MAY 12, 2011 · 1:02 PM
The other day, on a list to which I subscribe, I found this light-hearted observation about a very serious subject – hape.
Hape is not, as the Quick Reader may imagine, the last of the items (the others, like hape, were also evil) to emerge from Pandora’s box, in the Greek myth.
All round the world, business content and business processes are being digitised and made available to stakeholders, often in a highly restricted manner, ie on some kind of subscription basis. Subscribers do so, often paying money, in the faith that they get, in return, some sort of exclusive or protected access.
Quite often, in order to do so, they also voluntarily make available some kind of personal information (it varies, of course, depending on the need).The people who put together the service then store this digitised information and make it available as a part of their subscriber verification.
It works like this: if you are who you claim to be, then you should know [this] about you, because you gave us that information. If you don’t know [this], it isn’t you, so please (the please is optional, and mostly absent, I have observed) go away.
That’s all very well.
What underlies all of this is a tremendous amount of faith, faith in the quality of business processes used by the people who manage that service, that they will keep their part of the bargain – which is sometimes expressly stated, and sometimes implicit – to hold the information securely. And, a lot of the time, that is exactly what happens.
Until it hapens.
Hape is a word coined by my fellow listmember, Dinesh Bareja, to describe what happens when that trust is belied. To quote from his entertaining post:
“The Theory of Hape (abridged):
Every system or technology environment is built with known or unknown holes all over waiting to be penetrated and exploited.
After a hape, weak controls and dirty data are exposed to the world, and management [people] have to run around trying to save their reputation, jobs and more.”
These words somehow reminded me of India’s shining hope, the UIDAI, a quasi-legal organisation foisted on the country by the Planning Commission. The latter, an august body is not known for its cowboy antics, has quixotically chosen the path less often trod, that of mandating a special purpose vehicle, an Authority of India, with the task of ensuring that every resident of India gets a unique identification, a number that cannot possibly be allotted to anyone else, a passkey to all manner of delights, fancifully branded Aadhaar (foundation).
Of course, to work, whoever is charged with delivery of a particular service must be sure that the person quoting the number is actually who she says she is. Or he is, if you object to commonly accepted gender-free phraseology.
Enter the verifier. In its wisdom, UIDAI has decided that this shall take the form of biometric markers – fingerprints will do the job.
Or not. Turns out that fingerprints, the stuff of crime novels for well over a century, are well left there, in works of fiction.
Fingerprints have some problems: 1. they are not immutable, they can change with time, depending on the kind of work the person engages in, and also the state of health; 2. they may not be unique (no study of very large populations has ever been conducted, so the belief in fingerprinting is no more real than a belief in the Flying Spaghetti Monster – or no less real, to be sure); 3. fingerprint recording machines are not very pragmatic for countries like India (shaky electricity, poor hygiene, poor housekeeping); 4. digital fingerprints are based on algorithms that have never been applied to very large populations, so they may be even less perfectly unique than the fingerprint patterns themselves; and so on.
Enter the iris. This central part of the human eye turns out to be even more uniquely patterned than the fingerprint, and luckily, its digital version is also more perfectly matched to the real thing, than in the case of fingerprints. Except. This has also not been researched and scientifically established.
Somewhere in all this hype, one little factor seems to be missing – the act of verification. Each nodal point where it is needed will have to be equipped with a biometric scanner and Go/No-Go display device that will need to communicate very fast (oh, very fast indeed) with a digital store to immaculately match those credentials.
Recalling this, I was struck again by something in Dinesh’s post:
“THE EMPEROR’S NEW CLOTHES: A story about an egoistic king [who] believes he was wearing a robe that was invisible to the lower classes [ie anyone who wasn't royal enough], whereas he wasn’t wearing anything.”
Actually, the story is also about the clever pair of rascals who japed the king into blindly and faithfully accepting their story (and swiping a few bags of gold, but that is another story, hopefully not part of the modern Indian saga).
So, in the new clothes being sold, like a pup, to the country, there is some new fabric, that didn’t belong in the old fable. This is the inviolability of digital storage systems.
To be sure, there are millions of digital storage systems around the world into which no-one has ever been broken. Why then worry about the one (or ones) in which the digitsed personal information of 1.2 bn+ folks are going to one day be stored?
UIDAI has a simple answer: the information itself won’t be stored there, their store will only link the biometric information with the number. Of course, since other bits of personal information may or may not be pertinent, depending on the service being offered, the storage system will also provide a linking service between dozens of ‘silos’ of information. A silo is the charming geeky term for an information store that, figuratively, stacks up vertically, insulated from other such stores.
Interlinking stores is not a very good idea. That’s the general recommendation of security experts. Digital security experts. Dinesh’s post was probably triggered by the clever attack on the digital store maintained by an American cybersecurity firm (the matter is subjudice, so I’m protecting myself legally by not providing links here) by an anonymous network of cyberexperts (yes, there is a clue in that phrase, I’m not leaving my Devoted Readers entirely in the dark).
That story illustrates my point: the inviolability of digital stores is very closely linked to the value of the store. Very few people are interested in breaking into a pile of old clothes, they want stuff that can be traded for real value (in the case above, it was the value of very publicly embarrassing the cybersecurity firm, flushing out details of some very dodgy ethical practices).
What could the value of UIDAI’s store be, in this real world? After all, UIDAI is going to serve the poor. “Aadhaar will empower poor and underprivileged residents in accessing services such as the formal banking system and give them the opportunity to easily avail various other services provided by the Government and the private sector.”
Sounds very innocuous, and not very attractive to a would-be thief.
