Search Mailing List Archives


Limit search to: Subject & Body Subject Author
Sort by: Reverse Sort
Limit to: All This Week Last Week This Month Last Month
Select Date Range     through    

[liberationtech] a privacy preserving and resilient social network

David Golumbia dgolumbia at gmail.com
Sat Jun 29 09:28:45 PDT 2013


On Sat, Jun 29, 2013 at 12:05 PM, Eleanor Saitta <ella at dymaxion.org> wrote:

>
> I'm not.  I'm trying to solve specific technical problems which
> support larger social ends.
>

I don't think "privacy preservation" is a technical problem, or at the
least, not largely a technical problem. I think it's (mostly) a legal and
social problem.

>
> This is fine.  I'm not saying that using a network like this will make
> you invulnerable to HUMINT.  What I am saying is that networks can a)
> force your adversary to use HUMINT (which is a lot more expensive),
> and possibly even give you some tools to help maintain your social
> graph integrity, etc.
>

I don't think "forcing your adversary to use HUMINT" is what most people
understand by "privacy preservation."

Further, the snoops use HUMINT to get technical access. it only takes one
compromised friend on Facebook to allow downloading a huge amount of data,
for example. I don't even think it's clear that HUMINT is more expensive
than technical intelligence, and the budgets of snoop agencies are not so
constrained that cost is something we can take comfort in.


> If we build tools that force spooks to use HUMINT to get in, we've won.
>

I really disagree with this, and I don't think it's what most people
understand by "privacy preservation." I don't think members of WikiLeaks or
LulzSec feel their privacy has been preserved because the penetration
involved (but was not limited to) HUMINT.

>
> Privacy-preserving, as a property, doesn't mean "if you don't think
> about what you're doing in the world you can run black ops on this
> platform".  It means "you can keep what you're doing here private
> against mass observation by the motivated and targeted observation by
> the non-resourced".  Or, at least, I think that's a bar that's
> actually meaningful and can be achieved; what you're talking about can't.
>
>
I'm having trouble parsing the two properties you lay out here; they are
both much more complicated than I'd want to make them. I find privacy to be
a simple property: "I'm not going to be snooped on by the govt without a
warrant; companies are not going to collect my data and do inappropriate
things with it." These are matters of law and governance. I believe that
the world in which law and governance ensure these principles is not only
achievable, but the only meaningful kind of privacy we can hope for. Our
political sphere is governed by laws, not human beings.

Back to the original proposition, which did not appear to be yours:
building a social network and proclaiming it to be "privacy-preserving"
suggests to users that they will not be spied on. While there may be some
truth to the difficulty such networks would pose for commercial data
collection, any sense of security from government spying such a network
creates will be false. That will be true until and unless we have a legal
structure built to prevent that spying, in which case the technical methods
aren't necessary to begin with.

-- 
David Golumbia
dgolumbia at gmail.com
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://mailman.stanford.edu/pipermail/liberationtech/attachments/20130629/30fdb2b6/attachment.html>


More information about the liberationtech mailing list