Wikinews interviews DuckDuckGo, Opera, Mozilla, Wikimedia about DoNotTrack feature

February 5, 2013

Following the introduction of a "" feature in modern browsers at the end of last year, Wikinews interviewed several companies and groups about the feature.

DuckDuckGo
A crowdsourced  reviewed the feature and launched a whatisDNT microsite in December. The review involved checking answers to basic questions such as whether websites stop shaping a based on users' online actvitity or stop displaying. Wikinews interviewed DuckDuckGo founder Gabriel Weinberg about the microsite launch and the DuckDuckGo opinion on the feature.

What prompted the launch of DuckDuckGo (I think it was around 2006?)?


 * Gabriel Weinberg: We launched on Sep 25, 2008, though I had been working on it for about a year prior. The initial motivation was to try a search engine with a different UI that did a better job of using more structured content (like from Wikipedia) and also more aggressively removed spam.

What does the DuckDuckGo Team use instead of Gmail?


 * GW: We do not have company mail accounts (beyond forwarding ones), so everyone uses what they want. I personally use outlook.com right now.

When and how did DuckDuckGo start using Perl? What influenced this decision and language choice?


 * GW: We started out in Perl. I picked it up at MIT where it was prevalent in the late 90s, and pretty much never looked back. It worked really well for a project like this that is text heavy and can use a lot of existing helper libraries.

How did you first hear about "DoNotTrack" feature?


 * GW: I do not remember, but it was a long time ago :). The Do Not Track concept was proposed many years ago.

When did you launch http://whatisDNT.com/ ? Who worked on it? What researches, tests and studies did it involve?


 * GW: We launched the What Is Do Not Track micro-site on Dec 19, 2012, and various members of the DuckDuckGo Team worked on it (including myself). We had been following it closely for a while, and so have been up to date on everything that is going on with it. It honestly doesn't take any tests to prove its ineffectiveness since companies like Google tell you straight up they don't honor it.

A search engine that does not track users sounds good. What do you recommend users to do to not be tracked by *other* websites, such as blogs with targeted ads in sidebar?


 * GW: It depends on the Web browser you are using for specifics, but for each major Web browser there are tools you can install (besides DuckDuckGo extensions) to protect you in various scenarios. Some of those that are available in multiple browsers are Ghostery, DoNotTrackMe, and HTTPS Everywhere. Those help protect you while not really changing your browsing experience. Other tools do more, but do impact your browsing experience, like Tor and NoScript.

What is your opinion of ad block tools such as AdBlockPlus and NoScript that ideally aim to block third-party js?


 * GW: I think they are effective at doing what they set out to do, and that is great if you are a consumer who wants that experience. However, they are not for everyone because they do degrade browsing ability, especially NoScript.

Do you think that DoNotTarget option should exist (if consistendly adhered to)?


 * GW: Absolutely. My main problem with the DoNotTrack setting right now is it is misleading. If you use a major Web browser, you have this Do Not Track setting within it, but it really does next to nothing. It is a false sense of security.

What do you see the ideal behaviour of companies when it comes to making a user's profile? Would you like to ideally prohibit such activity entirely using legislative regulations in the long run?


 * GW: Yes, I think an analogy to the Do Not Call list is apt. Applying that analogy, if you were on a Do Not Track list then companies shouldn't track you unless you explicitly ask them to, e.g. by logging in and consenting to Terms of Service. That will only happen via legislative means, however.

Opera
Wikinews also interviewed Haakon Flage Bratsberg, Product Counsel from, the corporation behind Opera web browser.

Have you heard of "Do Not Track" feature support in modern browsers? What do you think about it?


 * Haakon Flage Bratsberg: Opera browser for desktop computers has built in support for Do Not Track since version 12, and Opera Software participates in the W3C Tracking Protection Working Group.

Given that not all companies respect this setting, do you feel its name misleading? Do you think it could better be named "Do Not Target" instead?
 * HFB: This is our major concern about the DNT functionality: Users can be given a false sense of security.

Do you think that all websites should respect DNT users' preferences in the long run?


