The thief, the question mark
and the revolutioner.

Introduction

"Man is least himself when he talks in his own person. Give him a mask, and he will tell you the truth" Oscar Wilde wrote.

These two different visualizations derives from the previous one and are meant to deepen the experimental research on images.

The glossary was necessary to show the variety of the objects in a picture. But which are the most used elements? How do they combine to each other? Which are the differences between the representation of anonymity in different countries?

How to read the visualization

The first part of the visualization is a network graph created to understand the relationship between the elements of the images. The objects contained on the images are now divided and positioned on the same level forgetting their hierarchy. The dimension of the bubbles show the amount of images containing that element; the connections between the bubbles are meant to underline the possible coexistence of the two objects linked.

Through the lateral buttons it is possible to visualize the network graph of the selected country. This is a way to figure out the differences between languages and dominions.

"Headless", "deleted face" and "covered face" are the starting points of the second part of the visualization. These three were the most relevant ways of showing a person (or eventually something else) on an image. This graph indicate for each (per country) google page result (100 images scraped) which is the median value and so where it is located on the page (from the 1st, to the 100th place). This is to understand their importance and to compare how they appear in different countries. In this visualization, as the previous one, the dimension of the bubble shows the amount of images containing that element.

Since the "covered face" is much more present in each country group, it has been chosen to concentrate the attention on that category to deepen the research. The new protagonists of the visualization are "balaclava", "shopping bag" and "Guy Fawkes mask". The graph can be read as the previous one.

How it has been done

The "picture glossary" of the first visualization could work as a legend to understand what the pictograms referred to. The objects of the images were represented through pictograms to easily show and understand them.

The purpose of these visualizations is to show and compare the differences between the representation of anonymity in the different countries. To do so, it was important to maintain the division into groups.

The network graph has been done through the help of Gephi software and then modified and visualized through Adobe Illustrator software. The second one has been made by manually calculating the median value for each element in each country result page.

Findings

People, question marks and computers are the protagonists of the anonymity representation on the web. They are presented associated to other elements or to each other.

The differences through countries are visible in the amount and in the connections between objects.

China and Korea, where "the real name authentication law" is strengthening the oppressive politics against freedom of speech, showed the higher number of "Guy Fawkes masks": the desire of being anonymous is probably seen as revolutionary. Otherwise Canada, the only country which has an anonymous protection law on behalf of freedom of expression, doesn't show anything that could be linked to this reality.

Generally the "covered face" element appears the most in each google page result and its median value is generally located around the thirty-third position and the sixty-sixth one. The "Guy Fawkes mask" is the heaviest group in all the countries but its median value is lower than the "shopping bag" element, even if its presence is not so large. In the second visualization too it's easy to understand how the "Guy Fawkes mask" predominates the page in Chinese and Korean results.

England, Usa and Germany are the ones which present the three categories in the lower amount but locating them into the first thirty-three positions.

Metadata

Timestamp: 18/12/2014

Data source: Google images

Related Protocol

Download data (530KB)