The power of words

Like a lot of people in my corner of the internet, I’ve recently become attracted to the daily Wordle puzzle published online. I first came across it via Twitter, when some of those I follow started around the turn of the year to publish some strange coloured runes accompanied by a wee bit of unfathomable text – and then someone posted a link to a short news item (now lost to me, I’m afraid) giving me a clue as to what it’s all about.

For the uninitiated, this version of Wordle (I first encountered the word years ago as a form of software for drawing word clouds to help make visual sense of large blocks of text, now with an address slightly shifted to edwordle.net) is a simple word game in which you have six goes to guess a five-letter (proper) word. The response to each guess you make is a green tile for a letter which is in the right place in the target word; an amber one for a letter which is in the target word but not in the same place as in your entry; and a grey tile for each letter in your entry which is not in the target word. It’s a bit like the ‘Mastermind’ board game in the 1970s, only with words as the target rather than coloured pegs; and you get fewer goes (but, as a result, with more specific information in response). I didn’t really get Mastermind (others did…) and probably the key for me is the use of words here rather than coloured pegs.

Each day’s Wordle is posted on a simple website – www.powerlanguage.co.uk/wordle – and its ‘inventor’ is ‘Josh Wardle’ who describes himself as an ‘artist, product manager and engineer’ and whose confessed purpose is to use the game to focus on human interaction. The result is that your Wordle results are easily shareable: the link to do so translates the outcome of your game into the runes I described above (and of which you may be able to see some examples down there on the left margin of this page) from which it’s a simple matter to paste into your social(s) of choice. Twitter, in my case. There is an unspoken code of conduct between players in which there are no spoilers; and the tiles within the ‘share’ button hide all the letters of your guesses so, on viewing others’ achievements, you know neither the solution nor the letters which are no longer in play.

It is, ultimately, just a bit of fun although these days there’s not that much on the internet that’s ‘just for fun’.

Yesterday’s Wordle indeed caused a bit of a kerfuffle, as you might well be able to judge from my outcome screenshot below (and which will lead to some merriment among my language editing clients who are, at first glance, bemused by my question as to whether they want what Microsoft, rather loosely, calls ‘US English’ or ‘UK English’). Some were angry at the apparent misdirection which had led to their stats of solved puzzles being undermined, or at solutions being found in more goes than hitherto, as well as to the numbers of words now in play which do not exactly ‘favour’ UK players; others were more resigned about the clues this gives to the state of the world (and, you might think, ‘global Britain’s’ place within it).

Given that you start with a blank sheet of paper, the game is surprisingly easy to get right and I wouldn’t be surprised if my ‘completed’ stats (75%; based on two out of eight games not being completed (I’m not exactly an early adopter…): one the first as I put in any old nonsense to check how it worked; the other another early attempt where I had a 50:50 at go six and plumped for the wrong one) were not among the lowest of regular users. I did, however, spend a lot more time on this one than on the others. After go two, I knew the word had an ‘a’ in place two and ended with an ‘r’; word endings of ‘-ur’ are not that common; while not that many end in ‘-or’ either (depending, of course, on your dictionary) although there aren’t many other choices than these. I got lucky with go three (I was thinking of valorise – and don’t get me started there, either!) which told me it ended in ‘-avor’ but I still wasn’t thinking of the US approach to spelling, running through my options twice before plumping, with a fair degree of trepidation, for ‘favor’ (as there was nothing else it could have been).

The feeling of being let-down was present, although brief: ‘Josh’ is American so this sort of problem and these sorts of feelings simply wouldn’t occur. And, of course, all these games have to start with some sort of a dictionary – and therein lies a world of debate between users of English. I’m not complaining – keeping all this in the air for other people does help to keep me in work 🙂

In my case, I was more upset about the apparent misdirection: the .co.uk suffix in the website’s domain name had, for me, led to a non-thinking, automatic presumption of the use of ‘UK English’ on the site (although I also know that country code top level domains haven’t been restricted to the countries concerned for over twenty years now – and which process has made a bit of money for Tuvalu (.tv) and Montenegro (.me). Being ‘UK English’ myself, that sort of presumption comes anyway as naturally as ‘US English’ would to an American. Other languages wouldn’t have this sort of problem – and I can easily see versions of Wordle working very well at powerlanguage.de, powerlanguage.ro and powerlanguage.ba, for instance. (Well, maybe not .de on the grounds that there may well not be enough five-letter words around in German.)

