It is fitting that Twitter – surely the closest thing in spirit to the 100m sprint in the Social Media world - has seemed omnipresent throughout the London 2012 Olympics.
The platform’s exponential growth was demonstrated by the statistic that more tweets were sent during London’s opening ceremony than during the entire Beijing Olympics in 2008. But with such growth inevitably comes significant challenges, many of which were in evidence during London 2012.
First and presumably foremost, at least from the company’s perspective, is translating its undoubtedly huge reach into greater advertising revenue.
A partnership with NBCUniversal, billed as a first step towards positioning itself as a media company, has seen Twitter take up the role of “official narrator” of the Games, curating an #Olympics page for its American audience and attempting to cross-drive traffic between Twitter and NBC’s Olympic coverage.
Of course, if there is one thing that Twitter is not, it is a respecter of “official narratives” and Twitter quickly became the focal point of fierce criticism of NBC’s Olympics coverage (much of which was shown on a considerable time delay) under the hashtag “#NBCFail”. Then a UK journalist who was amongst NBC’s most vocal critics had his Twitter account suspended after he tweeted the work email address of NBC’s Olympics Executive Gary Zenkel to his followers.
Twitter found itself at the centre of the controversy when NBC subsequently issued a statement highlighting that a member of Twitter’s staff working with NBC had reported the offending tweet to NBC. Many commentators also questioned whether Twitter in suspending the account had acted consistently with its own guidelines, which prohibit publishing “non-public, personal email addresses”.
Twitter’s General Counsel responded in a blog post which apologised for the proactive monitoring, but not for the decision to suspend the journalist’s account. The post stated that:
“we do not proactively report or remove content on behalf of other users no matter who they are. This behavior is not acceptable and undermines the trust our users have in us. We should not and cannot be in the business of proactively monitoring and flagging content, no matter who the user is — whether a business partner, celebrity or friend.”
It strikes me that there is a strong element of making a virtue out of a necessity here. After all, Twitter has a self strong interest from a legal perspective in distancing itself from any action which could imply that it has the capability or inclination to monitor content.
Monitoring policy and the law
US legal commentators interviewed by Reuters on the topic suggested that Twitter’s desire to re-emphasise their policy in this area could be a defensive measure to try and ensure they maintain the protection afforded to them under section 230 of the Communications Decency Act (CDA). The CDA protects operators of interactive computer services from being treated as the publisher of defamatory user generated conduct. The commentators argue that by editing tweets or giving users the impression that Twitter will assert general editorial control, Twitter could risk losing this general protection.
In the UK, the defences potentially available to online service providers against liability as secondary publishers of defamatory content can also be lost where editorial control is exercised. These defences are:
The “hosting defence” under section 19 of the E-Commerce Regulations 2002, which has served to provide a defence to Google in respect of libel actions for comments on blog posts hosted on Blogger and for Amazon in respect of a similar claim in respect of comments made in users’ book reviews.
The hosting defence will not be available where the online provider has “actual knowledge of illegal activity or information”, or where the customer of the service is acting “under the authority or the control of the provider”. Therefore, taking on a significant editorial role in respect of the service exposes the online provider to greater risk of being fixed with actual knowledge of illegal activity or being deemed to assert control over the users of the service.
The defence under section 1 of the Defamation Act 1996, which provides a defence for parties who adopt a secondary role in relation to the publication of a defamatory statement.
This defence has been characterised as containing a potential “Catch-22”, in that to avoid liability an online service provider must not be the “editor” of the statement but is also required to demonstrate that they took “reasonable care” in relation to its publication. However, there is no prospect of the defence being available where a significant editorial role has been adopted, whereas there is at least some prospect of benefiting from the defence where a notice and takedown procedure has been adopted and no editorial control asserted.
In conclusion therefore, there are sound self-interested legal reasons to think that Twitter will been keen to avoid any repeat of the kind of proactive monitoring of content which occurred in relation to its Olympic partnership with NBC, in order to minimise its risk of incurring liability for user generated content.
I think that there are also sound reasons under EU law for Twitter to defend its apparently blanket policy of “suspend first, ask questions later” for breaches of its private information policy – even in very marginal cases such as this. Adopting a very cautious approach in relation to private information makes sense because the hosting defence under the E-Commerce Directive (implemented in UK law by the Regulations mentioned above) doesn’t apply to breaches of EU Data Protection legislation – a fact which lead to three Google executives receiving suspended prison sentences for violating privacy laws after a video of an autistic boy being bullied was posted on Youtube.
Another pressure for Twitter to be censorious in its monitoring policies is the debate which has arisen in the UK during the Olympics about whether specific laws are required to deal with extreme cases of Twitter “trolling”, following the arrest of a teenager who had sent messages to Olympic diver Tom Daley telling him that he had let down his dead father.
A representative from the Association of Chief Police Officers (ACPO) responding to the controversy said that new laws weren’t required, pointing to existing powers under the Malicious Communications Acts of 1998 and 2003. Section 127 of the Malicious Communications Act 2003 creates an offence of sending a message via a public electronic communications network which is “grossly offensive or of an indecent, obscene or menacing character” and was recently confirmed to apply to Twitter messages in a High Court appeal case.
The ACPO representative also called on Twitter to do more to self-regulate arguing that “I think there is a case that if you are going to run it as a commercial organisation, then you have got to allow people to use it safely and securely, and have the processes in place where people are acting in a strange way – and the word troll comes to mind – then you get them off as quickly as possible”
Twitter therefore looks set to continue to face increasing and competing pressures in relation to its moderation policies, with users criticising it where it is quick to step in, while legal requirements can create an equal backlash if it is too slow in acting.
Better to be talked about…
The natural instinct of a risk adverse legal counsel might be to think that all of this “bad press” must result from some failure to properly mange risk and will have a negative impact on business. It was therefore interesting to read a story in US technology blog Techcrunch citing an unnamed source at Twitter claiming that high profile controversies such as the NBC suspension had led to a spike in Twitter’s user signup rate and were viewed internally as a “good thing”. This has the plausible ring of truth about it and should remind us that growing pains are better than the alternative for a tech company. Twitter would be forgiven for joining many of us in wishing that Rio was less than four years away!