Google: The Process of 'Forgetting' People Is Still Evolving In response to questions raised by EU data protection agencies, Google published a point-by-point response explaining how it is currently processing the 91,000 removal requests it has already received.
Opinions expressed by Entrepreneur contributors are their own.
In May, the highest court in the EU made the sweeping decision that people have the 'right to be forgotten.' More practically, the ruling meant search giant Google must delete "inadequate, irrelevant or no longer relevant" links in certain cases, when requested by an EU resident.
Lots of people, it turns out, want to exercise their right to be forgotten. Since Google began processing removal requests on May 30, the company has received over 91,000 submissions, which ask for the removal of 328,000 links. While not all of these links have been processed yet, of those that have been reviewed, Google says it has removed over half (53 percent), asked for more information for approximately 15 percent and deemed that 32 percent were of public interest.
So far, the ruling has proven both controversial – privacy groups applauded the decision, while the House of Lords, among others, denounced it as "unworkable, unreasonable and wrong" – and messy to enforce.
The central problem, of course: What, exactly, qualifies as inadequate, irrelevant or excessive information? How does Google decide which requests to honor and which to reject?
Yesterday, in an effort to keep this process "transparent," Google published a letter addressed to a group of European data protection agencies, in which the company outlined its strategy for reviewing and removing requests thus far.
The Mountain View, Calif-based company emphasized that this process is still evolving. "Our approach will not be static," wrote Peter Fleischer, Google's Global Privacy Counsel. "We know it will change over time as data protection authorities and courts issue guidance and as we all learn through experience. We also know some tough debates lie ahead."
The tech giant has faced scrutiny from regulators over its decision to only remove links from European versions of its search engine, such as google.co.uk, google.fr, while allowing the links to remain on Google.com (the ruling only affects European searches, but EU residents have access to Google's US domain), a concern the company addressed in the letter.
"We actively redirect European users from google.com to the appropriate Google France ccTLD, and European users overwhelmingly use those services," Fleischer wrote, adding: "Fewer than 5% of European users use google.com, and we think travelers are a significant portion of those."
Another hot-button issue with regulators is Google's decision to notify publishers when one of their links has been removed from searches because of the ruling; Critics argue the practice can, in certain cases, reveal the identity of the individual who made the removal request, reversing the ruling's stated intention. What, regulators wanted to know, is Google's "legal basis" for sending these notifications?
In the letter, Google responds by saying a) it doesn't require a legal basis for sending these notifications, as they do not contain any personal data, and b) even if it did need a legal basis, notifying publishers of removed links is essential to the process, because it ensures transparency:
"The notice to webmasters both ensures transparency and makes corrections possible when a removal proves to be a mistake. We have received information from webmasters that has caused us to reevaluate removals and reinstate search results. Such feedback from webmasters enables us to conduct a more balanced weighing of rights, thereby improving our decision-making process and the outcome for search users and webmasters."
This gets at the larger issue at stake: It's very difficult, almost impossible, for Google to determine the validity of a removal request because in most cases, it is relying on information submitted by the requestor without any outside assurance of the information's accuracy. Naturally, this breeds biased, faulty, and incomplete submissions.
"An example would be a request to remove an old article about a person being convicted of a number of crimes in their teenage years, which omits that the old article has its relevance renewed due to a recent article about that person being convicted for similar crimes as an adult," Fleischer wrote. Clearly, if you want an article about your criminal past expunged, you aren't going to draw Google's attention to a report detailing your recent criminal activity.
In addition, the ruling opens a whole other can of worms, namely businesses can use it to sabotage one another: "Historically, we have seen see many cases of business competitors trying to abuse removals processes to reduce each other's web presence, so this corrective is important," Fleischer wrote. "Abuse of such processes is a well-documented phenomenon -- one academic study based on Google's published information about copyright-based removals estimated that over 50% of removal requests originated with competitors targeting each other's sites for removal from search results."
Google ends the letter with a host of questions of its own, including 'How should we differentiate content that is in the public interest from content that is not?' and 'Should individuals be able to request removal of links to information published?'
These are basic, vital questions; you'd think the EU, in making such a sweeping ruling, would have provided Google with a little more guidance on how to go about enforcing it. In any case, Google concludes, it remains "open to your feedback as to how you propose these processes should work."