Let’s ask the professors UIDAI cited in its latest report: Do you agree with UIDAI’s assessment of Aadhaar? Do you share their confidence in the project? Did UIDAI ask you in advance, before using your name for their marketing purposes?
The Unique Identification Authority of India (UIDAI) have been accused of making false claims about the reliability of the biometrics that its unique identification number (UID) or Aadhaar scheme relies on. The report released earlier this week by UIDAI is in response to those criticisms.
UIDAI say that “… based on the analysis, it can be stated with confidence that UIDAI enrolment system has proven to be reliable, accurate and scalable to meet the nation’s need of providing unique Aadhaar numbers to the entire population. It is now safe to conclude that the system will be able to scale to handle the entire population”. But that is mere assertion, it begs the question, they would say that, wouldn't they.
They need independent and respected biometrics experts to agree with them, if this report is to boost confidence in UIDAI’s abilities. They mention several names. The casual reader may assume that these named experts all agree with UIDAI’s conclusion that Aadhaar will work. It would be instructive to ring them up and ask them directly for their opinion.
Does Professor John Daugman, for example, agree with UIDAI when they say that “… although [the false positive identification rate of 0.057%] is expected to grow as the database size increases, it is not expected to exceed manageable values even at full enrolment of 120 crores”? It seems unlikely—Professor Daugman is the man who first pointed out that any attempt to prove uniqueness in a large population of biometrics must drown in a sea of false positives, please see
And does Professor Jim Wayman, for example, agree with UIDAI when they say that “… based on the [receiver operating characteristic] model, the UIDAI expects the accuracy of the system to remain within the same order of magnitude as reported above. Hence it can be stated that system will be able to scale to handle the entire population without significant drop in accuracy”? It seems unlikely—Professor Wayman is the lead author of a paper which concludes that biometrics is a discipline out of statistical control, the results gathered so far tell you nothing about what to expect in future, please see
http://biometrics.nist.gov/cs_links/ibpc2010/pdfs/FundamentalIssues_
Final.pdf
If the two professors agree with UIDAI and renounce their earlier statements, well and good.
But if, on the other hand, they say that they have no reason to believe that UIDAI is right, they have not had a chance to assess the evidence that UIDAI claims to have, they do not understand why UIDAI has mentioned their names, then this schoolboy attempt to justify UIDAI’s waste of public money will fall humiliatingly flat on its face.
(David Moss spent eight years campaigning against the UK’s National ID (NID) card scheme, which was finally scrapped by the British government. Mr Moss is an MA in Philosophy from Cambridge University, MSc in Software Engineering from Kingston. With a career spanning of over 35 years, Mr Moss at present works as director at Business Consultancy Services Ltd and can be contacted at [email protected].)
You may want to read...
Inside story of the National Stock Exchange’s amazing success, leading to hubris, regulatory capture and algo scam
Fiercely independent and pro-consumer information on personal finance.
1-year online access to the magazine articles published during the subscription period.
Access is given for all articles published during the week (starting Monday) your subscription starts. For example, if you subscribe on Wednesday, you will have access to articles uploaded from Monday of that week.
This means access to other articles (outside the subscription period) are not included.
Articles outside the subscription period can be bought separately for a small price per article.
Fiercely independent and pro-consumer information on personal finance.
30-day online access to the magazine articles published during the subscription period.
Access is given for all articles published during the week (starting Monday) your subscription starts. For example, if you subscribe on Wednesday, you will have access to articles uploaded from Monday of that week.
This means access to other articles (outside the subscription period) are not included.
Articles outside the subscription period can be bought separately for a small price per article.
Fiercely independent and pro-consumer information on personal finance.
Complete access to Moneylife archives since inception ( till the date of your subscription )
If UID is meant for ensuring that welfare benefits reach the poor, beneficiaries, why is UIDAI enrolling those ineligible for such subsidies? How would the database be used for the poor? Will authentication be done at the 600,000 ration shops at delivery? If not how would UID be used to ensure the poor receive the rations? UIDAI and the advocates of the UID would never answer these and many such questions. Why waste time on technology, if issues of use and implementation are not included in the debate?
Perfect tautology.
Facts:
1. UIDAI results are the same across PoC, 100M DB and are consistent with other large accepted NIST like results.
2. UIDAI has presented results in over 6 papers in the the most prestigious biometric conferences.
3. UIDAI hired world's top most biometric experts in design and testing.
4. Two good multimodal biometric will generate 100 to 10,000 better results than single mode biometric. You will find this in any introductory biometric textbook.
Nothing is out of ordinary except India is able to execute a project efficiently, within schedule and in reasonable budget.
Mr. Moss, if you want to check out facts, please attend International Biometric Conference in Delhi in March 2012. You can check your statistics with 200 biometric experts including Daugman and Wayman.
Good to hear from you again, http://www.moneylife.in/article/how-uida...
