An analysis of the use of biometric identification using Australia as a case study
Social spending stays at historically high levels in many OECD countries averaging about 21% of GDP in 2016. While total public social spending covers all financial flows from public bodies for social purposes, total net social spending also considers taxes breaks as well as private spending for social purposes. In the US, the UK and Canada, total net social spending is much higher than public spending for social purposes. The costs of fraud and error in social security systems can be considerable. Given the current fiscal climate, efforts to combat fraud and error are at the forefront of policy agendas in many countries. A common assumption is that the economic recession has led to growth in social welfare fraud. Despite increased media interest, there is limited available data internationally showing actual levels of fraud in social security systems and their wider economic costs to society. Furthermore, there are important knowledge gaps on what works in mitigating against the issues of fraud and error. This paper discusses the solution for Australia of a biometric ID cards for all citizens to be shown and tested as time of receipt of welfare (also to be used for other identification purposes). The time is now ripe to consider such a solution. This also fits into increased demands for domestic and international security and refugee identification.
Key words: welfare costs, welfare fraud, biometric identification.
Biometrics covers a variety of technologies in which unique identifiable attributes of people are used for identification and authentication. These can be used to validate the identity of individuals seeking to control access to computers, airlines, databases, government services and other areas which may need to be restricted.[i]
1.1 Types of Biometrics
- DNA Matching (Chemical Biometric) – The identification of an individual using the analysis of segments from DNA.
- Ear (Visual Biometric) – The identification of an individual using the shape of the ear.
- Eyes – Iris Recognition (Visual Biometric) – The use of the features found in the iris to identify an individual.
- Eyes – Retina Recognition (Visual Biometric) – The use of patterns of veins in the back of the eye to accomplish recognition.
- Face Recognition (Visual Biometric) – The analysis of facial features or patterns for the authentication or recognition of an individual’s identity. Most face recognition systems either use eigenfaces or local feature analysis.
- Fingerprint Recognition (Visual Biometric) – The use of the ridges and valleys (minutiae) found on the surface tips of a human finger to identify an individual.
- Finger Geometry Recognition (Visual/Spatial Biometric) – The use of 3D geometry of the finger to determine identity.
- Gait (Behavioural Biometric) – The use of an individual’s walking style or gait to determine identity.
- Hand Geometry Recognition (Visual/Spatial Biometric) – The use of the geometric features of the hand such as the lengths of fingers and the width of the hand to identify an individual.
- Odour (Olfactory Biometric) – The use of an individual’s odour to determine identity.
- Signature Recognition (Visual/Behavioural Biometric) – The authentication of an individual by the analysis of handwriting style, in particular the signature. There are two key types of digital handwritten signature authentication, Static and Dynamic. Static is most often a visual comparison between one scanned signature and another scanned signature, or a scanned signature against an ink signature (often used by banks). Technology is available to check two scanned signatures using advances algorithms. Dynamic is becoming more popular whereby information is recorded as the signor writes the signature on a “signing device”. This data can be utilised in a court of law using digital forensic examination tools, and to create a biometric template from which dynamic signatures can be authenticated either at time of signing or post signing.
- Typing Recognition (Behavioural Biometric) – The use of the unique characteristics of a person’s typing for establishing identity.
- Vein Recognition – is a type of biometric that can be used to identify individuals based on the vein patterns in the human finger or palm.
- Voice / Speaker Recognition There are two major applications of speaker recognition:
- Voice – Speaker Verification / Authentication (Auditory Biometric) – The use of the voice as a method of determining the identity of a speaker for access control.
If the speaker claims to be of a certain identity and the voice is used to verify this claim. Speaker verification is a 1:1 match where one speaker’s voice is matched to one template (also called a “voice print” or “voice model”). Speaker verification is usually employed as a “gatekeeper” in order to provide access to a secure system (e.g.: telephone banking). These systems operate with the user’s knowledge and typically require their cooperation.
- Voice – Speaker Identification (Auditory Biometric) – Identification is the task of determining an unknown speaker’s identity. Speaker identification is a 1:N (many) match where the voice is compared against N templates. Speaker identification systems can also be implemented covertly without the user’s knowledge to identify talkers in a discussion, alert automated systems of speaker changes or check if a user is already enrolled in a system.
In forensic applications, it is common to first perform a speaker identification process to create a list of “best matches” and then perform a series of verification processes to determine a conclusive match.
