[Bigbang-dev] security and privacy-mentions at the IETF

[Amelia Andersdotter]

On 2019-02-18 01:01, Nick Doty wrote:
> +1, this is really interesting, Amelia! I’m not sure, but I think this
> was an email message stuck in a backlog, so we may be replying several
> months late. Be sure that that isn’t because of a lack of activity on
> the project or interest in this use case.
>
> I especially like the method of using counts for the word “the” as a
> kind of normalization. On another thread, Seb had suggested that we
> normalize by the number of messages sent (which I think is also
> useful) but using a baseline word like that could potentially do a
> better job at handling the situation where long and short messages
> could be very different. Do you know if there is published literature
> on using this technique?
>
Using "the" had two disadvantages:

- I ended up with very small numbers, so I couldn't actually do useful
computations before multiplying them by 1000 (the floating numbers were
too small and defaulted to 0 for some GARCH models)

- My variances were not completely random afterwards either, but rather
deterministically increasing. This appears to suggest that the measure
#mentions privacy/security simply gets more volatile over time, in a way
which is independent of the simultaneous increases of the word "the".

Tbh, I did not find a lot of material on time-series where volatility
strictly increases over time - normally the idea of making all the
transformations (moving average, normalization, filtering, etc) is
exactly to get rid of such effects.

> And I also think the periodicity testing is interesting, and I like
> the finding that maybe privacy advocates are more willing to work on
> the weekends. That might actually be an interesting way to identify
> volunteer participation. My impression is that grad students,
> advocates, people participating as unpaid volunteers are more likely
> to send emails on nights and weekends than those who participate in
> Internet governance as a formal part of their job.
>
Corinne had very interesting qualitative feedback when I told her these
results relating to her interviews with people. Maybe she would share?

best regards,

Amelia
> Cheers,
> Nick
>
>> On Feb 12, 2019, at 7:42 AM, Sebastian Benthall <sbenthall at gmail.com
>> <mailto:sbenthall at gmail.com>> wrote:
>>
>> Thanks for sharing!
>>
>> Would you be able to share any background on the project you are
>> working on?
>>
>> Any feedback on the tools that would make them more helpful?
>>
>> Thanks,
>> Seb
>>
>> On Tue, Feb 12, 2019, 5:07 AM Amelia Andersdotter
>> <amelia at article19.org <mailto:amelia at article19.org> wrote:
>>
>>     dear all,
>>
>>     perhaps of interest to some on the list. i was particularly amused by
>>     figure 6 and 7.
>>
>>     best,
>>
>>     a
>>
>>     -- 
>>     Amelia Andersdotter
>>     Technical Consultant, Digital Programme
>>
>>     ARTICLE19
>>     www.article19.org <http://www.article19.org/>
>>
>>     PGP: 3D5D B6CA B852 B988 055A 6A6F FEF1 C294 B4E8 0B55
>>
>

-- 
Amelia Andersdotter
Technical Consultant, Digital Programme

ARTICLE19
www.article19.org

PGP: 3D5D B6CA B852 B988 055A 6A6F FEF1 C294 B4E8 0B55