[tor-talk] pdf with tor

coderman coderman at gmail.com
Fri Jul 10 19:09:47 UTC 2015


On 7/9/15, flipchan at riseup.net <flipchan at riseup.net> wrote:
> couldnt we just code some protection against this

WhonixQubes with DeepLang semantic barriers between isolated temporal
processing pipelines.

you obtain the PDF inside a transient isolated VM via scrutinized path
through upstream Tor and Firewall VMs.

next hand-off to normalized representation VM instance, which produces
each page as a well formed bitmap no compression no metadata graphical
format.

then, each probably innocuous page image is paired to yet another
temporal VM to convert page into semantic description in ascii text,
english language.

finally, a semantic reconstruction VM uses the natural language
descriptions from above to re-construct a page for your viewing
pleasure as intended, now without risk of malicious surreiptitious
constructs...  perhaps with a puppyslug or two. :P


best regards,


More information about the tor-talk mailing list