Note: There is a difference between speaker recognition (recognising who is speaking) and speech recognition (recognising what is being said). These two terms are frequently confused, as is voice recognition. Voice recognition is a synonym for speaker, and thus not speech, recognition. In addition, there is a difference between the act of authentication (commonly referred to as speaker verification or speaker authentication) and identification.[ii]
1.2 History of Biometrics
The first biometric to be widely applied was the fingerprint, although there is argument about who developed the concept and when. Chinese records from the Qin Dynasty (221-206 BC) include details about using handprints as evidence during burglary investigations. Clay seals bearing friction ridge impressions were used during both the Qin and Han Dynasties (221 BC – 220 AD).[iii] The Persians (1400s) and Europeans (from 1600s onwards) were increasingly interested about the linkage between physiogamy and behaviour, using both fingerprints and phrenology to predict temperament. The British in India used the fingerprint (thumb often) or whole palm print in the execution of contracts between the government and natives who could not write.
It was in 1882 that Alphonse Bertillon, a clerk in the Prefecture of Police of at Paris, France, devised a system of classification, known as anthropometry or the Bertillon System, using measurements of parts of the body. Bertillon’s system included measurements such as head length, head width, length of the middle finger, length of the left foot and length of the forearm from the elbow to the tip of the middle finger.In 1888 Bertillon was made Chief of the newly created Department of Judicial Identity where he used anthropometry as the primary means of identification. He later introduced Fingerprints, but relegated them to a secondary role in the category of special marks.
Real progress was made by Sir Francis Galton, British anthropologist and a cousin of Charles Darwin, who began his observations of fingerprints as a means of identification in the 1880’s. Juan Vucetich, an Argentine Police Official, began the first fingerprint files based on Galton pattern types. Francis Galton published his book, “Finger Prints” in 1892, establishing the individuality and permanence of fingerprints. The book included the first published classification system for fingerprints. In 1893, Galton published the book “Decipherment of Blurred Finger Prints,” and 1895 the book “Fingerprint Directories.”
Galton’s primary interest in fingerprints was as an aid in determining heredity and racial background. While he soon discovered that fingerprints offered no firm clues to an individual’s intelligence or genetic history, he was able to scientifically that fingerprints do not change over the course of an individual’s lifetime, and that no two fingerprints are exactly the same. According to his calculations, the odds of two individual fingerprints being the same were 1 in 64 billion. A few of these same characteristics (minutia) are basically still in use today, and are sometimes referred to as Galton Details.[iv]
During the 20th Century wider application was made of fingerprinting and records kept for criminals, members of the defence forces, government employees and eventually for voluntary identification cards.
1.2.2 Facial Recognition
This utilises image or video in order to compare facial features from the chosen source to cataloged entries in a database. The technology works by creating a virtual grid and outlining the distance between defining characteristics on the face, as well as detailed information on the shape of the face: such as the contour of the nostrils, eyes, and even analyzing the texture of the skin.
It was initially pioneered by Woody Bledsoe, Helen Chan Wolf, and Charles Bisson in 1964 as part of their collective study on pattern recognition intelligence (PRI). After Bledsoe left the study of PRI, the research was continued at Stanford Research Institute by Peter Hart. In experiments performed using the collective work of the initial pioneers, as well as his own, the first major breakthrough happened in 1968 when a computer consistently outperformed humans in identifying human faces from a database of 2,000 photos.
In 1997, Christoph von der Malsburg as well as a team of graduate students from the University of Bochum in Germany developed a system known as ZN-Face that was (at the time) the most robust of its kind due to its ability to make facial matches on imperfect images. The technology was funded by the United States Army Research Laboratory and is used by customers ranging from major international airports, to banks and government offices.
Current facial recognition technologies rely on cross-referencing characteristics of several different facial recognition technologies and algorithms, and are said to be so powerful that they can not only significantly outperform humans, but also correctly identify the individual faces of identical twins.
