De-anonymising social media

There’s little doubt that the growth of online social media has conferred substantial benefits by helping friends and families to stay in touch, assisting groups of like-minded people to exchange views and the like.

However, there’s a darker side in that social media also provides a ready medium supporting undesirable and outright criminal behaviour.

This darker side is exacerbated by (i) easy anonymity, (ii) the personal nature of much of the information exchanged on social media, allowing personal details to be easily acquired by those of malign intent and (iii) its capacity for the rapid spreading of “fake news”.

Individuals can be targeted for bullying, slander/blackmail, fraud and the like while would-be terrorists, overzealous activists, noisy minorities and foreign powers can and do exploit “confirmation bias” and the gullibility of a section of the general population to promote their ends.

Examples from the newspapers include various suicides, life savings stolen, incitement to civil disobedience including farm invasions, alleged tampering in elections, faked tear-jerking pictures promoting a particular political view of the current bushfires and most recently, the children of Tasmania’s Premier, Will Hodgman, being targeted. 

It is certainly the case that the victims of online bullying would do well to “walk away” before contemplating self-harm but, even if they all did, the intrinsic difficulties with social media remain. What was a benign youth in 2007 has, 12 years later, become a difficult adult requiring a degree of legal oversight.

Before widespread electronic communication, and particularly before widespread use of social media, such offences were brought to the courts traditionally using evidence including documents and wiretaps.  With the widespread introduction of mobile phones, the recording of personal details associated with each account was mandated to ensure these devices did not become anonymous.

Where possible, and as a matter of principle, it is better to expose social media to existing law and existing enforcement methods rather than introduce overspecialised specific legislation relating only to social media.  This approach avoids needless nitpicking distinctions such as those between what is a phone and what is a GPS navigator introduced by legislation relating to the use of phones while driving.

The matter then reduces to one of ensuring the availability of verifiable evidence of social media interactions and the parties to them. A major problem here is that the social media platforms are operated by offshore companies whose primary interest is financial rather than addressing legal difficulties in another country and that their expressed concern is much greater than any effective action. Zuckerberg’s performance in the US Senate hearing says it all.

That control of social media is possible is amply demonstrated by examples such as Xi’s China and Putin’s Russia; the problem is to find an effective, yet less draconian, solution that is acceptable in a liberal democracy.

My proposal is to implement what is effectively a CCTV camera on social media transactions.  It is vital that (i) collected data remains secure in the hands of some trusted agency and (ii) that access is limited and granted only via a court warrant.

There is already legislation requiring that Internet Service Providers (ISPs) keep track of metadata. These providers also have identifying details of those who use their service.  All that is required is that either the ISPs store transactions between their users and the various social media providers or, much preferably, relay such transactions to a secure storage facility elsewhere.

As is often the case, perfection is both unnecessary and its implementation needlessly demanding of resources; all that is actually required is a sufficient successful prosecution rate to deter potential miscreants.  I’d therefore conjecture that no more than (say) a rolling 4-week window would be required for such a system to be effective with, by “big data” standards, a fairly modest data storage requirement.

These stored transactions then become available as evidence to subsequent investigations in the same way that wire intercepts and written communications were before social media.

There is a loophole in that use of a VPN would conceal all user activity.  That difficulty assumes a degree of user sophistication and applies to all user communications so is not addressed here.

Further, it is certainly true that dating sites, such as Tinder, provide a forum where those of criminal intent can seek victims. However the objectives of a typical user of a dating site are very different than those of a social media site such as Facebook, Twitter or Snapchat with evidence of face-to-face contacts likely to be available from, for example, mobile phone records or emails.  On this basis monitoring dating sites is likely to be both less effective and less necessary than monitoring social media.

Despite these weaknesses, the overall thesis remains that simply removing anonymity from social media transactions, thereby facilitating prosecution of miscreants, with no need for assistance from any overseas body, would help reduce the current Wild West nature of the medium.

While the collection and storage of this data might seem to some as the beginnings of a police state, the users of social media would do well to remember that all this information, and much more, is already held, and actively processed for profit, by those corporations that operate the social media sites they so willingly patronise.