I went to Biometrics 2009, a three-day biometrics conference and exhibition here in the UK and facts were hard to find, http://forum.no2id.net/viewtopic.php?t=2...
It was a groupthink event. Sealed off from the outside world, all the delegates thought that everything was rosy in the biometrics garden, and each one reinforced the other’s unfounded confidence. So, no thank you, I shan’t be in Delhi next month.
Re your facts, the FPIR we were all pointed at in the PoC report was 0.0025%. Now, in the enrolment report, it’s 0.057%, 22.8 times worse. Consistent?
Aadhaar is sui generis. UIDAI won’t learn from the experts and the textbooks, it’s the other way round.
And no-one will learn from the poorly-conceived and abysmally presented papers produced by UIDAI, all way below the standard demanded of respectable academic authorities. They inspire no confidence.
Your “Critics Argument” couldn’t be more wrong. If you are interested in investing in BAT shares, you don’t do so just because a BAT director tells you how marvellous the company is. You check the audited report and accounts first. That’s not to say that the BAT director is a liar. It’s just dutiful prudence.
Aadhaar is an investment. A big one, around Rs. 72,000 crore according to http://www.thehindu.com/opinion/lead/art...
Your money, Ram, is being invested in the project. Are you happy that that investment of your money is prudent? Do you think Aadhaar will curtail corruption? Or automate corruption?
Do you seriously think the banks will accept nothing but Aadhaar biometrics to authenticate payments? That’s the acid test. Maybe they will, but the history of uninterrupted failure of mass consumer biometrics is against it happening.
I choose BAT as an example because, back in 1980, I did some work at BAT (UK & Export) Ltd, here in the UK. They were converting their computer systems from Burroughs Medium Systems to something else and they used an Indian company to re-write the software. Their IT director told me he was staggered. Not only did the Indians do the conversion job, but they corrected all the errors in the system that BAT knew about and many they didn’t and they supplied an unparalleled set of test data.
That has coloured my view of India ever since and you’re wrong if you think I am trying in any way to denigrate Indian abilities.
For that, for a slab of high-handed Indian denigration, de haut en bas, we know best, take a look at The Economist magazine here http://www.dmossesq.com/2012/01/economis... and here http://www.dmossesq.com/2012/02/economis...
Or take a look at Morpho. You think, and I think, that Aadhaar is an Indian achievement. Morpho don’t. They think it’s a French achievement: “Morpho is in charge of all technological aspects of Aadhaar”, they say, http://www.morpho.com/references/identit... Maybe someone should tell Mr Nilekani he needn’t come into the office, he’s not needed, not now Morpho are here.
QUOTE
From: Planet Biometrics Editorial Team [mailto:[email protected]]
Sent: 27 January 2012 11:57
To: [email protected]
Subject: Planet Biometrics e-newsletter
Welcome to the Planet Biometrics e-newsletter
Dear Colleagues
Welcome to our next Planet Biometrics e-newsletter.
This week the Indian UIDAI released an incredibly positive report into the performance of biometrics at an unprecedented scale. Bearing in mind the number of naysayers that have continually doubted the performance of biometric technology and who predicted that India would "drown in a sea of false positives" [1], our industry should be waving thi9s report from the treetops.
Next week we will be running a full-length feature article on the findings of the report and the workings of the project.
****************************
Why not forward this newsletter on to a colleague so that they can also subscribe by clicking here [2]
Best regards
Mark Lockie
Editor
UNQUOTE
It's just a guess, but I bet that there will be no full-length feature article on the findings of the report and the workings of the project next week, or any other week. Mr Lockie's masters must be furious at and frightened by UIDAI's ill-considered report.
UIDAI faced an existential threat. They had to do something, to survive. So they produced this report, 'Role of Biometric Technology in Aadhaar Enrollment' [3], to demonstrate that their biometrics are reliable and that there is therefore a point to their existence.
Unfortunately, UIDAI have over-egged the pudding. And gilded the lily. Their ploy will be written up in the business school manuals for years to come in the how-not-to-do-it section, what-seemed-like-a-good-idea-at-the-time-blows-up-in-your-face category.
Look at it from the biometrics suppliers' point of view. UIDAI have claimed astonishing reliability for their biometrics, a reliability that has never been seen before and which no respectable expert has ever predicted before – "incredible" and "unprecedented", as Mark says. What are Morpho [4], for example, supposed to say now, when potential customers approach them saying that they, too, want a biometric failure to enrol rate of 0.14 percent and a false positive identification rate of 0.057 percent and a false negative identification rate of 0.035 percent?
Morpho and the others have eschewed the publication of performance figures for years. They look like warranties. And why give warranties if you don't have to?
Existing customers with worse performance figures will want to know why they haven't achieved the same reliability as UIDAI claim to have done. Again, what do Morpho say to them? "We sold you a defective product compared with UIDAI's"? Or maybe "UIDAI just made the numbers up, it's all politics, don't worry their performance is actually just as bad as yours"?
