Content uploaded by Thomas Beretich
Author content
All content in this area was uploaded by Thomas Beretich on Oct 11, 2017
Content may be subject to copyright.
Electronic copy available at: https://ssrn.com/abstract=2878750
How Online Tracking and the filter bubble have come to define who we are.
Thomas Beretich
University of Maine School of Law
From its early years as a principally research and education oriented resource, the internet
has evolved into a heavily commercialized and ubiquitous component (see figure 1.) of the
public’s daily routine.
1
In fact, research by the Pew Research Center’s Internet & American Life
Project shows the extent to which internet commerce has become a part of everyday life with a
proportion of the general public that has purchased services such as airline tickets or hotel rooms
online rising from 22% in May 2000 to 52% in May 2010.
2,3
What technologies have facilitated
this transition of the web to a commercially-focused research tool and what has been the
outcome in terms of the user experience and where it is heading? This paper discusses these
points and illustrates how the “web experience” has taken on a different aspect that was probably
not envisaged by its early creators and users. Far from being an efficient research tool to help us
make smarter decisions about what we buy, who we meet or even how we arrive at our moral
and political positions, the tracking technologies that are so instrumental in furthering the
1
This evolution did not occur without controversy. The Scientific and Advanced-Technology Act, , 42
U.S.C. § 1862(g), allowed the National Science Foundation to support access to networks with commercial interests,
thus further clouding what had been clear lines between commercial and non-commercial networks. As a result, the
research and education communities were worried that the internet would be less responsive to their needs.
http://www.gao.gov/new.items/og00033r.pdf . The final walls between the commercial and non-commercial world
on the internet fell in 1995, when the NSF ended its financial support to the NSFNET backbone service and helped
regional research and education networks transition to commercial Internet Service providers. A Brief History of the
Internet, http://www.walthowe.com/navnet/history.html.
2
The author of the paper, Jim Jansen, a research fellow at the Pew Research Center notes that “E-commerce is now
a 360-degree experience for shoppers - it begins with research that in turn leads to purchases that then trigger
commentary and reviews by shoppers. Every part of the online experience seems to have become second nature to
internet veterans.” http://www.pewinternet.org/Press-Releases/2010/Online-Product-Research.aspx
3
From the e-merchants perspective, the internet continues to make steady advancesas as a source of revenues that
cannot be ignored. See Figure 2. http://www.census.gov/retail/mrts/www/data/pdf/ec_current.pdf
Electronic copy available at: https://ssrn.com/abstract=2878750
objectives of the commercial players of the world wide web have arguably restricted our choices
vis-à-vis research and the user experience on the internet. The choices are not necessarily
smarter; they are reduced however, and certainly more efficient from a commercial standpoint;
but by allowing those services such as search engines
4
that are the gateway to internet research,
to continue to track, model and predict our interests, we have delegated to them not just control
of the important and much more debated matter of our privacy but also our ability to choose in
an unrestricted and unfettered manner those areas and people of interest to us and thus in the
context of internet socialization, the definition of our very personality.
4
According to FairSearch which references the Pew Internet & American Life Project research reports, as of May
2011, 92% of adults are using search engines to find information online.http://www.fairsearch.org/wp-
content/uploads/2011/10/Googles-Transformation-from-Gateway-to-Gatekeeper.pdf
Res ponde nts who us e internet.
40%
45%
50%
55%
60%
65%
70%
75%
80%
85%
90%
M a
r-0
0
Jan-0 2
Jun-0 3
F e b - 0 4
J a n - 0 6
Ma r - 0
7
M a
y- 0
8
A p r - 0 9
Jan-1 0
J a n - 1 1
Feb- 1
2
Figure 1.
Percentage (bold line) of respondents (with outer lines as margins
Research Center survey
asking whether they used the internet.
Pages/Trend-Data-(Adults)/Usage-Over-
Time.aspx
Figure 2. Quarterly E-
commerce Sales over time from 2004 to 2013. (3)
Tracking Technologies
and their uses
In Omer Tene and Jule
s Polone
Advancing Transparency and Individual Control in Online Behavioral Advertising
survey of tracking technologies and their evolution over the life of the internet was included.
