on tracking users

Audrey Watters has opted out of tracking people on her websites. It’s a good read. I agree 100%.

I’ve felt creeped out by the pervasive tracking networks online - analytics, ad networks, cookies, super-cookies, browser fingerprinting, etc…. This surveillance ecosystem is the end result of an arms race to find out about people reading web pages online. There are a few reasons, but my gut says it boils down to 2:

1. Monetization
2. Ego stroking

Monetization - to sell ads, in whatever flavour, requires metrics. How many people view a page? How many see an ad? How many click on it? How many then buy a product? How many return to the site? etc… So much data. I don’t sell anything, and don’t have ads¹ and I don’t sell product placement guest posts (never have, never will). So, this isn’t a reason to track people on my site.

Ego stroking - this is actually a good reason to track readers. Is it worth the privacy violation? I don’t think it’s worth feeding Google’s surveillance machine, so I used a self-hosted copy of Piwik for a year. But, after thinking about it after reading Audrey’s post, I’ve stopped that, too². My ego is just fine. I don’t need to be propped up. I do this for me. If people read it, hey, that’s great. If nobody (else) reads it, hey, that’s OK too. So, this isn’t a reason to track people on my site.

I just used the excellent Ghostery plugin for Firefox, to report web trackers. My site reports clean, with two exceptions: Gravatar and Google Analytics. Gravatar is from the WordPress comment system. It’s innocuous. Wait. Google Analytics? I don’t even USE Google Analytics! It’s there because I embedded a YouTube or Vimeo video. Which brings along all kinds of snooping trackers as part of the deal. Awesome. Once that post falls off the front page, Google Analytics will drop off (except on the pages that display that post and others with embedded media from YouTube etc…)

For work projects, though, it’s not as simple. There are variations of Reason #1 that are needed - although not monetization in the pure sense, I need to be able to answer the question “does anyone use the website or resource? Is it worth supporting it?” - there are 2 ways to answer that. The first is with web analytics. The second is with testimonials from users. I need to be able to provide both.

But, I won’t feed Google’s surveillance machine in order to meet my needs. So, I host a copy of Piwik on campus, and use it to track aggregated and anonymized web analytics. I’ve set it to not store full IP addresses - I have no idea which on-campus computers are accessing our stuff. I have no idea about any individual users. But, I can show traffic patterns and spikes, and that is important information when we’re planning support - I know exactly when spikes occur during semesters, and I know exactly when we need to develop support resources and have them online and available before the traffic spikes.

See Also

• Pearson and surveillance of students
• Audrey Watters on the nature of educational technology
• UGuelph and D2L sitting in a tree
• on a world with only 10 universities
• the web is what we make it