Variations of this technology are currently being used in consumer grade applications, such as the Xbox One. Using Kinect Sign-In, users can sign in to their Xbox profile after the Kinect scans their face and body profile in order to make a positive identification. The facial scanner is impressive, and even works in rooms full of others where it has to pick your profile from a group.[v]
1.2.3 The Eye
Although the modern technology surrounding iris scanning and recognition is rather new, the science behind the technology (iridology) dates back to ancient Egypt and Greece, and even appeared in the writings of Hippocrates. The modern pioneer of the technology is John Daugman, who developed and patented the first algorithms for computer-aided identification of iris patterns in 1994. Although the algorithms and equipment – such as scanners and the means to catalogue and retrieve samples – has improved since this time, Daugman’s algorithms are still the foundation behind all public deployment of iris recognition technology.
Retina scanning was the first ocular biometric technology used, but the technology has since been replaced by the iris scan, the more advanced – and reliable – of the two technologies. Iris recognition technology is an automated identification system that uses mathematical pattern recognition to map the complex patterns of an individual’s eye. The scanners attempt to match these patterns to a database using images or video representation of the individual’s eye.
Iris recognition has been, until recently, a relatively expensive technology that sort of priced its way out of consumer applications. That could be changing soon, as technologies like EyeLock – a device used to lock down your home computer with its own iris scanner – start to come to market.
1.2.4 Other Technologies
Every other biometric option in operation or under development relies upon computer technology and has its origin within the past 40 years. Some, like typing rhythm are used in tracking computer use and hacking, others like gait and voice are more likely to be used for security in buildings as an alternative to key cards.
For more abundant security, biometric systems are now employing a blend of technologies, e.g. face recognition and fingerprints, iris scanning and voice or sometimes more than two options. The drivers behind the adoption of biometrics are relatively obvious and are based upon the human proclivities of negligence and greed – passwords or identification cards can be lost, forgotten or stolen, the latter for personal, financial gain.
With increasing use of IT technology and need to protect data, everyone now has multiple accounts entailing passwords. Most people can only remember a few passwords, so it is likely that familiar prompts are used – birthdays, family nicknames, dog, cat, school friends and the like. It is remarkably easy to crack passwords because most of our passwords are categorised as weak. It is always recommended to include upper and lower case letters and a number but the creation of such strong passwords that have no particular meaning they are likely to be forgotten.
2.0 Leading Applications
2.1 Identification – India’s national ID program called Aadhaar is the largest biometric database in the world. It is a biometrics-based digital identity assigned for a person’s lifetime, verifiable online instantly in the public domain, at any time, from anywhere, in a paperless way. It is designed to enable government agencies to deliver a retail public service, securely based on biometric data (fingerprint, iris scan and face photo), along with demographic data (name, age, gender, address, parent/spouse name, mobile phone number) of a person. The data is transmitted in encrypted form over the internet for authentication, aiming to free it from the limitations of physical presence of a person at a given place. About 550 million residents have been enrolled and assigned 480 million Aadhaar national identification numbers as of 7 November 2013. It aims to cover the entire population of 1.2 billion in a few years.[vi]
2.2 Smart Phones – It is predicted that every smartphone sold in 2018 will incorporate biometric technology. The days of the four digit PIN will be gone (Apple is already there) and an index finger placed on a key will open the phone for calls, accessing contacts and use of all apps. The integration of smart phones into everyday activities including cardless cash and domestic security will make this biometric application all – pervasive in the future.
2.3 Airport Security – Iris recognition has been used in a number of large airports for several years. To participate, a traveler must sign up to the scheme and have his or her eyes and iris scanned. The details are then stored on an international database and instead of waiting in long passport queues, passengers simply walk into a booth and look into a camera. The software then scans the iris and matches the details with the information stored on the database.
2.4 Building Access – Fingerprint access for buildings is not new. Many high security facilities have used biometric technology for years when it comes to ensuring only authorised personnel gain access to the most well protected establishments.
2.5 Automobiles – It is estimated that over 2 million cars in the UK use biometric technology in one form or another. This can be anything from voice recognition when using Bluetooth or entertainment systems to unlocking the vehicle itself using digitalized fingerprints or face recognition.
2.6 Blood Banks – Security for blood banks is vital when proving identity of donors. Traditionally donors were issued with cards containing all information necessary following questionnaires. More and more these days that crucial data is being stored digitally – with donors using finger print or iris recognition to access their vital details. This data can be linked to the full healthcare profile of all citizens if approved by the donor.