Cookies, Flash Cookies, Browser fingerprinting, unique identifying numbers on mobile devices,
deep packet inspection, and history sniffing are all described from a technological standpoint,
while juxtaposing the level of transparency in their use by onli
expectations of privacy and control by internet users.
how the internet works today from a commercial standpoint and that is described in Tene and
Polonetsky’s paper, but from the perspe
5
Tene, Omer and Polonetsky, Jules, To Track or 'Do Not Track': Advancing Transparency and Individual Control in
Online Behavioral Advertising (August 3
http://dx.doi.org/10.2139/ssrn.1920505
6
Id., 300.
Percentage (bold line) of respondents (with outer lines as margins
-of-
error) answering positively to a Pew
asking whether they used the internet.
Data from
http://www.pewinternet.org/Static
Time.aspx
commerce Sales over time from 2004 to 2013. (3)
and their uses
s Polone
tsky’s exhaustive paper
To Track or “Do Not Track”:
Advancing Transparency and Individual Control in Online Behavioral Advertising
survey of tracking technologies and their evolution over the life of the internet was included.
Cookies, Flash Cookies, Browser fingerprinting, unique identifying numbers on mobile devices,
deep packet inspection, and history sniffing are all described from a technological standpoint,
while juxtaposing the level of transparency in their use by onli
ne businesses against the
expectations of privacy and control by internet users.
6
The underlying element that is crucial to
how the internet works today from a commercial standpoint and that is described in Tene and
Polonetsky’s paper, but from the perspe
ctive of third party analysis and services, is online
Tene, Omer and Polonetsky, Jules, To Track or 'Do Not Track': Advancing Transparency and Individual Control in
Online Behavioral Advertising (August 3
1, 2011). Available at SSRN: http://ssrn.com/abstract=1920505 or
http://dx.doi.org/10.2139/ssrn.1920505
error) answering positively to a Pew
http://www.pewinternet.org/Static
-
To Track or “Do Not Track”:
Advancing Transparency and Individual Control in Online Behavioral Advertising
,, a thorough
survey of tracking technologies and their evolution over the life of the internet was included.
5
Cookies, Flash Cookies, Browser fingerprinting, unique identifying numbers on mobile devices,
deep packet inspection, and history sniffing are all described from a technological standpoint,
ne businesses against the
The underlying element that is crucial to
how the internet works today from a commercial standpoint and that is described in Tene and
ctive of third party analysis and services, is online
Tene, Omer and Polonetsky, Jules, To Track or 'Do Not Track': Advancing Transparency and Individual Control in
1, 2011). Available at SSRN: http://ssrn.com/abstract=1920505 or
behavioral tracking. The use of online behavioral tracking has exploded and this explosion in
growth is a function of commercial demand as well as the arrival of big data technology which
has enabled the analysis of ever greater stores of data in predicting user behavior.
7
The ever
increasing quality of analysis and predictability has led to more targeted, efficient and effective
advertisement, according to online advertising companies and their executives.
8
Although the use of personalization agents in collecting and analyzing user activity on
the web to deliver tailored content is by no means a new phenomenon,
9
it is the explosion in their
use by large search engines and web advertisers which has created so much concern in terms of
privacy issues.
However, Ad Technology executives seem to make numerous and compelling arguments
for why the use of today’s tracking technology is a good thing.
10
Aaron Bell and Adam Berker from AdRoll feel that marketing becomes relevant and
effective because it is compelling and not intrusive, with the key element being that the user is in
control. This results in a true partnership between the internet user and advertisers which in turn
ensures that the content and information on the web remains free.
