Counting Down in 2011
by Vickram Crishna - an IIT Delhi Alumnus
For the first time since the year dot, the National Census is being carried out in conjunction with the creation of the National Population Register. The reason for this is simply that government policy today aims at tailoring subsidies and social benefits to those who need it the most. Unfortunately, nobody, meaning the government, knows who those people are. There are many reasons for that (most of which could have been addressed in the PM's recent press 'conference'), but let it be, this is 2011 and nobody expects miracles overnight, or even after sixty years. We should continue to ignore the fact that those people certainly know who they are.
However, and people who care about freedom should ponder this, joining the two exercises (the only possible reason being to save the cost of duplicate exercises, about which more later) is contrary to the very concept of the census, which guarantees complete privacy to Indians giving their personal information to the census-takers. Now, while 'preserving' privacy and the Constitution, the same census-takers also work for the NPR, which does not oblige any privacy at all.
Aside from personal information, as a secular country we decided many years back we would stop dividing people (even in as minor an aspect as a headcount) by religion or caste. Since this is contrary to the purpose of an NPR, the gaping chasm has been papered over by combining (not merging, they are supposed to be two separate forms, but since the same person fills in both forms, and is rather overworked and totally unpaid - go to jail if you don't agree to 'volunteer'...!). More on how actual census-takers are filling in forms below. Also take a look at this report on how the UK census operation, which has been bloated up from 41 to 56 questions from the previous census a decade back, is faltering under the welter of questions.
The 'sudden' appearance of UID in this melange happened sometime last year. Like it or not, convenient or not (and it is certainly inconvenient to be badgered time and again to give ostensibly the same information to the same people), the UID is as of the present moment a completely undemocratic (and possibly illegal, but certainly unethical) exercise, because it is conducted extra-Constitutionally, without any of the debate and discussion one might, in an ideal world, expect taking place in a democratic country. Examples of countries that have got by without debate and discussion until recent times are Tunisia, Yemen and now Libya. The Turkish Prime Minister, Recep Erdoğan, said recently that it is a mistake for countries to ignore the voices of the people, in respect of freedom and democracy, but since he wasn't addressing India, we shall continue to ignore such irrelevant advice.
Mr Nilekani has been appointed, with Cabinet rank, as a functionary of the Planning Commission (which has no role to play in the Cabinet, being by design at an arm's length from the executive). The Planning Commission is not supposed to be prey to the kind of pressures party politics brings in its wake, but in this case, the lines have been blurred.
There is a Bill pending to legitimise the UID implementation body, today called the UIDAI (authority of India), which will get renamed as the National Identity Authority of India, whose provisions make the US Patriot Act look like a social security scheme (and even Americans are scarcely aware that it was once illegal for anyone to ask for their own Social Security Number, except specifically in connection with the payment of Social Security benefits).
Nilekani himself has been at pains to clarify that the UID database he proposes to set up will not hold any personal information beyond name and number and some characterising biometrics, but between you and me, he is being no more straightforward about the role UID will play in India, than Mr Singh has been about stopping his fellow politicians from looting the country.
To put it plainly, UID will be used a data-connecting glue between databases that already exist, such as bank accounts, driving licenses, passports, ration cards and more, a total of some 21 databases. These contain everything about people in India who aspire to the modern life (as against eating berries in the forest). Today, it is well-nigh impossible, unless you are a criminal or an official high-ranking enough to subvert the checks and balances in place, to get at those databases and link up all the information about anyone.
Post UID, the world's top hackers will be camping (figuratively) in India, racing to find how many seconds it will take to breach Mr Nilekani's as yet untested security. It is untested, because no-one has ever created a database with one billion plus names, together with assorted bits and pieces of personal information, and tried to make it deliver a response to a query in microseconds. It has to be fast, because the reply (only a 'yes' or a 'no', in response to the equivalent of a question, "Does this [biometric] match this [person]?" must be given within seconds, after travelling across the country via any number of links - there may be a long line of people standing there waiting for their chance. A mammoth task, and one worthy of a global level technological challenge.
Unfortunately (from my PoV) this challenge is akin to climbing up a mountain of soft cowpats barefoot. Difficult, no doubt, a challenge, for sure, but worth it? Only Mr Nilekani and his camp followers seem to think so, while other democratic countries where this sort of idea has been floated before, such as the USA (REAL ID), Australia (Australia Card) and the UK (National Identity Card), have resoundingly turned them down, due to citizen protests.
The most recent, in terms of resolution, has been the UK. It cancelled its Identity Card plans last year, despite spending a whopping stg 850mn or thereabouts over 8 years, due to overwhelming concerns of privacy. The US faces this question over and over again, but consistently refuses to bow down to government pressure. [This is quite amazing, really, because it is their same house of representatives that endorsed the 'temporary' continuance of the Draconian Patriot Act once again, a few days ago]. Australia backed off on the Ozcard (they wanted to brand it the Australia Card, but it was more derisively known as the Auschtralia Card, with a nod to Hitler's infamous concentration camp) about 20 years ago, and no political party there is likely to touch this hot potato again.
