- From: Jules Polonetsky <julespol@futureofprivacy.org>
- Date: Sat, 22 Oct 2011 21:10:06 -0400
- To: "'Jonathan Mayer'" <jmayer@stanford.edu>, <public-tracking@w3.org>
- Message-ID: <008e01cc9120$808f2eb0$81ad8c10$@org>
Sorry to jump in late here. My take is that regardless of the debate about "targeting" or behavioral ads, one of the areas that is central to understand in order to assess the impact of not tracking is "measurability". Banner ads are a limited medium compared to the richness of some other media, but what makes them attractive to advertisers is that they can be measured. An advertiser using a third party ad server can deliver ads across many hundreds of publishers and by setting and recording cookies can learn how many unique users saw the ad, how many users who saw which ad on which site later visited their site, how many users exposed to the ad ended up purchasing or registering. And if they don't directly drive sales, they can survey users who have been exposed to ads across sites they can assess the branding value of the advertising. All this is dependent on tracking across a large range of web publishers. Most reporting by ad servers in this manner results in aggregate reports for each advertiser, but is based on the ad server's cookie level log files. In the early days of online advertising, there were publishers that served their own ads, and refused to allow third party ad serving/tracking. They were worried about the lag time of third party ad servers, the privacy of their users, or other companies having insight into user activity, or preferred to be able to report themselves on the number of unique users that saw the ad. Information across sites was apples and oranges, with advertisers getting widely varying results for the same campaigns. Large advertisers didn't want to have to do business with thousands of small sites and didn't want to have to ship creative to all those sites and get back reports that didn't fit together. Spending was consolidated on Alta Vista, Lycos, AOL, Yahoo and others. Strong pressure from advertisers eventually forced all publishers to accept third party ad serving and tracking by advertisers. At AOL, we had market power to hold out longer than others, but eventually needed to go along. Today, Facebook is the only site of note that is able to demand that it serve ads itself..third party ad servers cant track ads delivered with cookies (unless a user clicks, of course). Just some thoughts. From: public-tracking-request@w3.org [mailto:public-tracking-request@w3.org] On Behalf Of Jonathan Mayer Sent: Thursday, October 20, 2011 5:01 PM To: public-tracking@w3.org Group WG Subject: Web Tracking: Facts on the Ground I completely agree that a shared understanding of web tracking practices is important for the working group's progress. And an open dialogue about the facts on the ground will be invaluable for all stakeholders on these issues. To that end, I've moved this discussion into its own thread. I look forward to a productive discourse. I offer my take on ad blocking responses, the market impact of behavioral advertising, and company rights below. 1. "Some companies already block users from their sites who use ad blocking technologies because it fundamentally impairs their ability to monetize their content." I was agreeing with JC's comment that content owners already charge users for access to some content and services. This statement is not in question as we are all aware of content sites like the Wall Street Journal that charge for access or pay-for premium services like those offered by Linked In. I was also expanding this line of reasoning by pointing out that some sites have blocked users who use ad blocking technologies. Here is one example - http://arstechnica.com/business/news/2010/03/why-ad-blocking-is-devastating- to-the-sites-you-love.ars, but others have done so in a less public way. I am not aware of any popular website that is currently prohibiting or degrading service for ad-block users. To my knowledge Ars Technica was the only major website to experiment with such practices, and it quickly reversed course after facing a mob of angry users. My colleague Arvind Narayanan wrote a year ago about the lessons we might infer about responses to Do Not Track from responses to ad blocking (http://33bits.org/2010/09/20/do-not-track-explained/, pasted below). I agree with his views. . . . from the site's perspective, ad blocking would result in a far greater decline in revenue than merely preventing behavioral ads. We should therefore expect that DNT will be at least as well tolerated by websites as ad blocking. This is encouraging, since there are very few mainstream sites today that refuse to serve content to visitors with ad blocking enabled. Ad blocking is quite popular (indeed, the most popular extensions for both Firefox and Chrome are ad blockers). A few sites have experimented with tiering for ad-blocking users, but soon after rescinded due to user backlash. Public perception is a another factor that is likely to skew things even further in favor of DNT being well-tolerated: access to content in exchange for watching ads sounds like a much more palatable bargain than access in exchange for giving up privacy. 