7
In her article in the Wall Street Journal, Julia Angwin reports how in less than a two year span, the instance of data
collection from visiting 50 of the most popular websites went from 10 to 56 instances, an increase of 460%. One of
the reasons for this dramatic increase is the advent of real time bidding auctions which allow the winning bidder to
use online behavior to tailor advertisements more expediently and often in immediately.
http://online.wsj.com/article/SB10001424052702303836404577472491637833420.html?mod=rss_whats_news_tec
hnology&utm_source=feedburner&utm_medium=feed&utm_campaign=Feed%3A+wsj%2Fxml%2Frss%2F3_7015
+%28WSJ.com%3A+What%27s+News+Technology%29
8
http://www.adexchanger.com/online-advertising/why-is-tracking-good/
9
Mobasher in this seminal paper describes an approach to web-mining that includes co-occurrence page patterns
within usage data, using path profiles to predict future HTTP requests, and using data mining techniques to extract
usage patterns. These methods are combined within Mobasher’s Usage based Web-Personalization Architecture
which includes 1. the preprocessing of Web usage logs and the grouping of URL references into sets and 2. the
mining of these user transaction sets through the techniques of transaction clustering, usage clustering, and
association rule discovery. Bamshad Mobasher , Robert Cooley , Jaideep Srivastava, Automatic personalization
based on Web usage mining, Communications of the ACM, v.43 n.8, p.142-151, Aug. 2000
10
http://www.adexchanger.com/online-advertising/why-is-tracking-good/
Chad Little of Fetchback (Ebay) speaks in terms of the efficiencies of complete data
access and how cost savings that are ultimately passed on to the consumer. He proffers that
online behavioral tracking is to be embraced and not feared as it ultimately simplifies and
improves the consumer online experience.
Bill Todd of ValueClick Media highlights the real value that online behavioral
advertising brings to the web experience. Relevant advertising keeps the content and services on
the web free. Furthermore, he insists that consumers value the more personalized online ads and
prefer them to the more general and largely irrelevant ads that have been the industry standard in
the past.
David Nelson of Unanimis.co.uk speaks about the importance of content in making the
web so appealing to consumers. If not for this interesting content, there would be no web. If not
for online behavioral advertising,, content (ads) would irrelevant and uninteresting.
Eric Bosco of ChoiceStream feels that privacy is a good thing, but describes online
behavioral advertising as simply the same type of mechanism by which the local shopkeeper
would make a point of knowing what his loyal customers were interested in when entering his
store. In fact, Bosco insists that this degree of localized, friendly customer experience has
disappeared from Main Street USA but is thriving on the internet thanks to the use of online
behavioral tracking.
Maria Schimke of AudienceScience sums up the two major themes that seem to pervade
the logic of online Advertisers when they defend the use of online behavioral tracking. There are
two major consumer benefits: keeping the internet free and delivery of advertisements that are
relevant and add extra value to the consumer’s experience.
But are all these Ad Executives’ comments valid reasons for online behavioral tracking
or simply excuses for technology that has changed the very fabric of the internet. As Alex
Kelleher of CognitiveMatch says, Tracking is the only way for companies to figure out what the
consumer really wants without asking in person. And it is not the individual differences that
concern the ad company but the similarities amongst consumers which allow deducing what to
offer and ultimately how to get a better return on the ad companies investment.
If we look at tracking technology from the viewpoint of its use in online behavioral
advertising, the debate seems to be purely about a struggle between consumers and businesses
over control of advertising content and frequency. However, the uses of tracking technology
have now reached areas that touch more closely fundamental rights of privacy and freedom of
choice in associating with ideas and groups of people. A report by ENISA
11
, the European
Network and Information Security Agency, highlights how tracking technology use has created
the following areas of concern:
1. A diversionary debate – some, maybe most, advertising companies focus on Do
Not Track as an opt-out question for behavioral advertising, but ENISA maintains
that Tracking is the actual problem, not behavioral advertising,
11
http://www.infosecurity-magazine.com/view/29376/enisa-and-the-privacy-considerations-of-online-behavioral-
tracking/
Figure 3. Spying software manufacturer Finfisher is becoming popular with governments who feel the need to spy on their own
and other citizens. (13, 15)
2. User profiling for other purposes than behavioral advertising – user profiling now
goes beyond just behavioral advertising and now includes global surveillance
12
which can be performed by governmental entities to track its citizens
13
and those
of other countries
14,15,16
, but it also includes surveillance by companies for
commercial reasons.
