Advertising, Privacy and a Libre Internet

Discussion over the web’s dependency on advertising for revenue has over the years gained more and more attention with the development of various ad-blocking tools and their increasing use in various browsers and operating systems. Most recently, the topic resurfaced when it was announced that iOS 9 would allow ad-blocking applications to be installed by users and also force native apps to use HTTPS for all ads linking to websites or risk broken links. Media organisations and app developers reacted with various levels of concern on the friction between balancing revenue generation with the security of their users data. However, it is a rather blinkered mindset to assume advertising is the only means for income. Instead of settling on a place within this spectrum of how much user privacy platforms are willing to risk, we need to be thinking how to progress new ways for companies to earn revenue while also considering costs to users beyond those associated with payment.

On one hand, the Internet as we know it has thrived due to the prevalence of free products and services by establishing revenue streams based on advertising and sponsorship. Such models allow content creators and developers to offer their products for no monetary costs to the user while still making an income for themselves. However, thinking in terms of the cost to user privacy gives us a better perspective of the word ‘free’ and that free (as in free of charge) services can be not so free (as in freedom). This dual concept of the word free is specific within the English language and so for the purpose of this discussion, it is more useful to refer to the Latin words for these terms: gratis (free of charge) and libre (freedom). Software where the source code is open for others to modify, copy and audit is often referred to as libre to distinguish from those free of charge, but when considering the ability for open source code to contain vulnerabilities that reduce users privacy, is this an accurate use of the term? While an open source license and codebase is a step in the right direction, enabling auditing and improvements to code by outside developers, it does not go far enough and does not really apply to individual websites and services. Making the correlation between freedom and privacy is essential in understanding how to progress our ever increasing connected world, therefore it is useful to apply libre in a broader term to push for a better understanding of the effects on user freedom.

Due to the vulnerabilities within the structure of the existing Internet, finding a web service close to the definition of libre is rather difficult with the prevalent use of tracking cookies and third-party JavaScript. While advertising is not the only threat to user privacy on the web, it is a primary force as tracking cookies help maintain precise targeting for who to serve what ads to. Both Google and Facebook are highly dependent on such tracking mechanisms to correlate websites users have visited with the best advertisements and recommendations to serve. Services which implement Facebook Connect so that users can “like” their Facebook page directly from the website or “share” something to Facebook such as a news article are opening up direct routes for Facebook to track those who visit the site. Similarly, websites which either embed Google’s AdSense or advertise through AdSense are opening up a direct route to Google. So while websites are at least partially open source by design as anyone can review HTML and JavaScript code running in their browser, this does not guarantee user freedom from tracking and should not be considered libre.

To fix the pervasive use of tracking cookies on the web, we must take a step back and consider the incentives for services like Facebook and Google to use them in the first place. Going back to the concept of a thriving Internet based on free (gratis) services, these two companies are perfect examples of this process. Over the years, they have been able to leverage advertising as their main revenue streams to provide their users with free accounts and extremely useful services such as document storage, communication platforms, website analytics, event organising, web search, etc. Through this process, they have become large corporations serving most of the Internet population for much of our online experience which feeds a cycle where the more users they gain, the more data points they have for successful advertising. Unfortunately, we have become used to this gratis model at the expense of privacy and dependence on such centralized entities. The general lack of understanding in the effects on privacy with advertisements and tracking is a cause for concern. Not only does this type of tracking facilitate private surveillance by these companies, but it also feeds the monolithic public surveillance networks built by the NSA, GCHQ and others. Additionally, we have a bad user experience in current payment systems which do not work well for online services or content. By understanding better how we consume on the Internet versus non-digital consumption, we can begin to model new methods for improving this interaction and offer better solutions than advertising.

While advertising on websites and applications can be distracting and annoying for users, the fundamental reasons for pushing ad-blocking technology surrounds concerns for user privacy and the security of their data. That said, ad-blockers cannot be the only solution as it can very quickly become a game of cat and mouse where ad services are detecting the use of ad-blockers and ad-blockers detecting the detection of ad-blockers and so on. By reforming our perspectives on the dependence of advertising and work on generating better experiences for online payments, we can force a much larger shift towards a libre Internet.