2. "If we use a broad definition of tracking, similar to the FTC's definition, then we will be potentially impacting over 80% of the online ad market." The IAB conducted a survey of the buy side of the online advertising industry (we canvassed the major ad agencies) and found that when we used the FTC's definition of OBA, over 80% of all advertising campaigns would fall under this definition. We have published this survey in a number of venues over the years - http://www.iab.net/media/file/IAB_Comments_to_Congressman_Boucher%5C%27s_Dra ft_Privacy_Legislation.pdf (page4) - and as recently as last week when the DAA testified before the U.S. House of Representatives. This 80% figure has been repeated by a number of advertising industry representatives. I have not been able to locate the survey (Mike, I'd greatly appreciate a pointer). Here are some of the descriptions I've seen. -"However, an informal survey of agencies by the IAB suggests behavioral advertising is extremely widespread: up to 80% or more of campaigns conducted in 2009 involved some form of cookies or other tracking that could be so characterized. [Footnote: IAB interviews of advertising agency personnel regarding use of digital advertising targeting technologies.]" [http://www.iab.net/media/file/News_Media_Workshop_-_Comment_Project_No.P091 200.pdf] -"In an IAB survey of ad agencies conducted earlier this year, we found that 80% or more of digital advertising campaigns were touched by behavioral targeting in some way." [http://www.iab.net/public_policy/1296039] -"According to an informal survey by the Interactive Advertising Bureau (IAB), more than 80 percent of advertising campaigns in 2009 involved tracking of some sort." [http://www.cnn.com/2010/TECH/web/11/04/behavior.tracking.mashable/index.htm l] I believe the survey was intended to show that the vast majority of online advertising campaigns included at least one ad buy that set at least one cookie. That certainly is plausible to me. But, of course, that's far distant from behavioral targeting accounting for 80% of online advertising revenue, or Do Not Track negatively impacting 80% of online advertising revenue. I wrote a blog post in January detailing why I believe Do Not Track would not have a significant negative impact on publishers' online advertising revenues (http://cyberlaw.stanford.edu/node/6592). I would encourage interested readers to take a look. A few key points: a small minority (about 5%) of ad spend is behaviorally targeted, there are cross elasticities in online advertising demand, and there are good (admittedly not perfect) technology fixes available. 3. "Publishers and content owners have every right, in fact have fundamental rights, to offer their goods and services as they see fit." Content is copyrightable. In the U.S. this right was so fundamental (sometimes called national rights) that it was written into our Constitution: Article I, Section 8, Clause 8 of the United States Constitution, known as the Copyright Clause, empowers the United States Congress: " To promote the Progress of Science and useful Arts, by securing for limited Times to Authors and Inventors the exclusive Right to their respective Writings and Discoveries. Generally speaking, this right extends for the life of the author plus an additional 70 years. To own a copyright means that the author owns a bundle of exclusive rights (see Title 17, Chapter 1, Section 106 of the U.S. Code) - http://www.law.cornell.edu/uscode/html/uscode17/usc_sec_17_00000106----000-. html. One of those rights is the right to decide how your work is distributed, which includes the right to exclude others from seeing or using your works, i.e. to offer access to your work under your own terms, including charging a fee or requiring viewers to allow advertising. This is what I mean by fundamental rights of publishers and content owners. I could discuss contractual rights and the fact that most websites have Terms of Service that reserve the right to offer their content and services to consumers on the site's own terms, but that is not necessary a fundamental right. We could also have a discussion here about patent rights since many sites rely heavily on patents, but again, I think the copyright issue best illustrates the fundamental rights of publishers and content owners. To be sure, federal and state laws provide a number of legal rights to businesses. But they also impose a wide array of regulatory restrictions. I certainly do not share the view that copyright is a "fundamental right." The Constitution's Copyright Clause empowers Congress to pass copyright protections; it does not mandate such legislation. The same holds true for patents. As for website Terms of Service, they would be governed in almost all cases by state contract law - which regularly declines to enforce provisions on public policy grounds. For our purposes, the ultimate legal question is whether there would be any constitutional infirmities in requiring the same service for Do Not Track users. I do not see any.
Received on Sunday, 23 October 2011 01:10:36 UTC