The Information Commissioner’s Office issued guidance on how a Freedom Of Information request made via Twitter could comply with the statutory requirements. A subsequent decision by the First Tier Tribunal, which solicitors had hoped would provide more clarity in this area, has instead blurred the issue further.
Twitter launched ten years ago this month. In its first few years, it enjoyed a fast-paced, if fractious, relationship with the British legal world. There was the “Twitter Joke Trial” where a tweet from a frustrated traveller in January 2010 led to a criminal conviction and three appeals which finally overturned the conviction in July 2012. There was the anonymous tweeter who, in May 2011, revealed allegedly key details of several super-injunctions rendering them somewhat pointless as they then spread like wildfire over social media (which continues to be a thorn in the side of super-injunctions).
Even as far back as October 2009, Twitter played its role in the creation of a new type of court order called a “Blaney’s Blarney Order” whereby an order was permitted to be served against an anonymous individual via Twitter (Donald Blaney v Person(s) unknown, (not reported)).
Such forward thinking looked set to be adopted by the Information Commissioner’s Office who, in July 2011 via its monthly newsletter, opined that a Freedom Of Information request made via Twitter against a public authority could be valid as long as the public authority’s Twitter handle was cited, and the real name of the individual making the request was provided, if not in his/her Twitter handle then in a linked profile. In addition to making requests via Twitter, the ICO also suggested that public authorities could respond to requests via that medium: as to the 140-character limit imposed on tweets, the ICO suggested that the public authority could post its response on a website and tweet a hyperlink to that, or ask the individual for an e-mail address.
However, it appears that the courts are not so ready to embrace the ICO’s pragmatic approach. In November 2015, the First Tier Tribunal dismissed the appeal of a decision by the Information Commissioner, holding that the tweet in question was not a valid Freedom of Information request. The Tribunal held that (i) it did not contain the appellant’s real name (this was available by clicking on his Twitter handle which led to his profile, but the Tribunal stated that the Freedom of Information Act 2000 does not impose a requirement to search elsewhere for this information); and (ii) a Twitter handle did not constitute an address for correspondence (the Tribunal stated that this must mean an address that is suitable for correspondence between the public authority and the requester, and that a method of communication which is limited to 140 characters is unsuitable for such correspondence).
In August 2014, the Department for Work and Pensions sent a tweet from its Twitter account about the percentage of genuine jobs on its ‘Universal Jobmatch’ service. The appellant, Bilal Ghafoor, replied “@dwppressoffice FOI request: copy of internal report or assessment, including all data considered and method, for this assertion. Thanks.”
There was then an exchange by e-mail and Twitter, culminating in the DWP’s response (via e-mail) that there was “no set report or assessment” for its assertion regarding the percentage of genuine jobs on Universal Jobmatch and that “this percentage was calculated by the team responsible for removing non-genuine accounts from the service from their own records”.
Mr Ghafoor complained to the Information Commissioner that the DWP had not complied with his request relying on three grounds:
1. That the DWP was in breach of s.11 of the FOI Act because it had provided its response via e-mail, and not Twitter as Mr Ghafoor had asked;
2. That the DWP was in breach of s.10 of the FOI Act because it had responded outside of the statutory time limit; and
3. That the DWP was in breach of s.1 of the FOI Act because it was incorrect to state that no relevant information was held.
In opposing the complaint, the DWP contended that the FOI Act provided that (i) it could respond by any means reasonable in the circumstances; (ii) time did not begin to run from Mr Ghafoor’s initial tweet because it did not state the applicant’s name and correspondence address as required of a valid FOI request; and (iii) it was correct to say that no relevant information was held.
In relation to (ii) and (iii), the Commissioner decided in Mr Ghafoor’s favour and held that the DWP was in breach and required the DWP to issue a new compliant response or a valid refusal notice.
However, in relation to (i) the Commissioner agreed with the DWP. It held that the DWP was entitled to refuse to provide a response via Twitter on the basis that s.11, which allows an applicant to express a preference for the form or format of communication, did not apply to communicating simply – as was the case here – whether the requested information exists or not (known as the duty to confirm or deny). Rather, s.11 only applies to when the information requested does exist and is being provided.
Apparently unsatisfied with the decision, Mr Ghafoor appealed. He argued that:
(a) the drafting of s.11 must have been an oversight because permitting the use of one means of communication for the duty to confirm/deny, and then a different one to actually provide the requested information was counter-intuitive;
(b) permitting a public authority to ask for an address different from that stated in the FOI request would be a breach of the 3rd data protection principle (i.e. that an organisation should not hold more information than it needs for the purpose it is holding it for);
(c) where a FOI request had been made by a tweet, it should be responded to by a tweet where reasonably practicable;
(d) the DWP frequently published information via tweets which contained links, thus avoiding the 140-character limit problem, and therefore the DWP could do the same for a FOI response; and
(e) he had made his FOI request through the same medium through which the assertion he was challenging was made, and it was reasonable to expect the DWP to reply through this medium and give equal publicity for the evidence as it did for its claim.
The First Tier Tribunal maintained that the DWP had not breached s.11 as it did not apply to the duty to confirm/deny. However, it agreed with Mr Ghafoor that a public authority could not demand a different address for correspondence from the one stated by the applicant in a FOI.
Unfortunately, given the efforts to which he went, the Tribunal held that Mr Ghafoor’s FOI request was not valid in the first place. It did not contain his name or an address for correspondence as required by s.8 of the FOI Act. In so finding, it ruled that ‘name’ must mean ‘real name’ and ‘address for correspondence’ must mean an address that is suitable for correspondence between the public authority and the requester, and that a 140-character limit meant this method of communication was unsuitable.
Despite the ultimate outcome for Mr Ghafoor, this decision leaves open the possibility that Twitter could be suitable for FOI requests if certain hurdles can be overcome, for example had Mr Ghafoor signed off his tweet with his real name, or if his Twitter handle was @BilalGhafoor.
It is unclear, however, whether the Tribunal’s Decision applies only where individual correspondence would exceed Twitter’s 140-character limit, or whether it means no FOI requests made via Twitter can ever be considered valid even if both the FOI request and the response would be within the character limit. There is no consideration of whether, per the ICO’s guidance back in July 2011, attaching links (or screenshots) to Tweets would make a Twitter handle a suitable ‘address for correspondence, or of Twitter’s ‘direct message’ function which has a much bigger limit of 10,000 characters. There is also, of course, the possibility that Twitter may increase the character limit for Tweets generally. It has made various changes to its service in the past year: in addition to the increased character limit for direct messages, it also added an option for tweeters to permit anyone (not just followers) to direct message them and, in May 2016, Twitter announced that it would make changes to reduce what counts towards the 140-character limit on ‘public’ Tweets. This limit, which is largely a vestige from when Tweets were originally sent by text message, is a key issue for Twitter and its users.
However, until another Twitter FOI request comes before the courts, these issues will remain unresolved.