You are describing a gargantuan problem beyond the scope of Tor. A tool to compare the results from various exit nodes would be nice, though. That would be a powerful tool to determine what is blocked in various locations. That output could be exported to a large database keeping track of Internet censorship.
Check out Psiphon. It's purpose is undermine Internet censorship. http://psiphon.ca/ ----- Original Message ---- > From: Benjamin S. <[EMAIL PROTECTED]> > To: or-talk@freehaven.net > Sent: Monday, December 8, 2008 8:05:09 AM > Subject: technical solution for censorship [was: UK internet filtering] > > Am Samstag, den 06.12.2008, 19:49 -0500 schrieb Gregory Maxwell: > > > http://community.zdnet.co.uk/blog/0,1000000567,10009938o-2000331777b,00.htm?new_comment > > > > I've confirmed the reports of UK ISPs censoring Wikipedia using some > > UK tor exists. > > I think it's time to find a better technical solution to deal with > censorship in different countries. > Censorship is increasing all around the world and we should be prepared > that sooner or later nearly every country will censor the internet for > different reasons (cp, intellectual property, politcal reasons, etc.) > and in a different way. > > A perfect technical solution would make it possible to request any > ressource as long as there is just one exit-node which isn't affected by > censorship. > But at the moment TOR-Nodes doesn't know which ressources are censored > and which really doesn't exist, so it can happen that you use a circuit > which is not able to bypass censorship. > Putting them on the bad-exit-list is no solution, because first that way > they are lost for the network for all requests which are not censored at > all, and second the specific user has to know that the requested > ressource does exist, which nodes are able to access it and how he can > force TOR to do so. > > This could be done better: TOR itself should know which nodes are > affected by censorship and use another for the specific request. > > The list could be auto-generated by the exit-nodes. For example an exit > which gets back a 404 or a negative DNS-result could simply ask some > other exit nodes (in a different country) to check if this is "real" or > censorship. If last one, the specific request could be put on a list and > published to the directory. > Other TOR-Servers could use this list to check if they are also > affected; clients would be able to check if the existing circuit is > affected for the specific request and build up a different one. > > Not every censorship would be recognized this way, but for the other > ones there could be a button implemented to TOR-Client. If pressed, the > TOR-Exit is forced by the user to do the check. > > A nice side-effect is, that we get the neccessary data to check the > censor which could be useful at least in democratic countries which have > no public lists ('cause of index liborium prohibitum). > > So far, > B. > -- > Surf anonymously and reach Hidden Services by using TOR, JonDos (JAP) > and I2P with https://tor-proxy.net .