Category Archives: tracking

The Crown Estate and behavioural advertising

A new app for Regent Street shoppers will deliver targeted behavioural advertising – is it processing personal data?

My interest was piqued by a story in the Telegraph that

Regent Street is set to become the first shopping street in Europe to pioneer a mobile phone app which delivers personalised content to shoppers during their visit

Although this sounds like my idea of hell, it will no doubt appeal to some people. It appears that a series of Bluetooth beacons will deliver mobile content (for which, read “targeted behavioural advertising”) to the devices of users who have installed the Regent Street app. Users will indicate their shopping preferences, and a profile of them will be built by the app.

Electronic direct marketing in the UK is ordinarily subject to compliance with The Privacy and Electronic Communications (EC Directive) Regulations 2003 (“PECR”). However, the definition of “electronic mail” in PECR is “any text, voice, sound or image message sent over a public electronic communications network or in the recipient’s terminal equipment until it is collected by the recipient and includes messages sent using a short message service”. In 2007 the Information Commissioner, upon receipt of advice, changed his previous stance that Bluetooth marketing would be caught by PECR, to one under which it would not be caught, because Bluetooth does not involve a “public electronic communications network”. Nonetheless, general data protection law relating to consent to direct marketing will still apply, and the Direct Marketing Association says

Although Bluetooth is not considered to fall within the definition of electronic mail under the current PECR, in practice you should consider it to fall within the definition and obtain positive consent before using it

This reference to “positive consent” reflects the definition in the Data Protection directive, which says that it is

any freely given specific and informed indication of his wishes by which the data subject signifies his agreement to personal data relating to him being processed

And that word “informed” is where I start to have a possible problem with this app. Ever one for thoroughness, I decided to download it, to see what sort of privacy information it provided. There wasn’t much, but in the Terms and Conditions (which don’t appear to be viewable until you download the app) it did say

The App will create a profile for you, known as an autoGraph™, based on information provided by you using the App. You will not be asked for any personal information (such as an email address or phone number) and your profile will not be shared with third parties

autograph (don’t forget the™) is software which, in its words “lets people realise their interests, helping marketers drive response rates”, and it does so by profiling its users

In under one minute without knowing your name, email address or any personally identifiable information, autograph can figure out 5500 dimensions about you – age, income, likes and dislikes – at over 90% accuracy, allowing businesses to serve what matters to you – offers, programs, music… almost anything

Privacy types might notice the jarring words in that blurb. Apparently the software can quickly “figure out” thousands of potential identifiers about a user, without knowing “any personally identifiable information”. To me, that’s effectively saying “we will create a personally identifiable profile of you, without using any personally identifiable information”. The fact of the matter is that people’s likes, dislikes, preferences, choices etc (and does this app capture device information, such as IMEI?) can all be used to build up a picture which renders them identifiable. It is trite law that “personal data” is data which relate to a living individual who can be identified from those data or from those data and other information which is in the possession of, or is likely to come into the possession of, the data controller. The Article 29 Working Party (made up of representatives from the data protection authorities of each EU member state) delivered an Opinion in 2010 on online behavioural advertising which stated that

behavioural advertising is based on the use of identifiers that enable the creation of very detailed user profiles which, in most cases, will be deemed personal data

If this app is, indeed, processing personal data, then I would suggest that the limited Terms and Conditions (which users are not even pointed to when they download the app, let alone be invited to agree them) are inadequate to mean that a user is freely giving specific and informed consent to the processing. And if the app is processing personal data to deliver electronic marketing failure to comply with PECR might not matter, but failure to comply with the Data Protection Act 1998 brings potential liability to legal claims and enforcement action.

The Information Commissioner last year produced good guidance on Privacy in Mobile Apps which states that

Users of your app must be properly informed about what will happen to their personal data if they install and use the app. This is part of Principle 1 in the DPA which states that “Personal data shall be processed fairly and lawfully”. For processing to be fair, the user must have suitable information about the processing and they must to be told about the purposes

The relevant data controller for Regent Street Online happens to be The Crown Estate. On the day that the Queen sent her first tweet, it is interesting to consider the extent to which her own property company are in compliance with their obligations under privacy laws.