2.7 Education – Recently the UK Association of School and College Leaders estimated that about 30% of secondary schools in England were using biometric data. As well as ensuring that only pupils and authorised adults gain entry to school buildings, biometric data is also used for activities such as; recording attendance, checking out library books or even paying for meals. Due to the sensitivity surrounding children and privacy; schools require full, written permission from the parents of pupils in order to use the system.[vii]
3.0 The Case For and Against Biometrics
John Mears, Senior Fellow at Lockheed Martin offered a comprehensive discussion of biometric modalities and uses. Data collected from a biometric is captured and then translated into an electronic template. This template can be used to then compare a subsequent capture of biometric information to either authenticate that a person is who they say they are—or identify someone whose identity may be in question. The prior authentication is generally found, for example, in gaining access to a mobile phone or in facial recognition at a bank ATM. The latter identification is most typically found in law enforcement use or national security needs – matching fingerprints to a crime scene for example.[viii]
All this security and reassurance comes at a considerable cost, beyond the bounds of all but governments and big business. The promotional material emphasises the benefits of adoption in both physical & digital security control such as: streamlining operation, enhancing productivity, minimizing security related losses, and reducing losses.[ix]
The real potential for large scale usage really mimics the Indian model of Aadhaar, particularly in this age of terrorism and porous borders. Governmental applications are therefore under the spotlight with advocates stressing the advantages for citizens of a universal ID card containing everything from a person’s health records, educational achievements, employment history, taxation returns, criminal record, welfare benefits, marital status and, in fact, everything that a national census might discover and much more.
On the other side of the argument are those philosophically opposed to government intrusion into one’s private life – the Orwellian hatred of Big Brother. President Obama and former Australian Prime Minister John Howard both tested the public reaction to national ID cards and found the pressure against them too politically unpalatable[x]. Nevertheless, there is now so much information stored by governments and business about every person, their habits, spending patterns, banking information, etc. that such squeamishness would seem to be misplaced. Every time someone uses the internet, companies can trace their interests and respond with targeted advertising or, worse still, unsolicited tele-marketing.
Yet, many countries have already taken the plunge whether by upgrading the passport system or by issuing an all-encompassing, compulsory national ID card, the most prominent being Brazil, China, Germany, the Netherlands, India, Indonesia, Israel, Pakistan, Poland, Singapore, Thailand and the United Arab Emirates but there are many more.[xi]
In Australia one of the strongest cases for biometric ID is to prevent rampant voter fraud, as no ID other than an address is required for voting, unless voting outside of one’s electorate.
3.1 Biometric ID to promote Electoral Certainty
The Australian Electoral Commission has revealed that over 18,000 people are being asked why they voted more than once in the recent election. The Commission did not reveal which electorates are involved. This is of particular concern in those electorates where the candidate won by a handful of votes, in one case as low as 37. Moreover, the government enjoys a majority of only one seat. The AEC has also revealed that while there were just under 7000 cases of suspected voting fraud in the 2013 federal election, but not one conviction was recorded for electoral fraud. Where margins between winning and losing can hinge on a few votes, this situation can lead to electoral fraud.
The “reforms” introduced by the Hawke government to make it easier to vote, also made it easier to vote more than once. The crucial change was the abolition of the requirement that a voter should vote in a polling station in an electoral subdivision. In addition, no identification is required, such as the identification which is necessary to collect or send certain parcels at the Australian Post Office. One solution to this gap in security is that the AEC send voters a bar-coded card which could be read when voters’ names are ruled off. However, if people move, or letterboxes are not secured this is subject to security issues.
We have already permitted fraud by not closing electoral rolls when the election is called and by allowing extensive prepolling. GetUp!, a left wing organization, said 100,000 additional names went on to the roll, with many of these in marginal seats after a High Court decision to allow these practices. GetUp! claimed its action had ensured that the Gillard government survived, albeit as a minority government. The election was very close and it is clear that had the High Court ruled against the legislation the opposition would have won the 2010 election.
Electoral fraud remains rife in Australia. Given the indifference to this, the dismissal that it actually takes place and the strong opposition to any reform, it is reasonable to assume that some political parties regard electoral fraud as a useful campaign tool. If we introduce a requirement for a biometric ID card to be produced to be checked by scanning the person and for the person’s name to be ruled off on an electronic roll, this would block any second or further voting anywhere in Australia or indeed the world.
4.0 A Cost Benefit Analysis of Biometrics – Australia as a Case Study
The Australian “SmartGate” passport, border control and travel system makes full use of face recognition technology, although use of the data is governed by the nation’s leading edge Biometrics Privacy Code, which is established and administered by the Biometrics Institute. In addition to faster processing and a reduction in queues, contactless technology should deliver stronger security at this island nation’s borders.
