reply to post by _R4t_
Thanks, I understood all of that except the acronym 'MITM'; what's that mean?
Also, it's wisest to consider that keyword triggered 'sniffers' are a thing of the past, a compromise based on limited relative computation
capacity. Why not go whole hog?
For example, you may think 'well I have a 100GB hard drive and no one could track all of
that data'. However, how much of the data on your
hard drive is
unique, when looking at the set of 'all installed machines'? All the system files, common applications, downloads, etc., exist
in multiple places, on multiple machines. Why do more than identify 'this user has this common file', when constructing an terminal-state image,
relative to the Internet as a whole?
I've run 'duplicate file' calculations on a typical Windows installation, you would be
amazed at the amount of reported redundancy in the
file system layout....
I estimate that the complete activity log, of the changes a user affects to their system state in a single day, could
easily be represented
with less than 50KB of data, when correlated to a world-wide database of 'common files', and deltas thereof.
Have a look into government sponsorship of research initiatives into data checksumming and 'digesting' functions, al-la 'MD5'.
Edit to add: Oh! I guessed! MITM = "Man In The Middle" (research some cyptography sploits if this doesn't make sense)
[edit on 10-9-2008 by Ian McLean]