This post has been edited as a result of comments on the original, which highlighted that PECR does not, in strict terms, apply to Bluetooth marketing

4 Comments

Filed under consent, Data Protection, Directive 95/46/EC, Information Commissioner, marketing, PECR, Privacy, tracking

Privacy issues with Labour Party website

Two days ago I wrote about a page on the Labour Party website which was getting considerable social media coverage. It encourages people to submit their date of birth to find out, approximately, of all the births under the NHS, what number they were.

I was concerned that it was grabbing email address without an opt-out option. Since then, I’ve been making a nuisance of myself asking, via twitter, various Labour politicians and activists for their comments. I know I’m an unimportant blogger, and it was the weekend, but only one chose to reply: councillor for Lewisham Mike Harris, who, as campaign director for DontSpyOnUs, I would expect to be concerned, and, indeed, to his credit, he said “You make a fair point, there should be the ability to opt out”. Mike suggested I email Labour’s compliance team.

In the interim I’d noticed that elsewhere on the Labour website there were other examples of emails being grabbed in circumstances where people would not be sure about the collection. For instance: this “calculator” which purports to calculate how much less people would pay under Labour for energy bills, which gives no privacy notice whatsoever. Or even this, on the home page, which similarly gives no information about what will happen with your data

homepage

Now, some might say that, if you’re giving your details to “get involved”, then you are consenting to further contact. This is probably true, but it doesn’t mean the practice is properly compliant with data collection laws. And this is not unimportant; as well as potentially contributing to the global spam problem, poor privacy notices/lack of opt-out facilities at the point of collection of email address contribute to the unnecessary amassing of private information, and when it is done by a political party, this can even be dangerous. It should not need pointing out that, historically, and elsewhere in the world, political party lists have often been used by opposition parties and repressive governments to target and oppress activists. Indeed, the presence of one’s email on a party marketing database might well constitute sensitive personal data – as it can be construed as information on one’s political opinions (per section 2 of the Data Protection Act 1998).

So, these are not unimportant issues, and I decided to follow Mike Harris’s suggestion to email Labour’s compliance unit. However, the contact details I found on the overarching privacy policy merely gave a postal address. I did notice though that that page said

If you have any questions about our privacy policy, the information we have collected from you online, the practices of this site or your interaction with this website, please contact us by clicking here

But if I follow the “clicking here” link, it takes me to – wait for it – a contact form which gives no information whatsoever about what will happen if I submit it, other than the rather stalinesque

The Labour Party may contact you using the information you supply

And returning to the overarching privacy policy didn’t assist here – none of the categories on that page fitted the circumstances of someone contacting the party to make a general enquiry.

I see that the mainstream media have been covering the NHS birth page which originally prompted me to look at this issue. Some, like the Metro, and unsurprisingly, the Mirror, are wholly uncritical. The Independent does note that it is a clever way of harvesting emails, but fails to note the questionable legality of the practice. Given that this means that more and more email addresses will be hoovered up, without people fully understanding why, and what will happen with them, I really think that senior party figures, and the Information Commissioner, should start looking at Labour’s online privacy activities.

(By the way, if anyone thinks this is a politically-motivated post by me, I would point out that, until 2010, when I voted tactically (never again), I had only ever voted for one party in my whole life, and that wasn’t the Conservatives or the Lib Dems.)

6 Comments

Filed under Data Protection, Information Commissioner, marketing, PECR, Privacy, privacy notice, social media, tracking

I DON’T KNOW WHAT I’M DOING

As surprising as it always is to me, I’m occasionally reminded that I don’t know everything. But when I’m shown not to know how my own website works, it’s more humbling.

A commenter on one of my blog posts recently pointed out the number of tracking applications which were in operation. I had no idea. (I’ve disabled (most of) them now).

