Electronic Profiling

Dave Halstead, Helen Ashman, Computer Science and Information Technology, University of Nottingham, Nottingham NG8 1BB, U.K., hla@cs.nott.ac.uk

Abstract

In this paper, we review the various practices relating to the collection, dissemination and use of personal data. Information about people is becoming increasingly useful to marketing agents, and vast digital collections of personal information are being collected. In some cases, the data is sold, but in others, it is freely available. Now that the Web has become the ultimate data dissemination tool, databases of personal information can be queried very quickly, easily, and above all, anonymously. Never has the task of an investigator (or a casual interrogator) been easier.

It also is clear that data collection is increasingly a core activity of private companies, rather than government organisations. Orwell overshadowed a generation of readers with the "Big Brother" of 1984, and the spectre of governmental databases often provokes a Big Brother scare. However the public is less inclined to view commercial data collections as a threat to personal privacy or liberties. This could in part be because such data collections are hidden, and their extent and use are not publicised. So while Identity card proposals, such as the Australia Card of the early 1980s, are almost universally condemned, more insidious electronic profiling such as supermarket loyalty cards meet with indifference, despite the fact that even more detailed data is available.

We set out to discover how much information could be available on a single person through publicly accessible data sources. The aim is to highlight the boundaries of what information can be collected and to consider how this fits in with the current legal position in the UK. The can also include information on a specific company, rather than a person. This is interesting because a company has the same legal rights as an individual under British law.

1. Introduction

The aim of this research is to determine the necessity for a tightening of UK data protection laws, or, optimally, to reassure us that these laws are indeed protecting citizens' information. We additionally establish a manual data seeking process, from which we will create a simple automated system. This system will query and cross reference specific information servers, outputting information on the given person or company. The purpose of this part of the work is to consider means of rendering software of this nature ineffective. The end result of this would enable a user to enter the persons'/company's details, for instance a name, a valid email address, or current home address. The possible output may consist of, but may not be limited to, the following:

• Full name
• Address/es
• Telephone number
• Social security details
• Driving licence details
• Income details
• Education
• Family details (possibly details of individual family members)
• Mail order/online purchases

2. Data Collection

2.1 Off-Line: CD-Rom - Containing Commercial Software Packages

The software gives a list of matches, and for each match there is the personsâ/company's initials, full address including house number and postcode, and of course the telephone number. This software is extremely powerful, and arguably useful for everyday life. Simply knowing some very basic facts about a person can enable any user to locate their home address, their telephone number, and even who they live with.

The UK government currently sells the "Register of Electors", to any company or individual who requests it. The Data Protection Registrar identifies the large scale usage for non-electoral purposes as follows [2]:

• Credit Referencing - the register provides information to Credit reference agencies used in the initial construction of individual's reference files. The information is combined with County Court Judgement information and information from Lenders who have extended credit to the individual (used by agencies such as Equifax and Experian).
• Other Risk Management - the Register is commonly used for basic identity checks, but the Registrar states that it is also combined with other information and can be used for vetting insurance claims, checking current/prospective employees, and even money laundering checks.
• Direct Marketing - The registrar identifies that there are two main marketing usages of the register:
  1. Targeted offers - Individual receive marketing offers because their details appear on register. These offers can be more carefully targeted with careful analysis of the register. A simple example targeting could be of individuals living on their own in certain areas - using the register, a marketer can identify addresses of individual who live alone and in the identified postal area. This information can be combined with telephone directories to give the individual's phone number. More complex analysis of the register, and how it changes from year to year, can enable marketers to identify targets such as "new movers", or those who fall in a particular age group.
  2. "Cleaning" - a marketer who has names and address from another source, can check them for accuracy by comparing the names to those on the register, removing those that do not appear on the assumption that they have moved.
• Law enforcement - the Police use the register to assist in tracing individuals. Local authorities use it to identify those who may be evading their legal responsibilities, such as tax avoidance.
• Tracing activities - can be used for any tracing activity from tracing individuals who have gone away leaving unpaid debts, to tracing those who have gone away to escape violence.
• Research - register used as a source to draw samples for research.
• The information provided in these search CDs must be from a computerised Electoral register and telephone directory combined.

2.2 Online: The Internet

If the majority of the popular sites are commercial, this raises the question of why there should be any personal information available on the Web. This section identifies the motivation for constructing profiles of individual users. The commercial Web companies are constantly looking for more and more ways to make money.

From outside of the industry, looking at how people make money online, simple intuition would suggest two routes:

• Selling goods or services online - basically an enhanced mail order company
• Charging to access information they provide, in similar way to subscription journals and magazines.

