[Cryptography] Digital dyes for tracing digital leaks ?

[Phillip Hallam-Baker]

On Thu, Jan 10, 2019 at 11:22 AM Henry Baker <hbaker1 at pipeline.com> wrote:

> If I'm trying to find a leak in my sewer or water pipe, I
> can choose from a large variety of colored and/or fluorescent
> dyes, which are easily detected when they appear outside the
> pipe.
>
> Ditto for radioactive tracers for even tinier leaks.
>
> Are there standard digital *codes* (other than 0xdeadbeef)
> for performing the same function in detecting use-after-free
> and other types of data leaks?
>
> Ideally, the codes should be random-looking enough that
> they are picked up as possible crypto keys by key-hunting
> software, but also easily traced back to different sources.
>
> I was thinking along the lines of linear codes -- e.g.,
> some sort of CRC code, where a long enough subsequence
> can not only determine the polynomial, but also the
> position (i.e., exponent).
>
> Thus, "uninitialied" memory would be initialized to such
> a code in such a way that if some of the same bit sub-
> sequences showed up again, it might be indicative of a
> data leak.
>

There are many schemes of that type, see Blue Spike etc.

Basically they use steganography and massively redundant coding to encode a
hidden tag into audio and/or video.

Also there are schemes for Word Documents and such that ensure each
recipient of a classified document has a slightly different one and so on.
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://www.metzdowd.com/pipermail/cryptography/attachments/20190110/ec8121b4/attachment.html>