17
12
Of course the recent well publicized endeavors of the National Security Agency and its spying program PRISM are
such an example where a government has been able to obtain search history,content of emails, file transfers and live
chats directly from the servers of the most important U.S. Service Providers.
http://www.guardian.co.uk/world/2013/jun/06/us-tech-giants-nsa-data
13
Nigeria is one of many countries who plans apparently to ramp up surveillance of its own citizens, with the
purchase of equipment and software that will allow it to “conduct surveillance on an unprecedented scale” by using
products such as Finfisher which can “obtain passwords from your computer, monitor Skype calls, and even turn on
your computer’s camera and sound recording so as to watch you at work.”
http://advocacy.globalvoicesonline.org/2013/07/12/nigerian-government-to-ramp-up-internet-surveillance/
14
The United States has not cornered the market on extra-territorial surveillance of foreign citizens. Broad networks
of “infiltrated computer systems including foreign ministries, news media, NGOs and political dissidents” point to
IP addresses based in China that are receiving the purloined information. Even supposedly backwater regimes such
as Gadhafi’s were spying on dissidents based in the United Kingdom, as well as, foreign activists and political
adversaries amongst others. https://www.eff.org/deeplinks/2013/06/spies-without-borders-i-using-domestic-
networks-spy-world
3. The wrong data – user profiling that contains incorrect data on a user creates
potential harms such as “error, abuse, lack of transparency and accountability.”
In fact, incorrect tracking data in the hands of government could “prove
terminally wrong.”
4. Reality/Physical Mining – the automatic and continuous gathering of information
on location and users of smartphones raises many questions about privacy and
leads to the image of a surveillance society.
5. Augmented reality – using inference techniques from different websites to
identify what is presumed to be anonymous information. An example would be
the use of facebook photos to identify anonymous photos on a dating website.
6. Service and price discrimination – most notable in the realm of insurance where a
profile might reveal risk of a certain disease thereby resulting in the denial of
insurance coverage.
7. Personalization – as to be discussed further in this paper, this is where users get
trapped in a filter bubble. As described by ENISA, as well as George Orwell, in
authoritarian states, this filter bubble is the manifestation of the states aims at
15
Angola, Bahrein and Germany are amongst those nations that are implicated in the use of extra-territorial
espionage with the help of commercially available software such as Finfisher. (see footnote 13 also).
http://www.edri.org/edrigram/number11.2/germany-finfisher-spyware
16
http://www.businessinsider.com/countries-with-finfisher-spying-software-2013-5
17
This of course is one core aspect of the debate - what is surveillance? Countries may have fewer problems debating
the intricacies of what constitutes Personally Identifiable Information when there whole purpose is to monitor and
gather information on an individual, while companies (who cannot legally spy on an individual unless told to do so
by the jurisdictional powers to which they find themselves subject), can arguably gather as much information as
they deem necessary for marketing purposes, as long as they follow governmental guidelines, laws and industry
practices which purportedly protect the privacy of the individual. Of course many times one doesn’t know what PII
one has until it is observed. Much like the Schrödinger's cat paradox, the entanglement of unobserved PII in
customer data can help associate concretely the interaction between seemingly separate customer data, thereby
helping to cluster disparate data sources and then after its removal, remnants of its presence still remain as part the
unique user profile created. http://online.wsj.com/article/SB10001424052748703940904575395073512989404.html
censorship by the careful selection of news and information that can be showed to
specific users.
The Filter Bubble
The ”filter bubble” is the term invented by internet activist Eli Pariser in his book “The
Filter Bubble: What the internet is Hiding from You”.
18
It is described as the result state in
which internet users are separated from information that disagrees with their viewpoints, thus
isolating them in a cultural or ideological bubble.
19,20
This tailoring of information to the internet
user is a result of advances in tracking technology and data mining algorithms. Specifically, the
website algorithms use information about the user such as location, click behavior, search history
and any other specific information (gender, education level, sexuality, etc.) about the user either
explicitly or implicitly gathered and then chooses information to show to the user based on the
algorithm’s predictions.
