Monday, April 30, 2012

National Level Biometric Deployment Trees & Forest

Though it's not my intention, I know it can seem like "All India, All the Time" around here. That's because India has set for itself the most ambitious ID management in human history, and therefore, India is providing more real world examples of the challenges inherent in very large scale biometric deployments than anywhere else. But India isn't necessarily unique in terms of the problems it is attempting to address with biometrics. All the BRICS (Brazil, Russia, India, China, South Africa), for example, face very similar challenges.

Forest and trees:
The information upon which blogs rely (the news articles below) provide great descriptions of individual trees. The four articles from the Indian press below can be seen as individual trees. The four sections in bold below the articles describe the species to which each different tree belongs.

The biometrics forest is made up of these, and more, species.

Examples:
I. Can Technology Fix India? (Dawn.com)
The dreams of modern India rarely make it to Rayagada. The Indians of these eastern forests forage for sago leaves and wild mango to survive. Barely a third can sign their names. Most live without electricity. Many have joined a Maoist insurgency fighting to overthrow the system.

Now, modernity is creeping in. Smart cards, fingerprint scanners and biometric identity software are transforming Rayagada into a laboratory to test a thesis with deep implications for the future of India: Can technology fix a nation?
Please click through to look at the pictures (and don't miss the comments).

As readers of this blog know, the short answer to the headline question is: No. Technology cannot fix India. ID management is about people. Biometrics can help (immensely!) with identity management challenges, but they cannot manage anything all by themselves.

II. Aadhaar card fraud exposes registration process errors (Times of India)
"The fraud proves that in a 100% Aadhaar-enrolled city like Hyderabad, 30,000 cards have been registered in a part of the city by exploiting those loopholes.
If:
♦ The Hyderabad population is 4,010,238
♦ The fraud number is 30,000 (article)
♦ 100% of Hyderabad is enrolled (article)

Then the rate of fraud is 0.0074 — less than 1%.

III. Identification of slum residents suspended (The Hindu)
The Chennai Corporation has suspended biometric identification of thousands of slum residents along 16 canals in the city, following stiff opposition from AIADMK councillors. The civic body had begun biometric identification of residents on slums near the Mambalam canal a few weeks ago, in order to issue smart cards, but AIADMK councillors opposed it, alleging that the list prepared by the previous DMK-led council had failed to include many actual beneficiaries. Once the biometric cards were issued, the residents of these slums were to be resettled.
I'm not even sure exactly what this is about but I have no doubt it's of significant local importance (I think it's about accounting for people who reside in areas we in the West would describe as "not Zoned Residential" so that they might be humanely resettled or legitimized as residents — such as when squatters are given title to their abodes).

IV. Persons Throng Aadhar Enrolment (Deccan Herald)


These four specifics are open to a more general interpretation.

I. The scale of the problems better ID management can help address
In the simplest terms, biometrics can really help people reduce the amount and severity of human suffering in their midst. What nobler aim can there be? Biometric Identity Management is about people.

II. Perfect is the enemy of Good
The proper metric for measuring biometric deployments is Return on Investment, not perfection. If half of the funds devoted to a current welfare program are stolen and a proposed biometric ID management system will be defrauded 10% of the time, what should you do?

III. Biometric technology is extremely flexible
Biometrics give talented and imaginative managers a tool that can be applied more problems than any one person can possibly comprehend.

IV: People want it.



Many (but obviously not all) of the posts here fall into one of the four categories above. India just happens to be providing a whole lot of examples.