Yes, but when that Web site gets traffic
from the banned IP address, what is the
evidence that the person who got the
C&D letter was the one responsible for
the traffic? The traffic could have
been from any customer of the
ISP used by the banned user.
While I think that your argument is flawed in its understanding of the law, a technological approach could be: let x be the suspect IP address. When we receive a request from x, we send them some messages that we don't send to other users; or maybe we change messages slightly (like leave out some footers, or with a different image link, whatever). When the fake/modified messages show up on the suspect website, we can reasonably assume that they accessed our website. At that point, it doesn't even matter any more whether they did it through x or another.
Right. Now are beginning to make a little
sense -- a big change from the OP.
So, the problem was not getting the data from
Craigslist but publishing it elsewhere.
But your point still will not make Craigslist
at all happy: Craigslist has a lot of
data. Maybe someone, Tom, wants to get
a copy of a lot of that data and do some
things with it. The data at issue need not
be in any meaningful way subject to
the marking you describe. That is,
what Craigslist sends is HTTP, HTML,
CSS, simple text, and maybe some files
in JPG, PNG, GIF, etc. Tom can grab
just the text, put it into a database,
analyze it, and republish leaving
the marking far behind.
So, right, if Tom does something sufficiently
stupid, then he can be slapped down
for violating the C&D letter.
Otherwise, not stupid, it's essentially
impossible for the Web site to know
if Tom honored the C&D letter.
Then, searching Tom's computer, etc.
will be based on next to no go evidence
and be wildly unfair to Tom.