 * HFB: In general we think websites should comply with web standards, including the current proposal for DNT.

Would you ideally see the respect for DNT users' preferences legally enforced?


 * HFB: There are obvious limitations to a pure self regulatory approach, but our hope is that self regulation will provide a sufficient good outcome in this case.

What do you see as an ideal solution to resolve the misleading situation with DoNotTrack that you acknowledged in your response to the second question?


 * HFB: First of all, I personally do not think the name of DoNotTrack is misleading. However, the challenge is to have a meaningful standard that users can still be aware of the risks of that may be involved without giving a false sense of security. For example "private mode" in your browser does not prevent that your internet access provider can still keep track of what websites you visit, websites can collect information about you and so on. It only removes the trace of the sites you visited in your "browser". Similarly, the websites can by accident or intention to be set up in a way so they do not comply with the DNT signal. There is always a risk of rouge agents.

How do you envision the "self regulatory" approach? How would it work?


 * HFB: Self regulation basically means that the different stakeholders without having a legal obligation to do so, comply with a set of rules, for example an industry technical specification as W3C tracking protection standards. My take is that DNT would be a success if all major stakeholders involved in the process would comply with the standards. It is more an open question to what extent compliance to the technical standard is sufficient to comply with applicable laws in its respective country, for instance, in EU, or in the USA.

Mozilla
Wikinews contacted the Mozilla press office and got replies on behalf of the.

When and why did you decide to start supporting the Do Not Track feature?


 * Mozilla: Mozilla introduced Do Not Track in the desktop Firefox browser in January 2011, and in June 2011 Firefox for Android became the first mobile browser to support Do Not Track.


 * M: We support Do Not Track because we believe it is crucial to put users in control of their online experience. Do Not Track is intended to give users choice and control in a persistent, accessible way without preventing the customization and valuable advertising that powers the Web economy.  We are seeking ways to give users better insight and control into the ways their personal information is collected, used, stored and shared.

Do you think that all websites should respect DNT users' preferences in the long run?


 * M: Mozilla is actively working with companies that have started to implement Do Not Track and with others who have committed to doing so. Ultimately, the goal is for all industry stakeholders to trust and respect the signal in the long run.

The mobile Firefox attitude to DNT is ...interesting! Thank you for the replies.

Wikimedia
Wikinews contacted the press office and got replies from Jay Walsh, senior director of communications at the Wikimedia Foundation.

Have you heard of "Do Not Track" feature support in modern browsers? What do you think about it?
 * Jay Walsh: We're familiar with this feature of browsers. This provides the user with the option to remove user tracking, including cookies. It disrupts a lot of the technology used by digital ad services to follow users across sites etc. For the most part any users of the Foundation's projects would have no issues using the projects if they had Do Not Track enabled on their browser. The only issue they might have is in dismissing message banners on the projects (such as the fundraising banner) may not be completely predictable - banners may reappear after dismissing.

Given that not all companies respect this setting, do you feel its name misleading? Do you think it could better be named "Do Not Target" instead?


 * JW: I'm not sure if you mean browser-making companies or web companies. Many users who activate this service are aware that by disabling user tracking they are effectively preventing ad or content targeting. We're not a browser software organization, so it's not a topic we spend a lot of time considering. However I think it's safe to say that our community (who are very active, opensource and free knowledge advocates) would like the maximum amount of transparency and clarity in these kinds of features for users.

Do you think that all websites should respect DNT users' preferences in the long run?


 * JW: That's not really a question we're in the best position to answer. Obviously we take this kind of thing pretty seriously. From a software development perspective we would never want to confuse or frustrate a user by side-stepping a feature like do not track. We'd like all websites to honor a web user's preferences and desire for privacy of course. Obviously we're all looking at an Internet that relies increasingly on tracking technology to support commercial goals.

Would you ideally see the respect for DNT users' preferences legally enforced?


 * JW: That would be a question for our legal counsel. But I'm sure it's a topic we'd want to research in more detail before reflecting on, and given that we're a non profit and we have pretty limited resources it's not always possible for us to have legal opinions on all matters.