On 10/26/05, Alphax <alphasigmax(a)gmail.com> wrote:
Isn't Tor back-tracable by timing attack in China
due to the size of the
network, making any "anonymity" useless?
I think he's less concerned with anonymity than he is about getting
around the Great Firewall and being able to edit Wikipedia. If he's
worried about prosecution, well, that's his own fear and he should be
aware of what is or what is not legally/technically possible. Anyway,
that's an issue for Tor and him to worry about, not us.
There are a host of reasonably simple technological solutions which
could mitigate most of this (i.e. allowing users with reasonable edit
histories to edit from tor accounts) but as Jimbo said all of that
means very little unless we sit down and hack out the code for it.
On 10/26/05, David Gerard <dgerard(a)gmail.com> wrote:
Looking at
http://bugzilla.wikimedia.org/show_bug.cgi?id=550 , the
problem isn't having the code to hand - we have code patches on hand
for lots of possible options. The problem is deciding what is actually
a good idea socially. What's an obviously elegant solution to the
problem? If there isn't one, what compromise would make things better
rather than worse?
I took a look at it briefly and it seemed to be 80% of the people
saying, "This might be a good option to have in some circumstances"
and another 20% saying "I dunno, vandals might be able to still abuse
it!" Personally, I think it looks like an instance of the perfect
trumping the good.
If the code is there to do the "block certain IPs from allowing anon
edits but not from account creation" trick, why not just turn it on?
If all hell breaks loose we can turn it off again. We don't have to
get it perfect on the first try, do we? Whatever happens, the
experience would definitely inform future discussion and speculation
about these sorts of things.
FF