In message <[EMAIL PROTECTED]>, Ian Clarke <[EMAIL PROTECTED]> writes >On Tue, Oct 01, 2002 at 09:14:00PM +0100, Matthew Toseland wrote: >> Shouldn't be necessary unless/until I implement encrypted native FS dirs >> - there is a strong argument that we shouldn't concern ourselves with >> this, leave it to the OS. > >After our previous bad experience with DIYing stuff that the OS (and >specialized encrypted file-system softare) can handle, I think that this >should be a very low priority. > >Anyone paranoid enough to need an encrypted datastore should already be >familiar with encrypted file-systems. > >We may want to add a "README.PARANOID" document to the distribution >which gives advice on further securing the Freenet experience (such as >firewall modifications, encrypted/stego file systems, choice of >web-browsers, modifications to web-browsers etc). > >For the moment, I would argue that we don't try to implement additional >security measures ourselves when there is specialized software out-there >which already does the same job. > >Ian. >
Have I misunderstood? Does this mean Freenet content cached by the node appears as clear text? If so, the lack of OTB encryption would seem very harmful to the plausible deniability aspect of relaying Freenet content. Or are the individual Freenet items of content in themselves encrypted unless requested from one's node for downloading? In some (?most) jurisdictions, some Freenet content would be illegal to possess knowingly. External encryption would not help, as it would involve a deliberate external modification of the stored files, which would look more like concealment than ignorance, and in many cases would be transparently decrypted when the data store contents where examined by the user interface. -- Roger Hayter _______________________________________________ devl mailing list [EMAIL PROTECTED] http://hawk.freenetproject.org/cgi-bin/mailman/listinfo/devl