The players:
Google – considered the pioneer in personalization with its personalized search results.
Google’s personalized search results were based on a logged in users search history. As opposed
to earlier web search algorithms which looked at the relevancy of the search terms to web pages
and thus a larger universe of possibilities, this focused more on the user’s past activity in terms
of surfing history and made inferences based on this information.
21
As of December 4, 2009, the
Personalized Search was introduced to all users of Google, even those not logged in to a Google
Account.
22
Concerns brought up regarding this development in search technology included
18
Eli Pariser, The Filter Bubble: What the Internet Is Hiding from You, Penguin Press (New York, May 2011) ISBN
978-1-59420-300-8
19
http://en.wikipedia.org/wiki/Filter_bubble
20
http://www.searchenginejournal.com/the-google-filter-bubble-and-its-problems/29879/
21
http://news.cnet.com/Google-automates-personalized-search/2100-1032_3-5766899.html
22
http://googleblog.blogspot.com/2009/12/personalized-search-for-everyone.html
decreasing the likelihood of finding new information by bias based on previously found search
results, privacy problems for those users unaware of the personalization aspect of their search,
and a disruptive effect to the search engine optimization industry which relied standardized
rankings across users and not individualized ones.
23
Facebook – touted by some as the feature that allows the wall between the social network
and the world to be broken
24
, Facebook’s instant personalization feature was launched in
February of 2011and (by default) allowed sites such as Pandora or Bing to personalize their sites
based on data gathered from your account. By allowing third party apps and websites to gain
access to Facebook user’s personal data, Facebook in turn will gain access (own) the data that
results from this interaction and further its strategy of Data Dominance
25
. Facebook from a
commercial perspective of course, has no choice to do otherwise since it is in direct competition
with Google and must develop ways to keep its users engaged with Facebook even when they are
not on the site (through for example an external ad network that uses the data assets acquired by
Facebook Exchange – a realtime bidding ad system
26
).
Yahoo – In early 2012, Yahoo has begun a push towards the increased emphasis on
personalization of content for the user. For new users, the news feed starts off in a generic
manner, but as more stories are clicked on, then the feed will begin to adjust based on the user
choice of articles and stories.
27
There is some active and conscious user input however, in that
hovering over the right side of each story and clicking an X will de-emphasize that specific topic
in the news feeds whatever that may be. In a further nod to the importance of integration with
23
http://www.networkworld.com/news/2009/120709-google-personalized-results-could-be.html
24
http://www.zdnet.com/blog/igeneration/facebook-instant-personalization-how-to-disable-it-and-why/8006
25
http://www.geekwire.com/2012/facebook-personalization-engine-web/
26
http://techcrunch.com/2012/06/13/facebook-exchange/
27
http://techcrunch.com/2013/02/20/yahoo-personalized-front-page/
other social media sites, Yahoo will allow users to interact with sites such as Flickr, Facebook
and Twitter and presumably exchange customer data about these social transactions.
Microsoft – like Facebook, Microsoft has a common enemy named Google, and the
personalization strategy created for its search engine Bing serves as a counterweight to Google’s
dominance on the web. Bing now includes search results with Facebook activity by a user’s
friends related to the search term.
28
And as Bing’s web algorithm thinks the social results are
relevant they show up (within the Facebook module) higher in the page. Interestingly, those
users who feel uncomfortable with the Facebook instant personalization and Bing combination
technology, will be able to opt out via a pop-up message within the first five times they access
the feature.
It is apparent from the four examples above, that competition is driving the use of
personalization online, as much as consumer demand. In fact, 66% of business respondents
surveyed in one study answered that both an improved customer experience and improved
business performance are the main drivers of personalized web services.
