On Wed, 2015-12-02 at 07:05 -0500, Karl wrote: > I guess I'm just saying that the number of faces we have to compare > to > live normally implies that we can easily distinguish a lot more > information for this specific task than for other tasks.
I have not looked at the research, but individuals vary wildly in their ability to remember different sorts of information. And not everyone is great at faces. There are probably going to be people who remember a phrase, poem, or very short story generated by a fixed markov engine much better than an abstract image. There was a discussion of using visualizations to protect users against .onion fishing attacks* on tor-dev here : https://lists.torproject.org/pipermail/tor-dev/2015-August/thread.html# 9302 In that conversation, there was a feeling that (a) different people might need different visualizations and (b) a per browser salt should help lots. One could bind a regular non-generated image with a public key, in such a way that the image acted as a per contact salt instead of a per browser salt. If your contact did not care about anonymity, then you' might see a real photo of them, an abstract photo, possibly arranged around them, and a short story. Basically, a lol-contact key remembering and description tool. Jeff * In TBB's situation, there is an easy way to record identities one intentionally visits using browser bookmarks, but these do not fare so well when visiting another url on the same site using an href from another site. It'd be tricky to expand this as the Tor community is loath to make Tor Browser remember more about the user's browsing patterns.
signature.asc
Description: This is a digitally signed message part
_______________________________________________ Messaging mailing list [email protected] https://moderncrypto.org/mailman/listinfo/messaging