2.2.1 Banner advertisements

2.2.2 Transaction trails

• Logs of emails sent and received.
• Logs of Web pages visited.
• Logs of file transactions using other Internet tools - FTP, Telnet, IRC, MUD etc.

Large advertising companies such as DoubleClick serve advertisements on hundreds of sites (including freeserve, Disney, Infoseek), and it is possible for them to construct a profile of the pages a user views (forming similar data to that of the ISP Web page visiting log), and the frequency in which these pages are viewed, all by identifying a user from their cookie file. Using this information, advertising companies can offer highly tailored marketing solutions to their customers. The difference here between the ISP's log file and the marketer's cookie-based log is that the marketers can only collect non-personally- identifiable information (unless a user has registered with the advertising company to receive tailored advertisements and has volunteered the personal details). DoubleClick collects the following types of non-personally-identifiable information when it serves an advertisement [3]:

• IP address - further information can be inferred from this:
  • Users' geographic location.
  • Size of organisation whose network the user is accessing through.
  • Type of Organisation
  • Company/Organisation name
• User's domain type (i.e. .comâ .eduâ .ac.ukâ .govâ etc).
• How a user utilises pages visited within a customer's site

2.2.3 Email address collecting

Although the majority of marketers claim to only collect non-personally-identifiable information, some less scrupulous marketers have been exploiting a security flaw in many email programmes. One security problem is caused by email programs allowing HTML- formatted messages to write a cookie file on the user's computer. These programs include MS Outlook, Netscape Messenger, Eudora, and online HTML mail services such as Hotmail and Yahoo Mail. Companies send out "spam" emails, which include code to write a cookie file with a unique identification number. When the user subsequently visits a site hosting advertisements maintained by the marketer, the cookie file is found along with the profile cookie file (mentioned earlier) and the user's email address, which was associated with the new cookie file can be associated with the user's profile. This enables the marketer to send targeted marketing offers by email to the user, rather than just banner ads. This is a far more invasive means of advertising [5].

2.2.4 Volunteering personal details

Many of the most useful "free" services on the Internet require users to forfeit personal details in order to access the services or areas of their site. Examples include free e-mail providers, such as Hotmail.com, Yahoo-mail etc, and information providers, such as FT.com etc. Many also require users to complete a small survey, asking questions of marital status, job, salary etc - making users feel that it is a fair exchange for the ability to access the services.

2.2.5 Summary of Information collected

The following graph based on figures from a report to the Federal Trade Commission², which studied 100 commercial web sites and their privacy policies, shows clearly the sort of data that is collected.

2.2.6 Children being targeted

With rapidly increasing numbers of children, regularly surfing the net, especially in the UK, where schools are getting special Internet connection deals, concerning privacy issues are being raised. Recent surveys have discovered that some large companies have created sites aimed at children with specific intentions to gather information for marketing purposes. The companies manipulate children into divulging information, often personally identifiable, about themselves and their parents. To get this information, companies entice children with games and competitions that require the filling in of a questionnaire, before they can participate. Companies practising this include household names such as Lego (http://WWW.lego.com) and Crayola (http://WWW.crayola.com) .

With the rapid expansion of the web and companies' desire to build a base of loyal customers for the future (even though initially the customers are of course the parents!), these marketing techniques are becoming more widespread. CME's recent survey³ of 75 randomly chosen kids' sites and 80 top commercial children's sites, highlight how extensive these practices are - the results found that 95% of the random sample collect personally-identifiable information from children, with only 27% displaying a privacy policy! A similar survey by the FCC (US Federal Communications Commission) found some alarming statistics. Of the 212 web sites included almost all (96%) solicited the children's email address with almost half (49%) requesting their postal address. The following chart shows a clear summary the information solicited by sites in the survey⁴ .

With companies marketing specifically targeting children, we must begin to consider whether this is a legitimate use of their position (in providing sites that attract children) or whether children need protecting. CME's survey found that less than six percent of the sites they surveyed asked for parental consent. However even if more asked for consent, it would be extremely difficult to ascertain whether the consent is genuine - rather than the child forging consent.

3. Maintaining Privacy

There is no simple way to maintain privacy whilst online. There are services offered that help to avoid some of the data collecting techniques used by marketing companies, these are explained below. The only real way to maintain your privacy, is to be very careful of what you disclose and to whom, for a lot of people this means inventing a 'pseudo' identity which they can use whenever personal information is required.

