Tor

Agree/Disagree: Anonymity systems like Tor provide some degree of censorship-resistance, but not for all users

Agree, a censoring country can block virtually all Tor traffic

T/F: To avoid timing-correlation attacks, Tor routers delay and sometimes reorder packets before forwarding them

F, Tor does not delay

T/F: Because Tor routers "peel off" layers of encryption, packets to the entry node are larger than packets to the middle node, which in turn are larger than packets to the exit node

F, all packets are padded to the same size

T/F: Hidden services reveal their identities to their Introduction Points, and trust the Introduction Points not to reveal them

F, hidden serves are anonymous to their intro points too

T/F: Tor has high latency because Tor routers delay sending packets so they can "mix" (reorder) them before sending

F, it does not mix

T/F: Tor does not use public keys because doing so would reveal its users' identities

F, it does use public keys

T/F: To maintain anonymity, Tor does not reveal any IP addresses that are running a Tor router

F, it makes the list public

T/F: Broadcasting (such as over FM radio) a message to a large, highly populated geographic region will naturally achieve sender anonymity

F, this gets you receiver anonymity

Does website fingerprinting attack work when applied not between client and entry node, but rather between middle node and exit node?

No, it only reveals destination not source

Why does the client (host originating the connection to the hidden service) create a Tor circuit to some rendezvous point instead of simply letting the hidden service create a circuit to the client?

So that he doesn't reveal his identity to the hidden service

T/F: If you participate in Tor and agree to serve as an exit node, it may look (say, to the FBI) that you are the one visiting the website that Tor clients are directing you to go to

T

T/F: In a mix-net, sending "junk packets" is necessary to ensure anonymity even when not everyone has something to send

T

T/F: Timing attacks that deanonymize a Tor circuit are possible without having to observe any traffic to or from a Tor middle node

T

T/F: Tor is not secure against a global passive observer

T

T/F: Tor routers with higher bandwidth are more likely to be chosen to be on circuits (all other things being equal)

T

Why does the hidden service create a Tor circuit to the rendezvous point instead of simple letting the client create a circuit directly to the hidden service?

The client doesn't know who the hidden service is

Why is Tor susceptible to website fingerprinting attack?

Tor does not send junk packets

website fingerprinting

database of websites based on packet uploads/downloads