Some interesting data has been released today by the Forget.me service, regarding Europe’s so-called ‘right to be forgotten’ (rtbf) court ruling which seeks to balance the region’s individual privacy rights against the algorithmic power of search engines — especially Google — to determine which personal information is foregrounded within public search results.
The rtbf ruling has drawn the loudest and angriest criticism from search engines, free speech advocates and the press — with attacks generally declaiming the ruling as ‘censorship of information’. However new data on the type of requests being submitted suggests perceived risks to the freedom of the press and ‘human knowledge’ are being exaggerated.
In fact this data suggests most requests for de-listing personal information focus on far more parochial targets of social media sites and directory websites.
The European Court of Justice ruling requires search engines to accept requests from private individuals to ask for outdated or irrelevant personal information to be de-indexed. The Forget.me online service offers to hand-hold Europeans through the process of submitting these de-listing requests, although requests can be submitted direct to Google or other search engines.
So far Reputation VIP, Forget.me’s parent company, says it has garnered 21,000 registrations for the service after three months of operation.
Today it’s released data based on a study of 10,787 URLs submitted to Google via its service in three countries: France, the U.K. and Germany. While this is only a partial study of its data, and only a portion of the overall de-listing requests being submitted to search engines in the region, it nonetheless sheds some light on the types of content Europeans are most objecting to. And is especially interesting in light of the most vocal critics of the rtbf.
Reputation VIP told TechCrunch it focused this study on the three named European countries — rather than using all requests submitted via its service — because it couldn’t find a robust and comprehensive list of press websites for additional European countries (which are also covered by the rtbf ruling). It said it used the following Wikipedia lists to identify press websites in France, the UK and Germany.
A spokeswoman for the company added that the three countries also represent more than 65% of the de-listing requests Forget.me has fielded thus far.
Aside from a catch-all ‘others’ category, the largest proportion of rtbf requests covered by the Forget.me data-set ask for links to be de-listed on social networks and communications platforms (21.3%); followed by requests pertaining to directories/content readers (14.1%). While just 4.8% of requests target content on blogs; a further 3.6% target press websites; and only 0.2% directly target Wikipedia.
The Wikimedia Foundation, the not-for-profit organization behind Wikipedia, strongly condemned the rtbf ruling back in August, warning it poses a threat to its mission of providing ‘free access to the sum of all human knowledge’. Wikipedia co-founder Jimmy Wales has also been a vocal critic of the European Court of Justice ruling, telling TechCrunch in June that it amounts to “suppression of knowledge”.
However the Forget.me data suggests those criticisms are overblown, and that the individuals who are concerned enough about their personal privacy online to make a de-listing request are largely not people whose personal information has been published by mainstream online media.
Wikimedia publishes a list of the de-listing notifications it receives here — which also indicates that its content has so far been targeted by very few of these de-listing requests.
Frankly if anyone should be worried here it’s Facebook — which, according to Forget.me, comes out as the top domain for type of website targeted by a de-listing request, followed by Google.com (covering Google+, Google Groups, Google product forum, Picasa, Google profile); YouTube, Twitter and LinkedIn.
The European Court of Justice ruling specifically pertains to search engines — judging them to be data controllers and therefore requiring them to accept and process de-listing requests. But a social networking platform of the massive size of Facebook arguably holds a similar power to foreground users’ information as Google does.
Mainstream online media not likely to be forgotten
The Forget.me data also shows a breakdown of de-listing requests that have been refused based on content type. This shows Google is most likely to refuse requests to de-list content on mainstream online media channels — with 100% of Wikipedia de-listing requests refused thus far, and 93% of those pertaining to press websites denied.
On the flip side, the type of online content Google is most likely to agree to de-list a private individual’s person information from is the directories/content readers category — so sites that, for instance publish individuals’ contact details. Here just 28% of requests to de-list data were refused by Google, according to Forget.me.
On the social network side, Google is still denying the majority of de-listing requests — turning down 61% of Forget.me requests thus far.
The most commonly targeted directories/content readers for de-listing requests are yasni, 192.com, yatedo, profileengine.com and dastelefonbuch.com, according to the data.
For press websites the U.K.’s Daily Mail comes out on top, accounting for 8% of submitted requests, followed by The Guardian (6%) and stokesentinel.co.uk (6%); The Telegraph (5%); and nouvelobs.com (3%).
For blogs, blogspot.com (36%) and wordpress.com (23%) websites account for the lion’s share of de-listing requests, followed in a distant third place by tumblr.com (6%).
Reputation VIP has also crunched data on where within Google results de-listing requests fall. Close to half (45%) of URLs being submitted for de-listing appear on the first page of Google results; while a fifth (20%) appear on the second page; and just 10% appears on the third page. A quarter of results pertain to Google results that lie beyond the third page of a search result.
This suggests the greatest concern for individuals who are making de-listing requests is the prominence of their personal information on Google — i.e. rather than a desire for total erasure.
Again that stands in contrast to the most vociferous criticisms of the rtbf ruling which have focused on the binary idea of ‘knowledge deletion’ or censorship — with free speech campaigners refusing to see value in the concept that some information storage should be layered or sedimented. Aka the idea that gradually forgetting something can be a core enabler in societies by facilitating change to personal identities, and allowing our social relations to adapt and grow.
Individuals’ relationships with their personal information — as reflected in this Forget.me data-set, with its focus on social network outlets and prominent search results — certainly appears more nuanced than critics of the ruling have suggested.