As privacy is rapidly becoming a hot topic, there is a range of software products and services being developed that aim to help Internet users hold onto the privacy. Some of the more interesting ones are as follows:

3.1 PermissionTracker

This is a system developed by FollowUp.net, a provider of online, permission-based marketing for companies such as IBM, KPMG, Priceline.com, and Intel. From their publicity material, PermissionTracker creates a database of email addresses and possibly IP details, from users who specify how they will allow their information to be used. When a web company who has agreed to participate in the PermissionTracker scheme, receives a visit, they check the PermissionTracker database to see what privacy rights the user has allowed. Currently it appears that users who join the scheme can only specify that sites agree:

1. Not to sell or share any personal information collected from me on the Internet with other parties.
2. Not to aggregate or combine any personal information collected from me on the Internet with data from third parties.

There is also an option to opt out of DoubleClick's tracking.

On the PermissionTracker site (http://permissionTracker.com) there is a list of companies who have agreed to participate and also facilities to recommend companies who should be approached. At the time of writing it appears that the only company who has agreed to use the scheme is FollowUp.net, the creators! However, it is however early days yet.

3.2 Zero-Knowledge Systems

Zero-Knowledge Systems, offer a system that aims to give users total Internet privacy, its is called 'Freedom'(http://www.freedom.net). On the face of it, Freedom is a small client you launch from the desktop when you want to secure your Internet activities, be it browsing, mail, newsgroups, telnet, or IRC.

The Freedom system uses three components to conceal your online identity:

1. Nyms- pseudonyms you create and use online, these conceal your personal information by creating alternative details.
2. Encryption- on all outgoing data and messages.
3. The Freedom Network- a group of servers that route your Internet traffic through a series of "privacy-enhancing detours" that further encrypt your data and strip out all location information for your ISP, leaving only the 'Nym'.

The Freedom system appears to be very secure, not even Zero-Knowledge have the ability to open or expose your traffic. The system is double-blind, although the company have your credit card details (if you purchase on line) they cannot associate it with any individual Nym identity. According to Zero-Knowledge, as well as the Nym system, Freedom hides the source and destination IP addresses of your communication and encrypts the data flow (Freedom uses multiple cryptographic algorithms, such as Blowfish 128-bit encryption, DSA 1,024-bit keys, DH 2,048-bit keys). Apparently the most an outsider could find coming from a Freedom-equipped system is an encrypted stream of packets travelling to a Freedom server.

Zero-Knowledge have realised that the Freedom system could be abused by spammers and have implemented 'spam control' going into and out of the Freedom Network. The system will refuse to send excessive amounts of mail from any single Nym.

The Freedom system has had some excellent reviews, which are summed up with comments like the following, from Patrick Norton of PC Magazine :

"If online privacy is an issue for you, we can't think of a better option, except for not going online at all."

3.3 SiegeSoft

SiegeSoft offer three products to increase users' privacy, they are as follows(http://www.siegesoft.com) :

SiegeSurfer - This is a web-based proxy that relays pages either through clear text or through encrypted SSL.

SiegePipe - is a secure, private and anonymous Internet connection. All a users Internet activity appears to come from SiegeSoft.

SiegeWindowWasher - cleans up any user activity stored on Windows, such as Internet history, cache, recent documents and log files.

4. Current legal position

The following sections give a summary of the features of the Act that are relevant to an electronic profiling system.

4.1 The Data Protection Acts 1984 and 1998 (introduced 1/3/2000)

• Data User's Name
• Data User's Address

• Those about whom personal data are held.
• The items of data held.
• The purposes for which the data are used.
• The sources from which the information may be obtained.
• The types of organisation to whom the information may be disclosed i.e. shown or passed on
• Any overseas countries or territories to which the data may be transferred.

The penalty for not registering is a fine of up to 5000 pounds, if prosecuted in the Magistrates Court, or unlimited if convicted by the Crown Court [8].

The act has eight "Data Protection principles of good information handling" that registered users must comply with, these are summarised as follows, personal data must be:

• Obtained and processed fairly and lawfully.
• Held for lawful purposes described in data users' register entry.
• Used for those purposes, and disclosed only to those people, described in the register entry.
• Adequate, relevant and not excessive in relation to the purposes for which they are held
• Accurate and, where necessary, kept up-to-date.
• Held no longer than is necessary for the registered purposes.
• Accessible to the individual concerned who, where appropriate, has the right to have information about themselves corrected or erased.
• Surrounded by proper security.