What drives Nilekani and his backers in the government? And who are they anyway?
The latter would seem to be the Prime Minister himself, together with MS Ahluwalia, the affable deputy chairman of the PC. They also have the backing of the BJP, which mooted the UID as a card post Kargil, 1999, that would be distributed in border areas and prevent infiltration (Huh? Hey, it isn't my idea, if you want to know more, talk with LK Advani). Had we only listened to him in 1999, we might have all had cards, which we could have thrown at the terrorists wielding automatic rifles, had we too, been commuting via VT, or tippling at the Taj, or ogling lovely lassies at the Oberoi, on 26/11, as the 300+ people killed by foreign terrorists who, unaccountably, did not apply for entry permits when they came here.
Said terrorists came ashore by boat, unlike at Kargil, so there is a smidgeon of a chance that Mr Advani may have changed his mind about the effectiveness of a UID card as a border control mechanism. Actually, no such card is proposed even today by the UIDAI, but Nilekani has made sure that if 'others' want to issue a card, he has no objection. Since this card might also cost a bit of cash, we may remain a nation of haves and have-nots, but with respect to the 'new' need now to prove our Aadhaar.
Arraigned against them is Congress president Sonia Gandhi, who with the National Advisory Council, is bitterly against the idea of targeted subsidies. And why not, indeed? One might be tempted to believe that any modern country would be ashamed of the majority of its people starving, while foodgrains are exported, earning GDP raising forex, or rot in the open, whipping up a storm in a teacup (that being about how much public attention is being paid to the foodgrains still rotting today in badly constructed or non-existent godowns).
Yet our economics-centric 'clean' prime minister would have us believe that it is better to help a few of the most badly affected (because they are a 'burden on the economy') and to blame the rest for the non-delivery of even the subsidies that are 'targeted' nowadays. The 'blame' aspect comes from the fact that identification as a solution only applies in the case where false identities are being used to scam off most of the missing money, ie money that missed its targeted destination, but even a cursory reading of the published information on the subject indicates this is rubbish.
Further, he wants (or at least he does not object) to spend billions of dollars (a currency with which Mr Nilekani may be more familiar than our indigenous rupee) to establish the foundation of a (yet uncosted) national verification network, one that does not exist today, but will be built (or is planned to be built) on a connectivity solution conceived by Sam Pitroda, who has, as everyone probably knows, bounced back from the wilderness [where he was sent, pretty unfairly, after the fall of the Rajiv Gandhi government in the late '80s].
There are two types of subsidies, cash and kind. Cash is paid in return for work done, the world's largest scheme of its kind. After huge hiccups, cash is now paid to beneficiaries via bank accounts, and names of beneficiaries are nowadays posted publicly on blackboards, to ensure transparency. This does not prevent leaching of public subsidy schemes by middlemen (including panchayat heads, in some reported cases, and, according to a Bombay High Court order reported in the Asian Age of Feb 24, 2011, fake identity cards produced en masse by ministers, police and district officials), nor does it help make for better infrastructure (many of the actual work schemes are either badly designed or completely useless) but it does cut down on the invention of non-existent work and non-existent beneficiaries.
Granted, UID would also prevent the latter, but since this issue is already being dealt with, it is hard to see how any value addition is proposed. Certainly, all discussion of value-add are up in the air, as no feasibility study has ever been done to determine the effectivity of a nation-wide unique numbering scheme.
Beyond the cost, will it deliver its supposed benefits? The people who are supposed to deal with poverty alleviation certainly don't think so (see page 35 of pdf report - MHUPA Strategy Report) . Security hawks possibly still do, but then we are talking about a lot of money, to be spent endlessly into the future.
Will it even succeed in assigning every Indian resident an unique identity number (the UK government decided this was an unreachable goal, for a far smaller and better educated population, while exposing their citizens to an unacceptable risk of permanent breach of privacy)?
The enrollment process is critical. Laughably, instead of doing away with existing documents (riddled with problems, hence the perceived need for a new identifier), the formal enrollment process actually demands them. For people who don't have any of them, an 'introducer' will fill the gap. Witnesses have seen such 'personal verifiers' introducing 60 and 70 people at a time, whilst admitting to outside observers they haven't a clue about these 'identified' folks. They have also seen that if the fingerprinting system 'fails' (ie, does not produce a recordable scan in four tries), some enrollers calmly move on to the next step, 'capturing' the irreproducible fourth scan.
Unendorsed by democratic consensus, complete lack of clarity about feasibility, and plagued by procedural flaws: this is the world's greatest IT project at work.
