(in)Secure is a weekly column that dives into the rapidly escalating topic of cybersecurity.
Trackers are something that most internet users tolerate, though few have much fondness for. At a time when Facebook and other online entities are being called out for their privacy problems, trackers are certainly part of that conversation. But a new study from anti-tracking extension maker, Ghostery, is equally concerned about trackers’ ability to slow down web browsing itself.
Hidden from view behind the pretty front end of modern websites, trackers make up a major portion of the web browsing experience. From managing personalized adverts, to social network integration, they have an important part to play in how the internet works today, but does all of that come at the cost of a fast and efficient internet?
The tracker tax
Ghostery is a tool that is decidedly anti-tracker (it’s not the only one), in that it makes it easy for web users to block any they want, and anonymize the information harvested by ones that they don’t. With its new “Tracker Tax” study though, Ghostery looked into the performance benefit of doing so, as well as the implicit privacy implications of not.
“We were fairly confident that websites perform better and work faster when trackers are blocked,” Ghostery’s director of product management, Jeremy Tillman told Digital Trends. “The purpose of the study was really trying to quantify that impact. Is it a minor thing, is it somewhat rare, or is a major thing that’s widespread?”
The study focused on the top 500 sites as listed by Alexa, which they claim is where the vast majority of traffic in the US spends its time. In addition, Ghostery identified them websites as the websites that implement a lot of third party tracking technologies.
The results were rather stark too. Ghostery found that just under 90 percent of the websites investigated contained some form of third-party web tracker. Those trackers, whether small or many in number, proved to have a significant impact on page load speed too.
The average page load times were around 20 seconds , but when all trackers were blocked, that time dropped to just under nine seconds.
To draw these load times, Ghostery employed a custom-built web crawler that was said to be comparable in usage to a consumer browser like Chrome or Firefox. However, the web crawler didn’t look at websites in the same way a human would. It was focused entirely on a “complete” state of a website’s generation, where there was nothing left to load. In reality, many sites would be functional before then, but that, according to Tillman, is half the problem.
“The truth is that a lot of websites are never that functional, in no small part because the trackers on those pages don’t stop loading,” he explained. “Many sites out there, if you go there without Ghostery, there’s almost like a slow leak because they […] never stop communicating with their servers and for some of these sites the lagginess that results […] is persistent and it never goes away.”
Ghostery’s study suggests that some sites never do quite finish loading. In the case of the slowest sites out there – Ghostery calls out Cracked.com, Si.com and Chron.com as the worst offenders – they could take as long as two minutes to register as “complete.”
“A lot of websites are never that functional, in no small part because the trackers on those pages don’t stop loading.”
Even if those sites become functional for the user before then, Tillman suggests that there is still an unnecessary drain on the website load speed, its responsiveness and on the end user’s system resources.
“New ad placements are being loaded, new trackers are being loaded, information is constantly being sent back and forth, so it’s a constant, persistent drag on the browser’s processing power,” he said.
Living tracker free
Clearly Ghostery’s intent with this study is to highlight not only the impact of trackers, but the effectiveness of its own tracker-blocking tool at remedying the problem. But do we actually want to live in a world that is entirely devoid of trackers? What kind of sites did the study discover that have no trackers at all?
“Craigslist would fall in that category,” Tillman said. “If you’ve ever been to it, it’s almost like a website from the 1990s, just a bunch of hyperlinks.”
No one is suggesting all websites should be like Craigslist, though. Indeed Tillman said that in some cases, trackers can offer benefits to people. Though typically they’re website owners, rather than website visitors.
“It’s hard to say that any one tracker is good or any one tracker is bad,” he said. “There are some that can sort of enhance the functionality of a page. A lot of website owners benefit from this out of the box nature of trackers, so without having to build that feature yourself, within minutes you can have a fully featured video player by using a third party technology.”
“It’s hard to say that any one tracker is good or any one tracker is bad.”
While those widgets may provide a quick ‘upgrade’ for a website’s functionality though, is that worth their cost in privacy and performance impacted by the trackers that go hand in hand with them?
“In those situations where a user is getting a deeper, more meaningful user experience, you could say that those trackers are providing utility to the user,” he said. “[That said,] from a user’s perspective, are any trackers good? It’s a different question.”
Meeting in the middle
Often times the modern web browsing experience can seem like a battle of two camps: The website owners and the users. One side wants to make money from their service, track how its used, and continue to improve upon it. The other wants to use websites with as little intrusion into their personal privacy or finances as possible.
Both sides are entirely justifiable in their goals, but the difficulty comes from how opposed they are. One cannot exist without the other and in some cases, many modern web features require the use of trackers to facilitate much of what we’ve come to rely on as mainstay internet features. How then can we create a web that doesn’t impact the privacy or performance of web users, whilst still giving website owners the necessary tools to run their sites well into the future?
Ghostery’s solution is the “trust” system, that allows users to whitelist sites that they want to support by allowing them to run trackers and adverts.
“The homespun example we give is if your friend has a blog and you want to support it, you can use the trust function in Ghostery to do just that,” Tillman said. If a publisher or website owner does have a very clear sort of value exchange in mind, as long as it makes that easy for the user and the user says yes, then there’s nothing wrong with that.”
Ultimately, Tillman said, Ghostery would like to see more websites be upfront and honest with users about how the content is monetized. After all, if a website is free to access, then the costs of running it must be recouped somewhere, whether through adverts, selling user data, or something a little different like cryptocurrency mining.
“Most of our users […] support publishers finding ways to make money.”
“For the most part, most of our users, more so than the average user, support publishers finding ways to make money,” he said. “The problem is most of them are opposed to the models that are available today. They are opposed to the data collection (especially implied data collection ) practices of website, as well as the forced ad model especially when it’s coupled with poor website performance and really creepy ad targeting technologies.
While a solution to the current tracker climate of web browsing will need cooperation from users and website owners on a grander scale than seems imaginable at the moment, Ghostery does – of course – have recommendations for those wanting to avoid the “tracker tax” in the meantime.
“I recommend they use a browser like Cliqz [developed by Ghostery’s parent company] and a tool like Ghostery and those two things together provide the most complete privacy protection solution,” Tillman said, whilst also highlighting the availability of Ghostery extensions for Chrome, Firefox, and mobile platforms like iOS and Android.
“Fundamentally [people] need to proactive to protect their own privacy,” he said. “Relying on government regulations and self-regulating by companies like Facebook, users will never fully be protected and I think that creates a false sense of security. Using tools to protect your privacy is the best and most direct way to do that.”
