault of the author, since they inserted both.
> > > It doesn't strike me as a serious potential problem.
> > You're forgetting that KSKs are SVKs too. Doesnt' have to be the same
> > author.
>
> Well, we need to assume that an insert finding a reference to some data,
> but not finding the data, when the data *is* still in Freenet, is
> sufficiently unlikely as to make this.... unlikely.
>
Well thats a pretty terrible assumption. I'm more inclined to just reject
KSKs period in order to prevent namespace attacks. I can understand
relaxing this requirement on the other SVK derivitives since they would be
controlled by the same author.
Scott
_______________________________________________
Freenet-dev mailing list
[EMAIL PROTECTED]
http://lists.sourceforge.net/mailman/listinfo/freenet-dev
