Ad-Block, Game Theory and The Guardian

I read two blog posts this morning that seemed to be crying-out to be connected together. So all credit to their authors, and a tiny bit to me for the meeting.

The first was by Jamie Madigan, who writes the terrific Psychology of Video Games blog, looking into the reasons people do (or don’t) behave badly in multiplayer videogames. People discover little cheats in videogames that can advance their score but annoy everyone else. Whether to use them anyway is an example of the ‘Prisoner’s Dilemma’. According to Game Theory, the dominant strategy is to use these cheats.

[Explication: your opponent has the option to use the cheat as well. If they do, and you don’t, you lose. If you do, and they don’t, you win. If you both do, then it’s equal. The worst that can happen from using the cheat is that the stakes are even. On the other hand, if you don't use the cheat, then the worst that can happen is you losing. That's worse than the stakes being even: so use the cheat.]

However, the consequences of everyone using the cheats is mayhem and no fun for anyone, so it’s actually also an undesirable outcome, but less undesirable than losing. Everyone cheating rather than playing the game properly. But so long as the strategy exists and can be executed in a way that’s undetected, the rational decision is to continue the abuse. The way to counteract this for developers and publishers is to close down the cheat strategies or publically identify the abusers so that future potential opponents will either (a) avoid them or (b) use the same strategies as the abusers. Identification and iteration of the same game conditions turns the short-term gain into a long-term loss*. Creating a state of uncertainty over whether abusers will/can be identified can also work.

[*Actually, the maths says that continuing to cheat still remains dominant, even when the cards are on the table, but humans are rarely mathematical creatures. People are complicated and irrational: winning isn't always the overall goal for them. Some people don’t play the dominant strategy anyway, because of a sense of honour or fair-play. On the other hand, some people always will, despite the consequences, because they don’t care. (They’re ‘griefers’ in videogame jargon).]

The second post was by Bobbie Johnson on the Guardian website about the Firefox and Chrome extension Ad-Block. If you use Ad-Block, then it stops the advertising banners and MPUs on websites from loading. That makes for a faster and smoother browsing experience for you as an individual. However, the websites that you are looking at lose revenue, since they probably sell their advertising on a CPM basis – cost per thousand views – it doesn’t matter whether you click on the ads or not. Not all ads are intended to be clicked on anyway, such as branding campaigns.

If everyone Ad-Blocks, then the site you love goes out of business. If no-one does, then it thrives. The ‘cheat’ is the idea that Ad-Block is still pretty-much a secret, or that most other people are more honourable than you. That you can block advertisements, but because hardly anyone else is using it, then the sites will still be OK.

So here’s the obligatory 2×2 matrix:

The best outcome is that your favourite sites prosper and continue, and you don’t have to see the adverts. The worst – the ‘everyone cheats but me’ scenario -  is that they go bust despite you not filtering ads yourself. The dominant strategy is to Ad-Block and hope very few other people do that as well. It will continue to be dominant until enough of us perceive free web media as a long-term game, are identified as free-riders or learn the consequences to our short-term victory.

We want sites to prosper, yes? So what do they/we need to do? They need to make viewing and interacting with their content a long-term game. Part of that is achieved by Bobbie’s column – if Ad-Block is worthy of a column in the Guardian, then it’s certainly not some sort of hacker secret anymore. It is the most-downloaded Firefox Add-on and the leading Chrome Extension. Any certainty that ‘everyone else’ will play a dominated strategy ought to disappear. Thus, the ‘best’ outcome, where you get a free ride on sites that prosper has gone. Take that out of the picture and the game looks rather different: playing fairly together is the new best option. They should probably publish figures on the footer of every page of the revenue lost to filters; maybe scale that into an ‘articles we were unable to commission this month’ widget, if the loss is large enough. Arguably, it should be possible to identify the users of Ad-Block (if it isn’t already) and serve them altered content.

We need to switch off the extension, with the recognition that this is a long game, even if our identities remain masked: it’s the future of free media on the Web. Our best outcome is a free-ride, on sites that are free-to-access anyway. The worst outcome is our favourite sites going bust.

With long-termism brought to the front of our minds, the best outcome is removing a little inconvenience; the worst would be a disaster.

picture credit: HDR cafe

[As you might be tempted to point out: I used to use Ad-Block but I have stopped].

Possibly related posts:

• Facebook as Game
• Out of Touch or Moral Guardian?
• Web 2.0 in the Guardian