Morpho [5] must wish UIDAI hadn't published that report.
What are the banks going to say? "UIDAI's performance is so good, apparently, that we ought to be using the same technology to authenticate payments all over the world, not just in India. Still, better check first, can we see the independent audit of UIDAI's figures, please? Purely a formality but, you know, have to tick all the boxes ..."?
No, the attempt will be made to bury that UIDAI report as quickly and as thoroughly as possible (save your copy now [3]). And Mark Lockie will be asked/told to keep quiet. "No full-length feature articles please, Mark, as you always say, you're 'not technical', just forget it. Now. That's an order."
----------
1. Possibly a reference to David Moss, naysayer and author of 'India's ID card scheme – drowning in a sea of false positives', http://dematerialisedid.com/BCSL/Drown.h...
2. http://www.planetbiometrics.com/user/reg...
3. http://uidai.gov.in/images/FrontPageUpda...
4. http://dematerialisedid.com/BCSL/Garlic....
5. "Morpho is in charge of all technological aspects of Aadhaar ...", http://www.morpho.com/references/identit...
Every one of his articles on UIDAI has now been discredited, and yet he asks some of the most eminent professors guiding the UIDAI project (Daugman/Wayman) to reply. Not likely!
As for drowning in a sea of false positives, even a fool could see that Prof. Daugman made his comments warning that this could happen without iris. Our Indian project uses iris...
Moss should grow a backbone and admit he is wrong. Even in the face of facts he blindly refuses to accept that this multi-modal, multi-supplier project is delivering impressive results. Moss is happy to quote reports inaccurately when he feels like it. But disregards them as rubbish when they contradict his anti-biometric bile.
Mr Moss, you are, Sir, a Luddite in the extreme. Feel free to talk to yourself some more...
Posted anonymously for obvious reasons...
Not good enough. This is politics. Samir Kelekar clearly understands that. So does Indian. So does Mathew. Why doesn't Observer?
Because he isn't observant. Observer misses the promise already made, in the response to Dingy, willingly to apologise if and when there is a reason to do so.
In the meantime, the Prime Minister has given UIDAI and NPR equal shares of the Indian population, 60 crore each. Is the technology used by the two organisations equal? We don't know.
Instead of making a technology decision, a political decision has been made, a decision UIDAI have lobbied for worldwide, not least through The Economist magazine, whom UIDAI seem to have successfully suborned, http://www.dmossesq.com/2012/01/economis...
Next comes the financial decision. Will Parliament vote the money needed for Aadhaar's next phase? Perhaps the money men will be more observant. This is public money after all, and public money should not be wasted.
NREGA uses biometrics.
UIDAI uses biometrics.
UIDAI claims to have achieved a biometric failure to enrol rate (B-FTE) of 0.14 percent and a false positive identification rate (FPIR) of 0.057 percent and a false negative identification rate (FNIR) of 0.035 percent.
What are the equivalent figures for NPR and NREGA?
Lovely to hear from you.
UIDAI faced an existential threat. They had to do something, to survive.
They elected to produce a report claiming that their use of biometrics in Aadhaar is reliable.
They may have slightly over-egged the pudding or gilded the lily – rather than mere reliability, their report claims near-perfection.
Not only has a biometric failure to enrol rate of 0.14%, a false positive identification rate of 0.057% and a false negative identification rate of 0.035% never been simultaneously claimed before, but no respectable expert has ever suggested that it is achievable.
One must therefore look at these figures with some circumspection.
The biometrics suppliers such as Morpho must be having kittens. They don't like performance figures being published – they look like warranties, and that's the last thing they want to give.
The banks must be scratching their heads. If this sort of reliability is achievable, how come they aren't authenticating payments with biometrics all over the world?
By all accounts, UIDAI's ploy has worked and the Prime Minister has given the go-ahead for a further 400 million enrolments. Well done, UIDAI.
But now there's a price to pay. More and more people are going to want to see those figures audited.
Because if they are correct, Dingy, not only will I apologise but the world will change, this will be a revolution.
Conversely, if they turn out to be a cynical trick to stay in business, UIDAI are going to have to run fast, ahead of 1.2 billion Indians who want their money back and there may not be time for them to apologise first.
What do you think, Dingy? Which way will it go?
Best wishes
dm
But UIDAI has proved that, all that talk was non sense and they have successfully created 12 crore numbers. Mr Moss has to apologize for his past arrogance.
Indeed, UIDAI is also silent about how easily fingerprints can be faked. A criminal waste of money aka 2G scam. MMS doesnt seem to learn anything from past mistakes.
Indeed, the pathetic level of this govt. couldnt have been better revealed than this.
It is such decisions without any thinking that allow people such as Kasab to come and do what they want in this country. And who is going to pay for this collosal waste of public money? It is time to show the door to the Congress in this election in the states. I am out campaigning against Congress in my home state Goa.
samir