[liberationtech] Trackography future

ɣęƈƞą vecnamcclaudio at gmail.com
Fri Jan 9 04:40:25 PST 2015


On Thu, Jan 8, 2015 at 7:12 PM, Francesco Vitali <fv at futurevision.it> wrote:

> Dear Claudio,
> Trackography project is very interesting, thank you for working hard on it
> and for sharing it with us.
> I highlight some upgrades I think you may introduce in the future
> evolution of the project.
> I’d also like to better understand the geopolitical knowledge that may
> emerge through the data you analyze.
>

Thank you!


>
> 1) Your software should take into consideration also the mobile tracking.
> The most of the internet traffic will be generated through mobile devices
> such as smart phones and tablets. Furthermore the most comprehensive and
> pervasive tracking is done on mobile users. Have you got any plan to
> implement specific apps/sw for this purpose?
>

I agree that cover mobile traffic is extremely important, is the most of
the traffic. but:

**if we are considering website accessed by mobile, is enough use all the
current trackography settings, and just "emulate" a small screen browser:
this is recognized as mobile device, and if different tracking is involved,
it is spotted. Is something costing very few time in the collection side
and I want do it ASAP [1] but make harder the visualization: I've too many
data right now, we've some issue in designing functional experience that
drive an user in the data.

** if, instead of websites, I've to rank app (will be very good, all the
idea is started from this revelation:
http://www.valuewalk.com/2014/01/nsa-angry-birds-revelations/ ), will be a
dedicated research rank - download - use - profile them - put the server in
a list, collect the privileges request.

 ** when some tester has used mobile provider, I found out that ICMP are
filtered. so no traceroute  ( http://en.wikipedia.org/wiki/Traceroute ) is
possible. that's quite crippling at the moment, and I've to lookup on BGP
routing table to get insight on the routing path.



>
> 2) What kind of data is collected on the users? I think your tool should
> give some more insights on the kind of data is collected by each tracker.
> Otherwise Trackography risks to verify only the actual global penetration
> of digital advertising networks, missing important insights.
>

what is collected in their database: we cannot know. what is sent by us:
depends from the browser/javascript/previous login the user has performed.

there are one of the next step of the project, aimed to categorize "how
much a tracker is invasive". but for sure, at the moment I'm just crawling
the homepage, I'm not analyzing what effectively is sent. good point.


>
> 3) Regarding the impact of the data collected by the trackers: is there a
> way to cross-reference the data collected on digital-media web sites with
> the news that are then re-proposed on news-networks (news papers, news
> agencies, tv…) and on social media? It is a difficult task, but it could
> add very much more power to your counter-tracking project.
>
> is barely impossible for us :) one of the thing we've to include in the
data under analysis, is the percentage of connected users per country, the
amount of citizen and the daily access per site (using Alexa). but we've
just considering the online media, no old-style-networks ..



> 4) In order to understand which is the geopolitical power highlighted by
> the data collected, I think that it is not enough to identify which is the
> country “collecting” the data.
>
> As a first step it would be useful to try to have some more information on
> how the data collected is used. Is there any insight that you could add to
> the project to verify if the data collected by Google or Facebook (just as
> an example) is used for geopolitical purposes?


we cannot know...


> For public diplomacy actions? For actions of strategic influence?


idem, how ? only what is publicly stated like "the obama electoral campaign
based on facebook", but beside that...


> Otherwise we see the ingredients of a “digital political bullet” but we
> miss the point when they are assembled for a geopolitical purpose. May be
> it could be helpful, for this purpose, to cross-check (with a political
> marketing approach) which services are offered by the different advertising
> platforms in reference with Trackography analysis.
>

Yes, this is going to be done comparing kind of customer and services
offered by the tracking company:

https://github.com/vecna/trackmap/blob/master/csv/companies.csv

Those kind of informations are transformed in machine readable format by
us, in order to be used in visualization and algorithm.
but we've to address this collection in a collaborative way, is something
planned in this year.

Thanks for the interesting points :)
bye,
Claudio


-- 
This account is intended for mailing list only. Personal email via:
vecna at globaleaks dot org
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://mailman.stanford.edu/pipermail/liberationtech/attachments/20150109/4a2e32aa/attachment.html>


More information about the liberationtech mailing list