4.2 The EU Data Protection Directive

a series of fair information practices that define obligations and responsibilities with respect to the processing of personal information ... personal information should only be collected legitimately for specific purposes (Directive 95/46/EC, at Art. 6(1)(a) and Art. 6(1)(b)).

The Internet, however, challenges the establishment of obligations and responsibilities with respect to the processing of personal information. Under current practices, the basic principle of a "purpose limitation" for personal information has become the exception rather than the rule in the on-line environment. Where paper records themselves had provided a physical barrier of sorts to any further use, information generated by individuals on the Internet is digital from the start and available for any number of further kinds of sharing and combination. Once on-line, the individual generates enormous amounts of personal data and further use has not been limited to compatible purposes ... for example a large amount of transactional information is collected by service providers who make various kinds of further use of these data.

4.3 Regulation of Investigatory Powers bill

On 9th February 2000 a controversial bill was unveiled in the House of Commons, this is the 'Regulation of Investigatory Powers bill' (RIP bill). It is thought to give the UK authorities more intrusive powers than any other western democracy⁵ . Although the Home Secretary Jack Straw claimed that "the new powers would be used mainly to track down serious criminals", but the legislation enables authorities to collect huge amounts of data on ordinary citizens. Civil rights experts expect the RIP bill to be challenged in the European Court of Human Rights if it is adopted⁶ . Mr Straw insists however that· "in my view the provisions of the Regulation of Investigatory Powers Bill are compatible with the Convention Rights"⁷ . The sections of the bill that are most alarming from an Internet privacy perspective is that the bill requires ISPs to become party to secret surveillance of their customers, siphoning off Internet traffic into government computers. They can also be required to provide the authorities with detailed traffic analysis, this could include every email address and Internet site to which an individual had used, and possibly any correspondence between them. The bill also gives the authorities access to any encryption keys, with the presumption of guilt if anyone fails to produce their encryption key.

5. Public opinion

• In a recent survey of Web users 39% of respondents "agree strongly", and 33% "agree somewhat" that there should be new laws to protect privacy on the Internet. Interestingly respondents from Europe generally felt less strongly about this, with 18% of respondents being "neutral" on the issue and 39% "somewhat" agreeing [13].
• The same survey found that the most disagreed statement was that "Content providers have the right to resell user information". 63% of respondents "Disagree strongly" while another 19% "disagree somewhat". In total 82% of respondents are in disagreement with the reselling of their personal information [13].
• In a public opinion poll on medical privacy, 75% of respondents were concerned a "great deal" or "a fair amount" about insurance companies putting medical information about them into a computer information bank, that others have access to [14].

6. The automated system

In order to establish what information can be uncovered on individuals and individual companies by an automated system, we researched Websites that reliably provide information on individuals. The majority of the sites only provide small quantities of personal data, but by data-matching these small quantities of data can contribute to a richer data profile. In order to do the data-matching, the set of contributing sites needs to be as accurate and complete as possible. A non-exhaustive list of the sites potentially useful to the automated profiling system is included in the Appendix. We aim to document the theoretical maximum of data that can be achieved, and by what combination of these sources. The process may include an element of recursion because, as more information is uncovered, earlier sources in the chain may provide more output data given richer starting data.

6.1 Design Concept

One of the key aims of this project was to ascertain the feasibility of a software system that only using information available via the Web, would uncover personally identifiable information on a search target, thus highlighting how the Internet is affecting our privacy. Therefore although the easiest means of getting basic details would be using the electoral register (in the UK or other similar databases), this would require locally stored data, and hence would not fulfil this aim. The basic process that the program fulfils is similar to that of a human researcher or 'private detective' using the resources on the web to locate a target, the resulting software, will highlight how easy the process is, by giving the search a clear structure. The following diagram clearly illustrates the design concept:

Each layer in the search can be expanded to include different resources, and add to the variety of data found, as well as tailoring it for different locations. The initial steps in the process require user input to select the most likely target from the list produced. The system could perform a search on all the possibilities, but this would take a considerable amount of time, due to the slow Internet connections and frequently slow servers (due to high traffic) that the system uses.

6.2 Ethical issues of Software

The key aim in constructing this software was to establish the feasibility of an electronic profiling engine that would invade individuals privacy. As the implementation is primarily a feasibility study it was felt that it was very important that the information returned by the system wasn't 'too' invasive. However it should be alarming enough to raise public interest. After much consideration, information servers would chosen which, hopefully, adhere to these considerations. The system, given basic information, name details and/or email address, will return (where possible) full postal address of target including postcode (zip-code), telephone number, email address, and a series of maps showing the location of the targets house. Although this information is very easy to find, without any form of authorisation of justification required, it is all available from a county library and therefore can be seen as not 'too' invasive.