29
However, even
despite the level of personalization attained by the major players, it is still considered in its early
days
30
and thus it is difficult to see how far personalization will be allowed to extend into the
privacy of internet users and the control of their personality
The Inviolate Personality and the Right to Privacy
Within the philosophy of self, the personality can be described as those qualities in a
person that describes him or her uniquely as compared to others. Personality at least from a
modern understanding derives its roots from the Renaissance period, where introspection and the
concept self-development of the individual as opposed to the membership of a class defined the
28
http://gigaom.com/2010/10/13/bing-launches-facebook-instant-personalization/
29
http://econsultancy.com/us/blog/62583-94-of-businesses-say-personalisation-is-critical-to-their-success
30
http://econsultancy.com/us/reports/the-realities-of-online-personalisation-report
self.
31
This interpretation of self is in stark contrast to Stephen Greenblatt’s more Medieval
concept of defining self in the framework of the ”household, the kinship network, the guild and
the corporation” as described in his book on the history and recovery of Lucretius’ poem “De
Rerum Natura”, - The Swerve: how the world became modern.
In Warren and Brandeis’ 1890 article “The Right to Privacy”, the 20
th
century
foundations were laid to answer the question of what the Right to Privacy meant in the modern
era.
32
The publication of the paper, while clearly a response to what was the invasiveness of the
then recent technological innovation of photography and the rise of “yellow journalism” served
to define Privacy more as the “right to be let alone” with the focus on protecting the individual.
Later in Olmstead v. United States, Brandeis further developed the right to privacy from a
constitutional perspective, thereby implicating the government as a potential invader of the
individual’s privacy.
33
Prosser in his work, Privacy
34
, wrote that under the mantra of the “right
to be left alone”, there were four separate torts:
1. “Intrusion upon the plaintiff’s seclusion or solitude, or into his private affairs.”
2. “Public disclosure of embarrassing private facts about the plaintiff.”
3. “Publicity which places the plaintiff in a false light in the public eye.”
4. Appropriation, for the defendant’s advantage, of the plaintiff’s name or likeness.”
Finally Prosser, in his article “The Right of Privacy”
35
(and more recently in his
interviews)
36
, approaches privacy rights from almost a purely economic standpoint. He argues
that:
31
http://www.arcaneknowledge.org/histschol/renaissance.htm
32
Warren and Brandeis, "The Right To Privacy", 4 Harvard Law Review 193 (1890).
33
Olmstead v. United States, 277 U.S. 438, 1928.
34
Dean Prosser, 'Privacy' (1960), 48 California Law Review 383.
35
Posner, Richard A., "The Right of Privacy" (1978). Sibley Lectures. Paper 22.
36
http://bigthink.com/videos/judge-richard-posner-privacy
1. Privacy and curiosity are considered to be intermediate goods which have
instrumental as opposed to ultimate value.
2. The demand for private information is understandable where possible
relationships create opportunities for the demander such as the tax collector,
fiance’, partner, etc. It is also understandable in terms of casual prying into the
private lives and is motivated by self-interest of the person who is prying.
3. The supplier of private information also may make use of his or her private
information through misrepresentation, thus manipulating other people’s opinion
of them. Thus the wish for privacy is about controlling “others’ perceptions and
beliefs vis-à-vis the self-concealing person” or put more succinctly to “control the
flow of information about” one’s self.
4. As disclosure is resisted by those who wish to maintain their own privacy and is
valuable to others, this would seem to imply that property rights exist in people’s
own information about themselves. However, transaction costs “militate” against
assigning property rights to the possessor of a secret. What’s more, information
acquired by government entities have protections built in to assure against
disclosure.
5. In terms of discreditable information, the possessor can use this information to
mislead those with whom there is a transaction such as an employer, an insurer or
a potential spouse.
6. It is wrong on economic grounds for an individual to conceal material facts about
themselves. Although a duty of full and frank disclosure would be burdensome,
everyone should be allowed to protect himself from disadvantageous transactions.
In summary, Posner’s economic efficiency approach to the Right of Privacy is that 1)
trade and business secrets are protected to allow one to exploit superior knowledge or skills, 2)
no protection for facts about people,- health, temper, income but some ability to prevent their
discover if not unduly intrusive and 3) limitations of eavesdropping or other forms of
surveillance to only illegal activities.
