Ok,

When I said Google, I didn't mean a central indexing scheme, but rather the
look to the user.  Since users browse Freenet w/i a browser, presenting it
there makes sense to me. Unlike other p2p apps which are looking for files
(mp3's, doc's, etc.), Google is searching HTML, as would a search engine in
Freenet (at least at first) so I think the presentation should be closely
tied to the browser.

So, the only thing that could be indexed would be the data that is locally
viewed (through FProxy for instance).

The problem with that is, depending on how it is stored, it would leave
traces of what the node operator himself/herself viewed, but I think there
is a very simple means of protecting against that problem. First, we can
hash the keyword, which would only allow a dictionary attack to result in
viewed content.

Then, we would mix in external content to the index in the same manner of
key content in the data store.  As the search result is passed back through
the chain, it would be cached in the local index of each node -- since the
index result would be very small, it could easily be cached all the way back
the communication chain.  This would insure the privacy of the node
operator.

Jeremy


_______________________________________________
devl mailing list
[EMAIL PROTECTED]
http://hawk.freenetproject.org:8080/cgi-bin/mailman/listinfo/devl

Reply via email to