Vickram
Fool On The Hill
For the first time since the year dot, the National Census is being carried out in conjunction with the creation of the National Population Register. The reason for this is simply that government policy today aims at tailoring subsidies and social benefits to those who need it the most. Unfortunately, nobody, meaning the government, knows who those people are. There are many reasons for that (most of which could have been addressed in the PM's recent press 'conference'), but let it be, this is 2011 and nobody expects miracles overnight, or even after sixty years. We should continue to ignore the fact that those people certainly know who they are.
However, and people who care about freedom should ponder this, joining the two exercises (the only possible reason being to save the cost of duplicate exercises, about which more later) is contrary to the very concept of the census, which guarantees complete privacy to Indians giving their personal information to the census-takers. Now, while 'preserving' privacy and the Constitution, the same census-takers also work for the NPR, which does not oblige any privacy at all.
Aside from personal information, as a secular country we decided many years back we would stop dividing people (even in as minor an aspect as a headcount) by religion or caste. Since this is contrary to the purpose of an NPR, the gaping chasm has been papered over by combining (not merging, they are supposed to be two separate forms, but since the same person fills in both forms, and is rather overworked and totally unpaid - go to jail if you don't agree to 'volunteer'...!). More on how actual census-takers are filling in forms below. Also take a look at this report on how the UK census operation, which has been bloated up from 41 to 56 questions from the previous census a decade back, is faltering under the welter of questions.
The 'sudden' appearance of UID in this melange happened sometime last year. Like it or not, convenient or not (and it is certainly inconvenient to be badgered time and again to give ostensibly the same information to the same people), the UID is as of the present moment a completely undemocratic (and possibly illegal, but certainly unethical) exercise, because it is conducted extra-Constitutionally, without any of the debate and discussion one might, in an ideal world, expect taking place in a democratic country. Examples of countries that have got by without debate and discussion until recent times are Tunisia, Yemen and now Libya. The Turkish Prime Minister, Recep Erdoğan, said recently that it is a mistake for countries to ignore the voices of the people, in respect of freedom and democracy, but since he wasn't addressing India, we shall continue to ignore such irrelevant advice.
Mr Nilekani has been appointed, with Cabinet rank, as a functionary of the Planning Commission (which has no role to play in the Cabinet, being by design at an arm's length from the executive). The Planning Commission is not supposed to be prey to the kind of pressures party politics brings in its wake, but in this case, the lines have been blurred.
There is a Bill pending to legitimise the UID implementation body, today called the UIDAI (authority of India), which will get renamed as the National Identity Authority of India, whose provisions make the US Patriot Act look like a social security scheme (and even Americans are scarcely aware that it was once illegal for anyone to ask for their own Social Security Number, except specifically in connection with the payment of Social Security benefits).
Nilekani himself has been at pains to clarify that the UID database he proposes to set up will not hold any personal information beyond name and number and some characterising biometrics, but between you and me, he is being no more straightforward about the role UID will play in India, than Mr Singh has been about stopping his fellow politicians from looting the country.
To put it plainly, UID will be used a data-connecting glue between databases that already exist, such as bank accounts, driving licenses, passports, ration cards and more, a total of some 21 databases. These contain everything about people in India who aspire to the modern life (as against eating berries in the forest). Today, it is well-nigh impossible, unless you are a criminal or an official high-ranking enough to subvert the checks and balances in place, to get at those databases and link up all the information about anyone.
Post UID, the world's top hackers will be camping (figuratively) in India, racing to find how many seconds it will take to breach Mr Nilekani's as yet untested security. It is untested, because no-one has ever created a database with one billion plus names, together with assorted bits and pieces of personal information, and tried to make it deliver a response to a query in microseconds. It has to be fast, because the reply (only a 'yes' or a 'no', in response to the equivalent of a question, "Does this [biometric] match this [person]?" must be given within seconds, after travelling across the country via any number of links - there may be a long line of people standing there waiting for their chance. A mammoth task, and one worthy of a global level technological challenge.
Unfortunately (from my PoV) this challenge is akin to climbing up a mountain of soft cowpats barefoot. Difficult, no doubt, a challenge, for sure, but worth it? Only Mr Nilekani and his camp followers seem to think so, while other democratic countries where this sort of idea has been floated before, such as the USA (REAL ID), Australia (Australia Card) and the UK (National Identity Card), have resoundingly turned them down, due to citizen protests.
The most recent, in terms of resolution, has been the UK. It cancelled its Identity Card plans last year, despite spending a whopping stg 850mn or thereabouts over 8 years, due to overwhelming concerns of privacy. The US faces this question over and over again, but consistently refuses to bow down to government pressure. [This is quite amazing, really, because it is their same house of representatives that endorsed the 'temporary' continuance of the Draconian Patriot Act once again, a few days ago]. Australia backed off on the Ozcard (they wanted to brand it the Australia Card, but it was more derisively known as the Auschtralia Card, with a nod to Hitler's infamous concentration camp) about 20 years ago, and no political party there is likely to touch this hot potato again.
