Wednesday, 29 October 2014

An app for everything and everything for an app

So tonight I've been watching how the Samaritan's announcement of their Samaritansradar app has been received.  It's given me a lot to think about and lucky old you, I'm putting my thoughts down here.  

First of all, this is not a take-down of the good work Samaritans do or a dismissal of the obvious good intentions behind the app's development in the first place.  I've called the Samaritans myself on more than a couple of occasions and found solace in the voice at the end of the phone, knowing that it's confidential and not judgemental . And so I guess this is the first thing that makes me feel uncomfortable about the app. It feels like a judgement is being applied to me.  Not just by the person who has chosen to use the app to monitor my tweets for 'keywords' that might indicate I am in a suicidal frame of mind but also by the computer-generated algorithm that identifies these keywords and by the people who put this list of of keywords together in the first place.  (Yes I feel judged by an algorithm - I'm a mental, go with it). 

The confidentiality thing bothers me as well.  My Twitter account isn't currently set to private so no, I can't claim a breach of confidentiality in the sense that someone shouldn't have read them and shared them. But when I phone The Samaritans I know what I say is between me and them. When I don't know who is using the app to monitor my tweets I feel like that confidentiality, and most certainly the implicit bond of trust I have with people who follow me on Twitter, has been broken. I choose to phone a helpline.  I don't chose to have any number of people give my twitter handle to an app. But then with this app, my choices are not considered.  In fact with this app I only get one choice  - to lock my account.  This is the solution from Joe Ferns, Samaritans Director of Policy, Research and Development, that you can just set your account to private. Right, well that is naive at best and hugely insulting at worst - yes, I could but THAT'S NOT THE POINT. Why should I have to stop being able to interact with Twitter as a whole in order to avoid an app I never signed up to?  (Incidentally, this is what the police told me way back in 2011 when someone was using my tweets to stalk me - plus ca change...)

Which leads me to another concern.  Usually for an app to be actively looking at your tweets you need to authorise them to.  Here, it's someone else identifying the need for an app to interact with me but I don't get a say and I don't get to revoke access. 

I've seen some tweets which say that this app is no different to any old Joe Bloggs scanning your tweets and picking out those with a particular subject. Ok.  First of all, this is an application developed by a commercial company on behalf of another organisation; and secondly, I don't much like the thought of old Joe Bloggs  (or Joe Ferns for that matter) scanning my tweets - it's downright creepy and exactly what my stalker did! I really don't think it's the role of the well-intentioned follower, and even less so of a well-respected charity's commercial partner. 

There are also some seemingly glaring data protection issues which Information Rights & Wrong has blogged about far better than I could (TL:DR - personal data is personal data - yes, even tweets in the public domain - and sharing/processing it with third parties is subject to the law).  I find it almost impossible to believe that no one involved in the R&D for this app, either at Samaritans or Jam, the digital agency who developed the app (I think it's the company I've linked too - it's hard to tell - happy to be corrected) had any notion that there might be any data protection issues involved.  

I'm a massively paranoid person, but is it too far fetched to worry what uses the technology behind an app that shares my personal data with third parties without my consent, could be put to by more nefarious organisations and people? is it? IS IT? 

So, to the app itself.  In the press release for SamaritanRadar it refers to Twitter followers as 'your friends' and judging by the introductory video it pre-supposes you are in physical contact with the person whose tweets you're monitoring you're concerned about. That's not really Twitter, is it? That's Facebook.  Or if you're lucky, Actual Real Life.  The fact that they've either massively missed the language and global reach of Twitter, or they've subverted it to make the app seem more touchy-feely and less intrusive, massively gets on my wick. 

To recap, I've not authorised the app, I've not consented to the sharing by a commercial entity of my personal data and I don't know which of my 'friends' are subscribed to the app which allows them to do this.  Presumably after all this the support given to them to help a person who really, truly might be about to end their life must be shit hot, yes?  Well. They get an email showing them the tweet and an option to get the following advice: 

Would it be wrong to suggest that your natural consideration and care for another you like well enough to be concerned about might cause you to try some of these things anyway? Would it be wrong to suggest it's a little bit of common sense?

I never expected when I first joined Twitter in 2009 that it would become a trusted and safe space for me to talk about the depths and despair of my mental health but it has.  It has for many.  It's a hard, sad fact that people kill themselves. That they genuinely see no place for themselves in the world, or the world they inhabit is just too damn hard. What is harder is to accept that it is their right to seek help or not.  You can show them support. You can be a friend  You can tell them you care. You really don't need an app for that.