[Noisebridge-discuss] door stats

Shannon Lee shannon at scatter.com
Fri Jun 21 21:27:33 UTC 2013


A flatfile of anonymized log fields would go better with my cut | sort |
uniq -c | gnuplot toolchain.

--S


On Fri, Jun 21, 2013 at 1:19 PM, Leif Ryge <leif at synthesize.us> wrote:

> On Fri, Jun 21, 2013 at 11:58:01AM -0700, Shannon Lee wrote:
> > No, if you generate a one-time pad of random hashes, such that...
> >
> > 1 -> foo
> > 2 -> bargain basement
> > 3 -> mickey mouse
> > 4 -> jet plane
> >
> > ...et cetera, then we maintain frequency counts and timing while
> > anonymizing specific codes, and you can't brute force it because there's
> no
> > hash function.  You can even throw away the one-time pad, because the
> > timing data makes it easy to map "mickey mouse" back to specific entries
> if
> > you have the original dataset.
> >
> > I think Curtis' suggestion actually has the same properties.
> >
> > --S
>
> I was talking about making pseudonymous (not anonymous!) door code logs
> public a few months ago, and started on a project to do so. I'm sorry I
> haven't finished it yet, and glad to see others interested in doing this
> now!
>
> (Previosuly I heard a few concerns which I think I addressed, but mostly
> a lot of meh.)
>
> My plan was/is to replace the flat file of codes + comments (which
> frequently have email addresses which obviously should not be made
> public) with a json file that has a 3rd field called "nickname" that
> users can optionally choose, knowing that it will be publicly logged.
> The nicknames can be randomly chosen initially.
>
> I began implementing that here:
> https://github.com/noisebridge/noisebridge-baron/compare/master...json
>
> The converter script works, but has boring names ("pseudonym 751dfb0"),
> and I have not tested the changes to baron.py yet.
>
> I was thinking of trying to generate pronounceable gibberish names, but
> haven't done that part yet.
>
> I am not going to have time to work on this today or tomorrow but this
> thread inspiried me to push the little bit of code I wrote so far in
> case someone might find it useful.
>
> If nobody else publishes something by Sunday, I'll probably work on this
> then.
>
> ~leif
>
>


-- 
Shannon Lee
(503) 539-3700

"Any sufficiently analyzed magic is indistinguishable from science."
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.noisebridge.net/pipermail/noisebridge-discuss/attachments/20130621/4cb5b08f/attachment-0003.html>


More information about the Noisebridge-discuss mailing list