In particular, the SmartGate technology is designed to address the occurrence of transnational crimes being crimes that are violations of law that involve more than one country in their planning, execution, or impact. These offences are distinguished from other crimes in their multinational nature, which poses unique problems in understanding their causes, developing prevention strategies, and in mounting effective adjudication procedures. Transnational crimes can be grouped into three broad categories involving provision of illicit goods (drug trafficking, trafficking in stolen property, weapons trafficking, and counterfeiting), illicit services (commercial sex and human trafficking), and infiltration of business and government (fraud, racketeering, money laundering, and corruption) affecting multiple countries. Transnational crimes are distinct from international crime, which involves crimes against humanity that may or may not involve multiple countries. Examples of international crimes are genocide and terrorism, which are also included in this guide to sources.[xii]
Aside from the obvious security benefits, including the exclusion of undesirable immigrants and efficient interaction with Interpol, the automated immigration clearance technology is expected to save the Australian government $32.9 million per year in compliance costs but the trial that commenced in 2015 has involved the installation of a number of biometric SmartGate systems at eight of Australia’s busiest airports costing A$158 million. The big question is how much is a nation’s security worth? The prevention of a single terrorist crime involving homicide is incapable of quantification.
The Migration Amendment (Visa Revalidation and Other Measures) Bill 2016 currently before the Senate proposed to introduce a visa revalidation requirement. This will support the pilot of the proposed new 10-year validity visitor visa to ensure that only genuine visitors retain the right to travel to and enter Australia. As tourism is a vital part of the Australian economy, providing employment for around 600,000 Australians and accounting for more than $120 billion of economic activity, this measure will boost tourism across the nation, which in turn helps grow our domestic economy. The introduction of a 10-year visitor visa will encourage repeat visits by genuine tourists and businesspeople choosing Australia as their preferred destination. The boost to the economy will swamp any costs of introduction and is estimated at $11billion.
A biometric ID requirement for all persons resident, or nonresident entering or leaving Australia, or applying for a loan, a license, rental accommodation or to vote, will help track and reduce the potential for such crimes in Australia, particularly human trafficking, to which Australia is particularly vulnerable. It will also enhance cross checking of refugee status and cooperation with other nations seeking criminals.
4.1 Summary of Cost Benefit arguments for
- The next generation of SmartGates use leading edge technology to confirm a traveler’s identity without the traveler having to present their ePassport. The live facial image of the traveler at the SmartGate will be compared with a verified image from departmental holdings to confirm a traveler’s identity. In addition to faster processing and a reduction in queues, contactless technology delivers stronger security at our border, as the identity of the traveler is based on a facial image as a unique identifier, instead of document based checks.
- Benefits of reduction in time, security of our borders, reduction in crime, are expected “improvements to contactless automated immigration clearance technology (SmartGates) are expected to save the Commonwealth $32.9 million a year in compliance costs. Offsetting that is the budget allocation of $93.7m to the seamless traveller initiative.
- Migration Amendment (Visa Revalidation and Other Measures) Bill 2016 is currently before the Senate proposed to introduce a visa revalidation requirement. This will support the pilot of the proposed new 10-year validity visitor visa to ensure that only genuine visitors retain the right to travel to and enter Australia. As tourism is a vital part of the Australian economy, providing employment for around 600,000 Australians and accounting for more than $120 billion of economic activity, this measure will boost tourism across the nation, which in turn helps grow our domestic economy. The introduction of a 10-year visitor visa will encourage repeat visits by genuine tourists and businesspeople choosing Australia as their preferred destination. The boost to the economy will swamp any costs of introduction and is estimated at $11billion
- The word transnational describes crimes that are not only international (that is, crimes that cross borders between countries), but crimes that by their nature involve cross-border transference as an essential part of the criminal activity. Examples of include: human trafficking, people smuggling, smuggling/trafficking of goods (such as arms trafficking and drug trafficking and illegal animal and plant products and other goods prohibited on environmental grounds (e.g. banned ozone depleting substances), sex slavery, terrorism offences, torture and apartheid. Transnational crimes may also be crimes of customary international law or international crimes when committed in certain circumstances. It is estimated 21million people are trafficked worldwide and this crime is linked to immigration fraud and use of corrupt agents to facilitate recruitment and travel of trafficked persons across state borders. As trafficking and slavery are commonly motivated by profit, crimes relating to money laundering may occur as part of the trafficking process. There is a lack of research about the overlap of human trafficking crimes with other types of crime, particularly in Australia’s Pacific region. However, previous AIC research on the vulnerability of people to human trafficking in the Pacific region noted that such risks can be heightened in localities where unregulated and/or illegal logging and fishing industries occur
- A biometric ID requirement for all persons resident, or nonresident entering or leaving Australia, or applying for a loan, a license, rental accommodation or to vote, will help track and reduce the potential for such crimes in Australia, particularly human trafficking, to which Australia is particularly vulnerable.
