To Forget or Not to Forget...

On the one hand the internet is so good at remembering things that we have to create legal frameworks to force it to forget. On the other hand, in those instances where we expect the internet to remember things in the same way we used to rely on libraries link-rot and content-drift has resulted in valuable information no longer being traceable.

This article was first published in The Mint. You can read the original at this link.

In a recent interim decision, the Madras High Court ruled that the accused in a criminal case could have his name redacted from a judgement that acquitted him of all charges. The court pointed out that anyone can type his name into a [[search engine]] and be taken straight to the case in question, and, while he might have been acquitted of all criminal charges, the fact that he even figured in the case could harm his reputation—an unacceptable violation of privacy.

This is not the first time an Indian court has ruled on something like this. Earlier this year, the Delhi High Court was faced with the same issue when it was petitioned to remove a judgement from major search engines and online case-law repositories on similar grounds. In this instance, however, the court applied reasoning that was slightly more nuanced. While recognising that prejudice was caused to the petitioner by the fact that the case was easily accessible on the internet, it pointed out that, at the same time, the public had a right to that information and courts an obligation to maintain transparency in their records.

Two Sides to Every Story

The fact that information is easily accessible online is a double-edged sword. On one hand, it offers tremendous convenience, allowing us get hold of a near-infinite store of information whenever we need it. However, what this also means is that information that we would prefer no one found is relatively easy to discover. That technology makes information easy to find is no reason to remove all trace of it from the public record—even if that information could prejudice the reputation and privacy of those to whom it pertains. Any demand to do so must be balanced by the public interest to ensure that such information remains accessible.

It is for this reason that privacy laws around the world approach this issue with some circumspection. The right to be forgotten is typically only invoked when the information in question is no longer needed to fulfil the purpose for which it was originally collected, where the consent on which its processing depended has been withdrawn, or where there is no longer any legitimate interest to continue with its processing. However, where there is a legal obligation to retain that information, or where such retention is necessary in the public interest—such as to establish a legal defence against a claim or to comply with a legal obligation—a request to ‘forget’ that information must be rejected.

Whenever I discuss the right to be forgotten, I cannot help but reflect on how the machines we built to help us remember information we were prone to forget have become so good at their jobs that we now have to devise legal measures to compel them into forgetfulness. It’s the sort of delicious irony that characterises so much of our digital interaction. But, while on the subject of the internet acting as the collective memory of all humankind, there is another aspect of the Internet’s hive mind that we should consider.

In a recent article in The Atlantic, Jonathan Zittrain made the startling claim that the world wide web was decaying. He said that link-rot and content-drift were now so pervasive in academic circles that it had become the norm rather than the exception.

We have always known that there is no guarantee that hyperlinks pointing to information on the web today will continue to reference that same information tomorrow. The internet is, after all, a decentralised network. Every page on the web is hosted by some person or organisation, and we can only access it if that host continues to keep it active. There is no master index of the internet that keeps track of where information is stored — that can point us, like librarians, to where it might have moved. This is why no one should presume that information that is available on a website today will continue to be accessible at the same place tomorrow.

This has become a very serious issue. According to a 2016 study of scientific articles, as many as 75% of the web URLs—uniform resource locators—cited by them no longer pointed to the content originally referenced.

To demonstrate why this is so important, Zittrain cites the example of a 2010 judgement of Justice Alito of the US Supreme Court in which he had used a website link to explain his judicial reasoning. Today, if you click on that link, it takes you to an error message saying the page you’re looking for no longer exists. It is particularly piquant to think that information necessary to explain the reasoning of a judge of America’s top court has vanished, making his judgement that much harder to understand.

As the internet serves more and more of our information needs, we have begun to think of it in the same way that we used to think of physical libraries—as a permanent storehouse of knowledge where we can reliably expect to find information that we had previously discovered. To the contrary, the internet shifts, like the sands of a desert, and no one should expect that the path taken to access information will reliably remain the same for all time to come.

We will only truly understand the internet once we fully come to terms with the contradictions that it contains. In certain instances, it retains information for so long and with such accuracy that we need build legal constructs to force it to forget. Yet there are numerous instances where it falls well short of reliable repository of knowledge that we expect it to be. How do we build a legal framework that can effectively stunt the internet’s capacity to remember things when, in fact, its elephantine memory is not all that it’s cracked up to be?

To forget or not to forget. That is the question.