Some graphs of OONI coverage (country, continent, test_name)
Here are some quick graphs visualizing the number of OONI reports in various dimensions. https://people.torproject.org/~dcf/graphs/ooni-coverage-20170428/Rplots.pdf https://people.torproject.org/~dcf/graphs/ooni-coverage-20170428/ (source code) My research group was interested in knowing, for example, which countries have high OONI coverage and which don't. These graphs show total volume of reports over time, also broken down by country and continent, and the most common test_names.
On Fri, Apr 28, 2017 at 02:09:44PM -0700, David Fifield wrote:
Here are some quick graphs visualizing the number of OONI reports in various dimensions.
https://people.torproject.org/~dcf/graphs/ooni-coverage-20170428/Rplots.pdf https://people.torproject.org/~dcf/graphs/ooni-coverage-20170428/ (source code)
My research group was interested in knowing, for example, which countries have high OONI coverage and which don't. These graphs show total volume of reports over time, also broken down by country and continent, and the most common test_names.
Hi David, These are very interesting. Is there any reason that you represented the per-country ones in the form you did rather than a choropleth? (I've always found them quite an intuitive way to put across this kind of data.) I've adapted your "plot_num_reports_by_country" function to output a choropleth. Happy to share the code if it's useful: http://www.pseudonymity.net/~joss/share/num_by_country-choropleth.pdf J -- Dr. Joss Wright | Research Fellow Oxford Internet Institute, University of Oxford Co-Director, Oxford University Cybersecurity Doctoral Training Centre Faculty Fellow, Alan Turing Institute http://www.oii.ox.ac.uk/people/?id=176
On Sat, Apr 29, 2017 at 05:04:17PM +0100, Joss Wright wrote:
On Fri, Apr 28, 2017 at 02:09:44PM -0700, David Fifield wrote:
Here are some quick graphs visualizing the number of OONI reports in various dimensions.
https://people.torproject.org/~dcf/graphs/ooni-coverage-20170428/Rplots.pdf https://people.torproject.org/~dcf/graphs/ooni-coverage-20170428/ (source code)
My research group was interested in knowing, for example, which countries have high OONI coverage and which don't. These graphs show total volume of reports over time, also broken down by country and continent, and the most common test_names.
These are very interesting. Is there any reason that you represented the per-country ones in the form you did rather than a choropleth? (I've always found them quite an intuitive way to put across this kind of data.)
We were interested to know, for example, what countries have more than 1,000 OONI reports. About a year ago when we looked, there weren't enough reports from anywhere in Africa, for example, to really say anything, but there is a lot better coverage now with the mobile client.
participants (2)
-
David Fifield -
Joss Wright