- It will enhance cross checking of refugee status and cooperation with other nations seeking criminals.
4.2 Arguments against
4.2.1 Unquantified costs.
The official estimates of government costs have varied widely during the course of the campaign. A variety of omissions and under-estimates remain, including the compliance costs of government agencies themselves.[xiii] There are also significant errors in calculating personnel requirements (the overheads of supervision, staff turnover and leave were omitted).
The cost and inconvenience to individuals in complying with requirements are totally ignored. Recent ABS statistics show that at the end of each year 15% of the population are at a different address within the same State, and a further 1.7% are at a new address interstate. After allowing for international movements, and multiple moves by the same family, the volatility of the 16 million addresses on the Register would appear to be above 20% per annum.
Costs to the private sector may be vast, since every company in the country would need to change complex and, in many cases, ancient payroll and creditors systems, and every investment system in the country’s banks, building societies, credit unions, trusts, insurance companies, solicitors’ offices and even real estate agents would have to be modified.
Both during the issue phase and subsequently, many employees “would need time off from work to attend interviews, collect cards, advise change of address and lost cards, and collect original and replacement cards.
These costs were entirely omitted from the Government’s considerations. Remarkably, there was an attempt by an academic economist to justify the exclusion of all non-government costs from the cost/benefit analysis.
4.2.2 Security threats of hacking – Domestic And Global
Biometric measurements could be held in a centralised data base. Even if decentralised and data stored on cards, there is still need for a data base for card checking purposes. There is concern that the data bases could be compromised and acquired by those wishing to use it for gain. In addition there is concern false acceptance and rejection of individuals.
However, in 2014-15, over 19 million travellers arrived into Australia, of which over six million self-processed through our automated immigration clearance system known as SmartGate. There have been no problems. Also there are a number of methods to prevent illegal access of databases, especially when held in clouds, such as file-level and share-level security, password-protection, EFS and disk encryption, hid with steganography, protection of data in transit with IP security, securing wireless transmissions and use of rights management to retain control. There are also a handful of existing web standards which companies in the cloud know about. Chief among these is ISO27001, which is designed to provide the foundations for third party audit, and implements OECD principles governing security of information and network systems. The SAS70 auditing standard is also used by cloud service providers.
Given the state of the art which has been around for decades this should not be a barrier to progress. Paper passports can be faked as can all other forms of ID. Biometric storage of data would be subject to so many controls and other backup forms that this should not be considered an obstacle.
The resistance against properly administered biometric ID cards whether as a small format passport or as a separate form of personal validation, will gradually disappear. Those with nothing to hide are most likely already resigned to the establishment of such a system. Those with strong views about privacy will need to take up residence on a desert island if they are to avoid the surveillance of business and government. It is merely a short step to connect the data readily available about every person who uses any form of computerised technology and to amalgamate it into a single card.
The protections of privacy legislation within a democratic system offers many safeguards, but the use of such material in a totalitarian regime is less comforting. However, it is the totalitarian or single party states that have embraced the ID card whether the populace approves or not. Common Law countries can rely upon a free press and an impartial judiciary for protection against governmental and business abuse of personal information.
 PhD scholar at the Fenner School, Australian National University.
[vi]Building a Biometric National ID: Lessons for Developing Countries from India’s Universal ID Program”, Alan Gelb and Julia Clark, The Center for Global Development, October 2012, http://www.cgdev.org/doc/full_text/GelbClarkUID/1426583.html
[x] http://www.aph.gov.au/Parliamentary_Business/Committees/Senate/Legal_and_Constitutional_Affairs/Biometrics/Rep ort/c02