NEWS4 May 2011

Do-not-track isn't just about advertising, say web users

Data analytics North America

US— Researchers might be looking for an exemption from do-not-track proposals, but for a large number of consumers do-not-track might mean no data collection at all, according to research presented at a recent workshop on online privacy.

Aleecia McDonald of Carnegie Mellon University gave a taste of some results from an ongoing research project at the Worldwide Web Consortium (W3C) event last week, showing a gap between what companies think web users want and what users say they actually want.

Although do-not-track has largely been discussed in the context of online behavioural advertising, McDonald saw a wide range of responses when she asked survey-takers to say what they thought would happen if they clicked on a hypothetical do-not-track button.

Thirty-two percent thought it would block history tracking, 27% said cookies, 5% mentioned IP addresses and only 3% mentioned advertising or data mining. Almost 40% of respondents said they would expect no data at all to be collected after clicking the do-not-track button.

But research industry bodies have argued that some data collection will still be necessary, and in a position paper prepared for the W3C workshop they called on do-not-track mechanisms to be limited in scope to advertising tracking and tracking where there is criminal or malicious intent.

Diane Bowers, president of the Council of American Survey Research Organisations, wrote: “Do-not-track tools that block researchers’ ability to access internet users and measure their online behaviours could degrade the quality of the statistical information and insights that we provide and on which decision-makers in the private, public and not-for-profit sectors depend to better understand consumers, customers and citizens.”

In her presentation McDonald suggested that companies and regulators have the option of either building do-not-track mechanisms to match user expectations, or to create a solid definition of what do-not-track means and communicate it clearly and explain where there are exceptions.

The latter approach would be similar to what preceded the introduction of anti-telemarketing ‘do not call’ legislation in the US and Canada, which was accompanied by research industry efforts to explain to consumers that research activities were exempt from the rules restricting marketing calls.