Why Popularity Based Social Media is Bad for Democracy | TaraElla Report Lab

And how to fix the problem.

Welcome back to TaraElla Report Lab. Today, I want to start talking about how we can fix the broken marketplace of ideas we have right now. As I said before, the key to a healthy liberal democracy is a truly fuctional marketplace of ideas, where every voice gets a fair hearing, and where the people get to make a truly informed choice. And right now, the reality is that we are far from this ideal. Clearly, there is a lot of work that needs to be done.

In recent years, there has been a lot of attention on the role of the internet and social media in the propagation of echo chambers, polarization and division. Some people have painted the role of the internet and social media in an almost wholly negative way, as if the pre-internet age was actually better. I think that is a mistaken view, because a media landscape dominated by corporate media clearly ain't a fair marketplace of ideas either. The internet and social media have at least partially democratized our media landscape, and that is a very good thing indeed. The problem is that it just hasn't gone far enough. Therefore, we should look to the possible future of a truly level playing field, rather than look back to the past of a highly restricted media, to resolve our current problems with social media.

There are a lot of problems with the way the internet and social media exist today, but I am going to focus on what I believe is the most important problem: the obsession with 'popularity'. Almost every big social media platform that exists today has a strong focus on the 'popularity' of both creators and content. We see it in the way the number of followers and likes are prominently displayed. More importantly, these metrics of 'popularity' are heavily used in the algorithms of these platforms, so that only content deemed 'popular' is visible to many people. 'Unpopular' content is quickly buried, as if it never existed at all.

The trouble with this model is, what is 'popular' can be manipulated in multiple ways. At the most basic level, content creators are incentivized to create content that cater to the narratives and emotions of particular echo chambers, so as to maximize the number of likes received, because that is the only way the algorithm will pick up that piece of content and spread it to new audiences. As a creator who refuses to play that game, I know very well how frustrating taking a principled stance can be, in the world of popularity-obsessed algorithms. Content creators who have built a following based on pandering to certain narratives generally can't walk away from those narratives either, because of the very well known phenomenon known as 'audience capture'. They are scared that their audiences will turn on them, potentially leading to a massive loss of followers overnight. This constant need to pander to certain narratives means that almost every popular creator stays in line with their chosen narrative, which creates an echo chamber effect on their audiences.

Even more sinister is the potential for money to be able to distort this whole 'popularity' thing. On the internet, one thing that we all hate is 'bots', i.e. automated accounts designed to inflate the numbers of particular accounts and spread particular messages. All the major platforms are constantly on the look out for bots, which are promptly banned when detected, but this has clearly not been able to eliminate the problem. However, all the focus on bots is perhaps distracting us from an even bigger problem: that people or organizations with an agenda, armed with adequate money, probably don't even need to use bots to achieve what they want. They can instead target their content to real people who they know will be receptive to their messaging, through advertisements, events, and even cross-promotion on mainstream media outlets. This is why the platforms can ban all the bots, and still not solve the problem of the 'popularity' model. Indeed, banning the bots likely only shifts the advantage to those who can afford even more expensive means of manipulating the metrics of popularity, thus favoring ideas and agendas backed by major financial power, effectively making the problem even worse. This shows us that the 'popularity' model can't be credibly reformed, and must be abolished or at least radically altered, if we are to have a free and fair marketplace of ideas that is good for democracy.

The pitfalls of the 'popularity' model also extend beyond the world of what we usually consider social media, even to relatively credible platforms like Wikipedia. In particular, Wikipedia's notability criteria is something that I have been campaigning for years to reform. Basically, for something to be on Wikipedia, it needs to be featured on at least a certain number of what is considered 'credible sources', which is most often mainstream media or mainstream media adjacent outlets. Hence, ideas not picked up by mainstream media are effectively excluded from Wikipedia, which in turn could affect the way they are displayed or ranked on Google and other search engines. Just in the past few years, I have seen many cases of Wikipedia deletion wars, in which pages of borderline-notable people are proposed for deletion. If the subject happens to have a dedicated following, or certain media connections, then they might be able to salvage their page by effectively generating a new article or two that can be used to prove that they meet the Notability criteria. Thus, the Notability criteria effectively makes Wikipedia a popularity-based platform, just like social media. And it can obviously be affected by financially backed interests too. Let's be honest: we all know that it ain't too difficult for rich and well-connected people to get their ideas published in so-called 'credible sources'. Wikipedia's Notability criteria thus ultimately pose the same problem as the likes-driven algorithms of social media platforms.

Now that we've identified the problem, we need to do something about it. The cynical among you might say that, internet companies are ultimately businesses, and they will do whatever necessary to promote their business interests, which would include spreading content deemed to be 'popular'. However, it is not like that their business interests don't include the need to respond to widespread demands for change. And it is clear that demands for change do work on internet companies: for example, just in the past few months, YouTube stopped displaying the dislike count for videos, a change which is quite small indeed, but clearly in the right direction. Ultimately, the key to pushing companies to reform their practices is popular demand from its customers. This is why we need to keep this discussion going, keep the ideas for reform coming, and effectively elevate the cause of social media reform to social movement status. Imagine that, if the social media reform movement were to grow to become as big as the climate action movement one day, the companies simply wouldn't be able to ignore our demands.

In recent years, there have been growing calls for getting money out of politics. But as I have demonstrated, there really is no way to get money out of politics without also getting the obsession with popularity metrics out of social media and the internet more generally. Thus fundamentally changing the popularity metric-based internet landscape should be taken seriously as a core pillar of making our societies more democratic.