The key question of ethics for these types of program is really what or how much information should be private and therefore shouldn't be available via any source? In most cases when we volunteer personal information we are not aware exactly what it will be used for - with most users (understandably) assuming that the current requirement for the data will be its only use. What is really needed is a globally defined structure for privacy statements. The current problem is that the majority of sites do not even have privacy statements, and users don't consider necessarily consider how their privacy is being affected. Where sites do have privacy statements, they are often quite long and written in a convoluted non-user-friendly manner. This means that users who are concerned about their privacy are put off reading the document.

If rigid format were to be given to privacy statements, it would make the statements easier to understand and perhaps the public would then expect all sites to have one. Ultimately features could be added to web browsers in a similar way to security certificates that would warn the user when they were accessing a site that doesn't posses a privacy statement or where the statements exist, if it contradicts the user's privacy settings. This would allow individual users to specify what purposes they are happy for any data to be collected for, and hands the control of personal data back to the individual.

In conclusion

The free availability of large collections of data does not in itself constitute an infringement of personal privacy. As discussed in [15], a major problem is the "transliteration" of existing databases and information searching and retrieval practices to the electronic media, particularly the Web. The nature and purpose of queries are no longer screened, and the identity of the agent requesting the information is not required. For legitimate uses, this is highly appropriate, but it also provides no protection, or audit trail, against misuse of the information.

Also, it is interesting to note that the greatest threat to personal privacy does not necessarily come from government agencies, but from the more hidden activities of commercial data collectors. These commercial databases have far more detail than any governmental Identity card proposals, as they include the most minor details of an individual's life, such as what items they purchase and what brand, time and place of purchase, Web browsing habits and trails, and many other seemingly trivial pieces of information which become important when taken in context with other such pieces of information. Orwell's Big Brother is a reality, but is not necessarily a government organisation.

Hypertext References

1. UK InfoDisk, Published by I-CD Publishing (UK) Ltd.
2. "The Sale of the Register of Electors", Submission by the Data Protection Registrar to the Home Office Working Party on Electoral Procedures, http://www.dataprotection.gov.uk/reg-elec.htm, August 1998
3. http://www.doubleclick.net/company_info/about_doubleclick/privacy/non_identify.htm
4. Dan Schiller: "Tradesmen launch Assault on the Internet", Le Monde Diplomatique, March 1997
5. "The Cookie Leak Security Hole", R.M. Smith, 30/11/99, http://www.tiac.net/users/smiths/privacy/cookleak.htm
6. Figures from "Privacy and The Top 100 Web Sites: Report to the Federal Trade Commission", M.J.Culnan, Ph.D, Georgetown University, June 1999
7. "Registration & Data Protection", http://www.dataprotection.gov.uk/register.htm
8. "Data Protection - a brief Introduction", Dec 1998, www.dataprotection.gov.uk/intro.htm
9. http://www.dataprotection.gov.uk/eurotalk.htm
10. "Defaults Guidance" - DP Act 1984, E. France - Data Protection Registrar, www.dataprotection.gov.uk/defaults.htm
11. "Data Protection Law and On-line Services: Regulatory Responses", J.R.Reidenberg Prof. Law - Fordham University, P.M.Schwartz Prof. Law - Brooklyn Law School, http://www.europa.eu.int/comm/dg15/en/media/dataprot/studies/regul.pdf.
12. "Consumer Privacy Concerns about Internet Marketing", Communications of the ACM, Vol.41 No.3, March 1998
13. Graphic, Visualization, & Usability Centre's (GVU) 8th WWW User Survey, www.gvu.gatech.edu/user_surveys/survey-1997-10/
14. "Health Information Privacy Survey", Harris Equifax, 1993, http://www.epic.org/privacy/medical/polls.html
15. H. Ashman, "Transliteration of Databases and Information Retrieval Practices to Public Websites", submitted for publication, 2000.

Further reading

Hard copy

• J.H.Ellsworth and M.V.Ellsworth, Marketing on the Internet, Wiley, 1997
  
• D.Banisar, S.Davies, Privacy and Human Rights 1999: An International Survey of Privacy Laws & Developments, Electronic Privacy Information Center, 13/8/1999.

