Thanks for your comments on this folks ... lots of food for thought.
   [1]coderman@gmail.com said:

     the longer discussion is how to make decentralized search useful.
     "Google style" search has a terrific performance advantage over
     decentralized designs by brute force. however, take advantage of
     massive endpoint / peer processing and resources combined with
     implicit observational metrics for reputation and recommendation,
     inside a well integrated framework for resource discovery in usable
     software, and you have something more robust and more effective than
     "Google style" could ever provide.

   Yes, it does seem like the speed advantage of centralized search will
   be a barrier to adoption of decentralized search. This is analogous to
   the difficulty getting people adopt systems like Tor because it is
   slow. But I think that as more people become aware of the extent of
   state/corporate surveillance, they will become more inclined to accept
   solutions that are slower in exchange for not having their search
   habits monitored, and also being able to receive uncensored search
   results. As long as decentralized search is (a) usable/simple and (b)
   provides quality results, I feel like speed is somewhat of a secondary
   concern. The key question to me is: "How do we build a search engine
   that is [2]simple enough for Grandma to use, that produces quality
   results without massive centralized indexing servers?"
   Standalone P2P search applications (e.g. Yacy) don't really make sense
   from a usability perspective. It's unrealistic to expect hundreds of
   millions of users to download a standalone Java app, and configure a
   P2P search node. What would make more sense, and would lead to much
   more rapid/widespread adoption, is to use protocols like [3]WebSockets
   / [4]WebRTC to facilitate [5]P2P connectivity in the web browser, so
   that everything can be done via a simple browser plugin that can be
   installed by anyone with few clicks, and would then just allow people
   to use the browser search bar as usual. This browser integration would
   also have the bonus of simplifying the choice of what to index -- it
   could just default to indexing bookmarked and frequently-visited pages,
   and then be optionally customized by more advanced users to create
   custom indexes (i.e. all of the complexity of setting up indexing could
   be hidden from the user, unless they choose to look for it).
   To help bootstrap the WebRTC nodes into the P2P network, and to deal
   with some of the instability inherent in P2P networks (i.e. by creating
   stable "[6]super-peer" indexing nodes), I like cathalgarvey's
   suggestion of utilizing something like a Wordpress plugin that would
   use the same index/search standard as the WebRTC clients, but could
   additionally bootstrap the web-based clients. As cathalgarvey said:

     A standard rather than a codebase. But there's a huge advantage to
     this line of thought, if you'll bear with me. A two-digit fraction
     of the web right now is powered by Wordpress.org, who explicitly
     advocate open/free culture. If you can convince them to include a
     social search/index standard of this type, which is virtually free
     in terms of computer resources, then you'd have it deployed across
     the web in days as the next update rolled out. Indeed, even if
     Wordpress seemed reluctant, a wordpress plugin could probably be
     written quickly enough to enable such a thing and make it available
     for casual use. Suddenly, a bunch of PHP-powered sites around the
     web start committing small bits and pieces of resources to a social
     search engine based on human-curated attestations of trust that flow
     through a web, helping to confine spammers to the fringes and to
     users with stupid taste.

   What would also be interesting is if this standard enabled some kind of
   "pingback" mechanism whereby WebRTC nodes could be associated with
   specific super-peer nodes (e.g. maybe people who have bookmarked the
   super-peer site in their browser, or subscribe to its feed), so that in
   addition to broad/random queries that target the entire P2P network,
   clients could also create more targeted custom searches that say
   something like "start the search with the nodes that are clustered
   around these super-peers". This would create an enormous diversity of
   search possibilities -- hundreds of thousands (millions) of different
   "search engines", each of which would return different results for the
   same query, depending on where you start your search ... This diversity
   is another reason I find P2P search interesting, in addition to the
   benefits re: censorship, traffic shaping, and surveillance.
   I've been looking around for some kind of WebRTC P2P search engine and
   haven't found anything yet ... maybe I've found a programming project
   for this summer :)
   -- [7]Jesse Taylor

References

   1. mailto:coderman@gmail.com
   2. http://freehaven.net/anonbib/cache/usability:weis2006.pdf
   3. http://mikemp.mit.edu/wss/wss-mikemp.pdf
   4. http://inet.cpt.haw-hamburg.de/papers/vws-lwpcd-13.pdf
   5. http://inet.cpt.haw-hamburg.de/teaching/ws-2012-13/master-projekt/maxjonas-werner_aw1.pdf
   6. http://infolab.stanford.edu/~byang/pubs/superpeer.pdf
   7. http://www.interference.cc/