And someone has just pointed out (and some others have confirmed) that, when visiting my blog on their iphone, it asks them whether they want to tell me their current location. I have no idea why. (I’m looking into it).

These two incidents illustrate a few things to me.

Firstly, for all my pontificating about data protection, and – sometimes – information security, I’m not particularly technically literate: this is a wordpress.com blog, which is the off-the-peg version, with lots of things embedded/enabled by default. Ideally, I would run and host my own site, but I do this entirely in my own time, with no funding at all.

Secondly, and following on from the first,  I am one among billions of people who run web applications without knowing a great deal about the code that they’re based on. In a world of (possibly deliberately coded) back-door and zero day vulnerabilities this isn’t that surprising. If even experts can be duped, what hope for the rest of us?

Thirdly, and more prosaically, I had naively assumed that, in inviting people to read and interact with my blog, I was doing so in a capacity of data controller: determining the purposes for which and the manner in which their personal data was to be processed. (I had even considered notifying the processing with the Information Commissioner, although I know that they would (wrongly) consider I was exempt under section 36 of the Data Protection Act 1998)). But if I don’t even know what my site is doing, in what way can I be said to determine the data processing purposes and manner? But if I can’t, then should I stop doing it? I don’t like to be nominally responsible for activities I can’t control.

Fourthly, and finally, can anyone tell me why my out-of-control blog is asking users to give me their location, and how I can turn the damned thing off?

UPDATE: 30.06.14

The consensus from lots and lots of helpful and much-appreciated comments seems to be a) that this location thingy is embedded in the wordpress software (maybe the theme software), and b) I should migrate to self-hosting.

The latter option sounds good, but I have to remind people that I DON’T KNOW WHAT I’M DOING.

UPDATE:05.07.14

The rather excellent Rich Greenhill seems to have identified the problem (I trust his judgement, but haven’t confirmed this). He says “WordPress inserts mobile-only getCurrentPosition from aka-cdn-nsDOTadtechusDOTcom/…DAC.js via adsDOTmopubDOTcom in WP ad script”…”Basically, WordPress inserts ads; but, for mobile devices only, the imported ad code also attempts to detect geo coordinates”.

So it dooes look like I, and other wordpress.com bloggers, who can’t afford the “no ads” option, are stuck with this unless or until we can migrate away.

UPDATE: 11.07.14

We are informed that the code which asks (some) mobile users for their location when browsing this blog has now been corrected. Please let me know if it isn’t.

3 Comments

Filed under Data Protection, Information Commissioner, Personal, social media, tracking

Right now, you are being monitored

This morning, as I was leaving the house for work, I wanted to check the weather forecast so started tapping and swiping away at my newish iPhone to find the weather screen. I was startled to see some text appear which said

Right now, it would take you about 11 minutes to drive to [workplace address]

(It looked a bit like this (not my phone I stress)).

It was correct, it would indeed take me about that long to drive to work at that time, but I was genuinely taken aback. After a bit of research I see that this was a new feature in iOS7, (and, indeed, the weather widget was lost at the same time). Sure enough, I find that my new phone has been logging frequently visited locations, but must have also been logging the fact that I travel between A (home) and B (work) frequently. It is described by Apple as being a way to

Allow your iPhone to learn places you frequently visit in order to provide useful location-related information

I’m not going to argue whether this is a useful service or not, or even whether on general principles it is concerning or not. What I am going to say is that, because I’ve not had much time recently to sit down and learn about my new phone, to customise it in the most privacy-friendly way, I’ve been saddled with a default setting which has captured an extraordinarily accurate dataset about my travel habits without my knowledge. And yes, I know that tracking is a prerequisite of mobile phone functionality, but I would just rather it was, as default, limited to the bare minimum. 

p.s. to turn off this default setting, navigate to Settings/Privacy/Location Services [scroll to very bottom]/System Services/Frequent Locations and set to “off”

 

Leave a comment

Filed under Data Protection, interception, Privacy, surveillance, tracking