On-line Resources

• A Guide to Developing Data Protection Codes of Practice on Data Matching, http://www.dataprotection.gov.uk/match.htm
• Data Protection Act 1984, http://www.hmso.gov.uk/acts/acts1984/1984035.htm, 1984
• Data Protection Act 1998, http://www.hmso.gov.uk/acts/acts1998/19980029.htm, 1998
• Data Protection and the Internet - Guidance on Registration, http://www.dataprotection.gov.uk/internet.htm
• Defaults Guidance - DP Act 1984, http://www.dataprotection.gov.uk/defaults.htm, 6/1999
• DoubleClick Privacy Statement, http://www.doubleclick.net/company_info/about_doubleclick/privacy
• Privacy Threatened By New Databases, http://www.dataprotection.gov.uk/dbasepr.htm, 15/7/97
• The 15th Annual Report of the Data Protection Registrar, http://www.dataprotection.gov.uk/99arcontents.htm, 6/1999
• The Eighth Data Protection Principle and Transborder Dataflows, http://www.dataprotection.gov.uk/transbord.htm
• The Sale of the Register of Electors, http://www.dataprotection.gov.uk/reg-elec.htm, 8/1998
• Using the Law to Protect your Information, http://www.dataprotection.gov.uk/intro.htm, 12/1998
• Consumer Privacy Concerns about Internet Marketing, H.Wang, M.K.O.Lee, C.Wang, http://www.acm.org/pubs/articles/journals/cacm/1998-41-3/p63-wang/, 3/1998
• You've Got mailing Lists, J.Duffy, http://www.news.bbc.co.uk/hi/english/uk/newsid_557000/557505.stm, 10/12/1999
• Data Protection Law and On-Line Services: Regulatory Responses, J.R.Reidenberg, P.M.Schwartz, http://www.europa.eu.int/comm/dg15/en/media/dataprot/studies/regul.pdf
• Internet Privacy Conerns confirm the case for Intervention, R. Clarke, http://www.acm.org/pubs/articles/journals/cacm/1999-42-2/p60-clarke/, 2/1999
• Privacy and the Top 100 Web Sites: Report to the Federal Trade Commission, M.J.Culnan, http://www.msb.edu/faculty/culnanm/GIPPS/oparpt.pdf, 6/1999
• Recommendation 1/99 on Invisible and Automatic Processing of Personal data on the Internet Performed by Software and Hardware, P.Hustinx, http://www.europa.eu.int/comm/dg15/en/media/dataprot/wpdocs/wp17en.htm, 23/2/1999
• Information Privacy on the Internet Cyberspace Invades Personal Space, R.Clarke, http://www.anu.edu.au/people/Roger.Clarke/DV/Iprivacy.html, 2/5/1998
• Information Technology and Dataveillance, R.Clarke, http://www.anu.edu.au/people/Roger.Clarke/DV/CACM88.html, 11/1987
• The Cookie Leak Security Hole in HTML Email Messages, R.M.Smith, http://www.tiac.net/users/smiths/privacy/cookleak.htm, 30/11/1999
• On-line Srvices and data protection and the Protection of Privacy, S.Gaurthronet, F.Nathan, http://www.europa.eu.int/comm/dg15/en/media/dataprot/studies/servint.htm, 12/1998
• How anonymous is the Web?, W.Rodger, http://www.usatoday.com/life/cyber/tech/ctg802.htm, 3/12/1999

Footnotes

1. A particularly worrying example is that of the Morgan Stanley Dean Witter bank who "collect", among other things, details about an individual's race, religious beliefs, sexual preferences, union membership, etc. As this information is never required as part of the credit application procedures, it is most likely inferred by analysing the individual's subsequent spending pattern. This is similar to the way supermarket chains infer such things as marital status, number and age of depdendents etc. using their "loyalty cards" to analyse purchasing patterns. Morgan Stanley Dean Witter also claim the right to disseminate this information .
2. Figures from "Privacy and the Top 100 Web sites: Report to the Federal Trade Commission", M.J. Culnan, Ph.D. Georgetown University, June 1999.
3. CME Assessment of Data Collection Practices of Children's Web sites", Center for Media Education, July 1999.
4. "Online Marketing to Children", US Federal Communications Commission (FCC), 1998.
5. "Leader: Spies in the Web", The Financial Times, 7/3/2000
6. "Ripping into UK Privacy Bill", Kalin Lillington, 23/3/2000, Wired News.
7. "On the Cover of the 'Regulation of Investigatory Powers Bill'", 9/2/2000, http://www.publications.parliament.uk/pa/cm199900/cmbills/064/2000 064.htm

Copyright

[ Proceedings ]