But we are precious about our stats and our record and achievements; and about a presumption of simple honesty of those who deal with us, even if we are partnering with an algorithm of some kind. We don’t like being played for fools; and being led to agonise over a combination of letters that is really quite simple does make us feel foolish. There are also cultural values at stake here, too, in terms of the word choices: interestingly, today’s word (achieved by this writer in four goes) is, I suspect, likely to be achieved more quickly and more readily by users of ‘UK English’.

I always start with ‘raise’, by the way: it’s the word that makes the best use of the most-used letters in the English alphabet. Though whether those are the same most-used letters in five-letter words is a different question – and, perhaps, a job of work already being engaged in by ‘Josh’. (NB I can’t see any data privacy protocols around the site or which become apparent on first use and, like anything else, Wordle – for all its desire to get ‘humans’ talking around shared issues – is surely likely to be a data collection operation somewhere along the line.) Others might find more fun in starting with a word inspired by current affairs – party, for instance – and making connections that way. Going from ‘party’ to ‘slump’ (the Wordle from a few days ago) would have been amusing, even if we know that the day’s word of choice is not made by a human but by an algorithm with no interest in the current affairs of ‘UK England’.

[Edit 27 January 2022: Wordle has been rumbled – though indeed that won’t stop me playing and, occasionally, sharing. It seems as though some of my suspicions in this piece were a little wide of the mark, though the thought that inspired them – that we need to be more careful online – still remains valid. And, yes, some things on the internet are indeed just for fun. And that’s a happy thought.]

[Edit 2: 1 February 2022: Well, indeed I spoke too soon since Wardle yesterday sold Wordle to the New York Times for what the company said was ‘in the low seven figures’. Not a bad return for an apparently simple piece of coding drawn up originally to give a bit of fun to his wife – and good luck to him. He promises that it will remain ‘free’ and that he is working with the NYT to ensure people’s stats are transferred (though it seems the NYT is only promising ‘initially free‘). But then, nothing is ‘free’ (or, indeed, in Wardle’s own words, can ‘just be fun‘) on the internet – the NYT exists behind a paywall and, while Wordle, once transferred, may well continue to be ‘free’ (at least for a while), it will be in return for at least your registration on its website and the cookies it will place on your computer as a result to track where you go. Plus, quite probably, a few ads. That may well be a ‘no thanks’ from me.]

Automated recognition software: your rights in the public space

This is the text of my summer 2021 column for BECTU’s Stage, Screen & Radio, slightly extended and with added links. Sometimes the column – especially when published several months later – gets overtaken by events; occasionally concurrent events give it added relevancy and that’s the case with this one, with news this week that the Information Commissioner is stepping in over the case of facial recognition technology in Ayrshire schools ‘to speed up the lunch queue’; and with Eurostar testing the same to give ‘seamless travel across borders’ and a ‘touch-free journey through border checks’ (under plans originally announced last summer). As always, the language is of course interesting focusing on the upsides with little consideration of the (considerable) downsides. Passport checks – which already incorporate biotechnology – are one thing; whether school children are in a place to give informed consent for something as quotidian as school lunches is another thing entirely.

Anyway, on with the column…

The European Data Protection Supervisor – an agency which reinforces data protection and privacy standards – has called for a ban on the use of ‘automated biometric identification in public space’. This means a number of things connected with the use of what, for simplicity, we’ll call here ABI to categorise a range of features including, most obviously, facial recognition but also gait, voice, keystrokes and our other biometric or behavioural signals.

The EDPS is not concerned with the use of AI to unlock your smartphone, but it is concerned about the public space: law enforcement and also the wider commercial and administrative environments in which it might be deployed – for example ‘smart’ advertising hoardings and billboards, attendance at sporting and other mass events, in airport screening or wherever users access public services.

The call for a ban is clearly serious – but so is the context in which it was made: the European Commission’s legislative proposal for an Artificial Intelligence Act. This, the EDPS noted, did not address its earlier calls for a moratorium on the use of ABI in public, however otherwise welcome the initiative.