The Filter Bubble and the Right to Privacy
It should be clear from Warren and Brandeis’ essay all the way to Posner’s Lecture that
Privacy Rights offer little in the way of protecting users against the Filter Bubble. Online service
providers and advertisers have integrated personalization technologies into the user experience to
the point where, we have lost the ability to control the information that is fed to us. The
Renaissance model of personality emphasizes the individual’s ability to “self-determine”
through the continuous and unfettered freedom of choice of association. The medieval model
instead associates self with class or more specifically with the ”household, the kinship network,
the guild and the corporation”. One needs not delve to long in thinking to understand that this
recent development in the personalization of the web-experience is much more medieval in
nature.
Solutions and Recommendations
ENISA makes several recommendations in its paper “Privacy Considerations of online
behavioural tracking:
1. Refocus on tracking not on OBA – the debate today seems almost diversionary as
if Online Behavioural Advertising was the bogeyman of internet privacy issues.
37
However, “creepiness” is not a harm against which the constitution protects us.
37
http://venturebeat.com/2012/02/22/the-7-creep-factors-of-online-behavioral-advertising/
Moreover, it has been hard for plaintiffs to make claims of harm due to OBA
stick. Given the more constitutionally protected rights being infringed by
government surveillance, this refocusing on tracking itself, may allow the legal
system to find other solutions to protect internet users.
2. More meaningful privacy policies – the notice and choice aspect of privacy
standards is seriously limited according to ENISA and more must be done to
overcome their limitations. However, given weak enforcement, a lack of
substantial restrictions, lack of real notice and the lack of real choice
38
, this
recommendation seems far from practical.
3. Easier to use tools for transparency and control – TETs (Transparency Enhancing
Technologies) are suggested to help improve user awareness about their being
tracked and profiled while engaging in web activity. This still however requires a
level of knowledge that is beyond the average user and probably impractical from
an implementation standpoint. At what level does the usability of these tools
have to reach to ensure that everyone is informed of their rights?
4. Compliance and monitoring initiatives – ENISA suggests Privacy impact
assessments and possibly privacy certification as important initiatives to show
how a service or application might affect a user’s privacy. More importantly, in a
shift away from the traditional notice and choice to what may become an
accountability standard, ENISA suggests that the monitoring and detecting of
violations and the enforcement of rules become important tools in protecting
online users from illegal privacy practices.
38
http://digitalads.org/documents/Schwartz_Solove_Notice_Choice_NPLAN_BMSG_memo.pdf
5. Anti-tracking initiatives for Mobile Apps – since current browser technology has
not adapted to mobile platforms, most tracking is occurring through mobile
applications which offer no DNT choice except for uninstalling of the application.
6. Privacy-by-design - although unsuccessful till now, regulation would have an
important role in pushing the burden of enforcing online privacy to businesses.
Until they are obligated through laws or regulation, it is unlikely that companies
will integrate privacy into their products or processes when they can simply
disclaim liability through privacy policies that are largely ignored by the
consumer.
As described from the ENISA study on online behavioural advertising, there are several
solutions to privacy issue of personalized tracking. Some solutions if endorsed by governments
would probably be successful, however, given that governments have come to know the power
of tracking in furthering their goals of citizen and extra-territorial surveillance, it is possible that
these new regulations and laws may never come.
What can work and has worked in the past is for the legal system to understand the nature
of the constitutional infringements that occur when user personalization stifles the free flow of
information and choice in association. If newspapers were to begin printing individualized
papers based on the subscribers religious, ethnic or sexual orientation, there would be a
resounding howl of protest from the public and legal community based on First Amendment
rights here in the United States.
39
And more explicitly from the perspective of the Universal
Declaration of Human Rights, Artice 19, “everyone has the right to freedom of opinion and
39
http://www.archives.gov/exhibits/charters/constitution_zoom_1.html
expression; this right includes freedom to hold opinions without interference and to seek, receive
and impart information and ideas through any media and regardless of frontiers.
40
40
http://www.un.org/en/documents/udhr/