What drives Nilekani and his backers in the government? And who are they anyway?
The latter would seem to be the Prime Minister himself, together with MS Ahluwalia, the affable deputy chairman of the PC. They also have the backing of the BJP, which mooted the UID as a card post Kargil, 1999, that would be distributed in border areas and prevent infiltration (Huh? Hey, it isn't my idea, if you want to know more, talk with LK Advani). Had we only listened to him in 1999, we might have all had cards, which we could have thrown at the terrorists wielding automatic rifles, had we too, been commuting via VT, or tippling at the Taj, or ogling lovely lassies at the Oberoi, on 26/11, as the 300+ people killed by foreign terrorists who, unaccountably, did not apply for entry permits when they came here.
Said terrorists came ashore by boat, unlike at Kargil, so there is a smidgeon of a chance that Mr Advani may have changed his mind about the effectiveness of a UID card as a border control mechanism. Actually, no such card is proposed even today by the UIDAI, but Nilekani has made sure that if 'others' want to issue a card, he has no objection. Since this card might also cost a bit of cash, we may remain a nation of haves and have-nots, but with respect to the 'new' need now to prove our Aadhaar.
Arraigned against them is Congress president Sonia Gandhi, who with the National Advisory Council, is bitterly against the idea of targeted subsidies. And why not, indeed? One might be tempted to believe that any modern country would be ashamed of the majority of its people starving, while foodgrains are exported, earning GDP raising forex, or rot in the open, whipping up a storm in a teacup (that being about how much public attention is being paid to the foodgrains still rotting today in badly constructed or non-existent godowns).
Yet our economics-centric 'clean' prime minister would have us believe that it is better to help a few of the most badly affected (because they are a 'burden on the economy') and to blame the rest for the non-delivery of even the subsidies that are 'targeted' nowadays. The 'blame' aspect comes from the fact that identification as a solution only applies in the case where false identities are being used to scam off most of the missing money, ie money that missed its targeted destination, but even a cursory reading of the published information on the subject indicates this is rubbish.
Further, he wants (or at least he does not object) to spend billions of dollars (a currency with which Mr Nilekani may be more familiar than our indigenous rupee) to establish the foundation of a (yet uncosted) national verification network, one that does not exist today, but will be built (or is planned to be built) on a connectivity solution conceived by Sam Pitroda, who has, as everyone probably knows, bounced back from the wilderness [where he was sent, pretty unfairly, after the fall of the Rajiv Gandhi government in the late '80s].
There are two types of subsidies, cash and kind. Cash is paid in return for work done, the world's largest scheme of its kind. After huge hiccups, cash is now paid to beneficiaries via bank accounts, and names of beneficiaries are nowadays posted publicly on blackboards, to ensure transparency. This does not prevent leaching of public subsidy schemes by middlemen (including panchayat heads, in some reported cases, and, according to a Bombay High Court order reported in the Asian Age of Feb 24, 2011, fake identity cards produced en masse by ministers, police and district officials), nor does it help make for better infrastructure (many of the actual work schemes are either badly designed or completely useless) but it does cut down on the invention of non-existent work and non-existent beneficiaries.
Granted, UID would also prevent the latter, but since this issue is already being dealt with, it is hard to see how any value addition is proposed. Certainly, all discussion of value-add are up in the air, as no feasibility study has ever been done to determine the effectivity of a nation-wide unique numbering scheme.
Beyond the cost, will it deliver its supposed benefits? The people who are supposed to deal with poverty alleviation certainly don't think so (see page 35 of pdf report - MHUPA Strategy Report) . Security hawks possibly still do, but then we are talking about a lot of money, to be spent endlessly into the future.
Will it even succeed in assigning every Indian resident an unique identity number (the UK government decided this was an unreachable goal, for a far smaller and better educated population, while exposing their citizens to an unacceptable risk of permanent breach of privacy)?
The enrollment process is critical. Laughably, instead of doing away with existing documents (riddled with problems, hence the perceived need for a new identifier), the formal enrollment process actually demands them. For people who don't have any of them, an 'introducer' will fill the gap. Witnesses have seen such 'personal verifiers' introducing 60 and 70 people at a time, whilst admitting to outside observers they haven't a clue about these 'identified' folks. They have also seen that if the fingerprinting system 'fails' (ie, does not produce a recordable scan in four tries), some enrollers calmly move on to the next step, 'capturing' the irreproducible fourth scan.
Unendorsed by democratic consensus, complete lack of clarity about feasibility, and plagued by procedural flaws: this is the world's greatest IT project at work.
Vickram
Fool On The Hill