The UK has of course left the EU, but the Information Commissioner’s Office – the UK’s own data protection and information authority – is also concerned about these issues. A reference to facial recognition technology appeared very early in the ICO’s 2019/20 Annual Report; while the Office issued an Opinion on the use of facial recognition technology in law enforcement in October 2019. It also intervened in a judicial review on the use of such technology by South Wales Police – a review which the police lost on human rights and data protection grounds.

We know – and have done for some time – of the problems of ABI in distinguishing between people: it has a much lower accuracy record in correctly matching people of colour, women and those aged 18-30. Partly, this speaks to the lack of diversity amongst those developing ABI software and amongst those on whom it is tested; in either case, were the base to be more representative, its accuracy record may well be better.

This, in turn, speaks to the need for software development standards also to be more representative and more inclusive, and to take serious account of tightly-drawn standards of ethics.

(Whatever the comical faults of the LinkedIn jobs algorithm, it is AI that is responsible for diverting job advertisements in a way which reproduces the extent of existing occupational job segregation, and which may contravene sex discrimination laws, by sending grocery delivery jobs to women and pizza delivery jobs to young men).

Furthermore the EDPS spoke specifically of its concerns that AI ‘presents extremely high risks of deep and non-democratic intrusion into individuals’ private lives’ while the ICO being similarly exercised – expressly, and in very similar language, about its potential for ‘unnecessary intrusion into individuals’ daily lives’ – indicates a worry among regulatory authorities that there are unsettling data privacy and state surveillance aspects surrounding the use of ABI in this way.

ABI works on the basis of matching scanned images against a ‘watchlist’, deleting those where there is no match and otherwise prompting human intervention. What the authorities are concerned about is whether an individual could anticipate, and understand, their image (or data) being processed in this way; and whether this is both a necessary and a proportionate response. What you and I might be concerned about is how someone could put us on a watchlist – was it because we went on strike, perhaps, or demonstrated against racism? – and how the authorities would then be allowed to track us wherever we go without us knowing.

Unquestioning faith

Additionally it’s true that we tend to place a large amount of unquestioning faith in the results that machines give us. If our trust is not to be abused, we need to be confident that the ABI which lies underneath has been developed, and is being used, in a socially just way.

The South Wales Police case highlights that ABI could identify large numbers of people and track their movements. Few trade unionists – or others organising protest actions – will need a refresher course on what that might mean. The decision in this case recognises the need for precise legal boundaries on the use of ABI, something which EDPS also openly acknowledges, although what these will be has yet to be defined.

Where we impose limits on the use of surveillance technology, in a law enforcement capacity and in terms of our knowledge of our data rights and our trust, is something in which we should all be taking a keen interest.

The spy behind the screen

Interesting news today that Facebook has denied Admiral the right to use its data for its Firstcarquote app.

The news is undoubtedly welcome from a data freedom point of view – it could indeed lead to decisions which have all kinds of implications as regards the perpetuation of social bias based on race, gender, religion or sexuality – although I’m not quite sure that Jim Killock of the Open Rights Group is right to give Facebook as much credit for the – perhaps temporary – withdrawal of permission as his press release does. It is a step in the right direction, admittedly, so I can see where ORG is coming from – but it is an infinitesimally small one.

There is undoubtedly more to this than Facebook suddenly deciding to uphold the data privacy of its users. The app might well, as referred to in the news pieces, contravene Section 3.15 of Facebook’s platform policy, but Admiral has been working with Facebook on this app for months and the principle of data privacy seems a strange one to crop up at this – very last minute – stage. Besides, the whole point of Facebook is to collect data on user preferences, likes and dislikes, and views, the mining of which to monetise at some future point. That’s why so much money has been chucked at it in the past. Note that the rest of Section 3 of the platform policy – which is titled ‘Protect data’ – is not so much to do with the protection of individuals’ data but to ensure that the data that third parties receive is not then compromised (resulting, presumably, in its value being undermined). And note also that it is rather arguable whether Section 3.15 itself is actually contravened in practice by what the app is reported to do.

It’s probably not quite as simple as Facebook deciding that it wasn’t going to make enough money from the app, and therefore deciding to withdraw permission; or that it has other plans with other insurers which might be jeopardised by letting this partnership with Admiral go ahead. Either (or both) might of course be true: time will tell.

But, quite clearly: beware the claims to data privacy mounted by organisations the sole purpose of whom is to collect your data. As Red Riding Hood found out, grandma has big eyes all the better to see you with.