Till one reads between the lines. Bank accounts, mobile phones, payment systems – all are, already or in the pipeline, features of the modern economy that will be ‘facilitated’ by Aadhaar. There’ll be more, but this is quite enough, eh?
Now that’s a treasure!
And to close, let me leave my Patient Reader with this last thought from Dinesh’s post:
“Hape is inevitable.”
Hape is not, as the Quick Reader may imagine, the last of the items (the others, like hape, were also evil) to emerge from Pandora’s box, in the Greek myth.
All round the world, business content and business processes are being digitised and made available to stakeholders, often in a highly restricted manner, ie on some kind of subscription basis. Subscribers do so, often paying money, in the faith that they get, in return, some sort of exclusive or protected access.
Quite often, in order to do so, they also voluntarily make available some kind of personal information (it varies, of course, depending on the need).The people who put together the service then store this digitised information and make it available as a part of their subscriber verification.
It works like this: if you are who you claim to be, then you should know [this] about you, because you gave us that information. If you don’t know [this], it isn’t you, so please (the please is optional, and mostly absent, I have observed) go away.
That’s all very well.
What underlies all of this is a tremendous amount of faith, faith in the quality of business processes used by the people who manage that service, that they will keep their part of the bargain – which is sometimes expressly stated, and sometimes implicit – to hold the information securely. And, a lot of the time, that is exactly what happens.
Until it hapens.
Hape is a word coined by my fellow listmember, Dinesh Bareja, to describe what happens when that trust is belied. To quote from his entertaining post:
“The Theory of Hape (abridged):
Every system or technology environment is built with known or unknown holes all over waiting to be penetrated and exploited.
After a hape, weak controls and dirty data are exposed to the world, and management [people] have to run around trying to save their reputation, jobs and more.”
These words somehow reminded me of India’s shining hope, the UIDAI, a quasi-legal organisation foisted on the country by the Planning Commission. The latter, an august body is not known for its cowboy antics, has quixotically chosen the path less often trod, that of mandating a special purpose vehicle, an Authority of India, with the task of ensuring that every resident of India gets a unique identification, a number that cannot possibly be allotted to anyone else, a passkey to all manner of delights, fancifully branded Aadhaar (foundation).
Of course, to work, whoever is charged with delivery of a particular service must be sure that the person quoting the number is actually who she says she is. Or he is, if you object to commonly accepted gender-free phraseology.
Enter the verifier. In its wisdom, UIDAI has decided that this shall take the form of biometric markers – fingerprints will do the job.
Or not. Turns out that fingerprints, the stuff of crime novels for well over a century, are well left there, in works of fiction.
Fingerprints have some problems: 1. they are not immutable, they can change with time, depending on the kind of work the person engages in, and also the state of health; 2. they may not be unique (no study of very large populations has ever been conducted, so the belief in fingerprinting is no more real than a belief in the Flying Spaghetti Monster – or no less real, to be sure); 3. fingerprint recording machines are not very pragmatic for countries like India (shaky electricity, poor hygiene, poor housekeeping); 4. digital fingerprints are based on algorithms that have never been applied to very large populations, so they may be even less perfectly unique than the fingerprint patterns themselves; and so on.
Enter the iris. This central part of the human eye turns out to be even more uniquely patterned than the fingerprint, and luckily, its digital version is also more perfectly matched to the real thing, than in the case of fingerprints. Except. This has also not been researched and scientifically established.
Somewhere in all this hype, one little factor seems to be missing – the act of verification. Each nodal point where it is needed will have to be equipped with a biometric scanner and Go/No-Go display device that will need to communicate very fast (oh, very fast indeed) with a digital store to immaculately match those credentials.
Recalling this, I was struck again by something in Dinesh’s post:
“THE EMPEROR’S NEW CLOTHES: A story about an egoistic king [who] believes he was wearing a robe that was invisible to the lower classes [ie anyone who wasn't royal enough], whereas he wasn’t wearing anything.”
Actually, the story is also about the clever pair of rascals who japed the king into blindly and faithfully accepting their story (and swiping a few bags of gold, but that is another story, hopefully not part of the modern Indian saga).
So, in the new clothes being sold, like a pup, to the country, there is some new fabric, that didn’t belong in the old fable. This is the inviolability of digital storage systems.
To be sure, there are millions of digital storage systems around the world into which no-one has ever been broken. Why then worry about the one (or ones) in which the digitsed personal information of 1.2 bn+ folks are going to one day be stored?
UIDAI has a simple answer: the information itself won’t be stored there, their store will only link the biometric information with the number. Of course, since other bits of personal information may or may not be pertinent, depending on the service being offered, the storage system will also provide a linking service between dozens of ‘silos’ of information. A silo is the charming geeky term for an information store that, figuratively, stacks up vertically, insulated from other such stores.
Interlinking stores is not a very good idea. That’s the general recommendation of security experts. Digital security experts. Dinesh’s post was probably triggered by the clever attack on the digital store maintained by an American cybersecurity firm (the matter is subjudice, so I’m protecting myself legally by not providing links here) by an anonymous network of cyberexperts (yes, there is a clue in that phrase, I’m not leaving my Devoted Readers entirely in the dark).
That story illustrates my point: the inviolability of digital stores is very closely linked to the value of the store. Very few people are interested in breaking into a pile of old clothes, they want stuff that can be traded for real value (in the case above, it was the value of very publicly embarrassing the cybersecurity firm, flushing out details of some very dodgy ethical practices).
What could the value of UIDAI’s store be, in this real world? After all, UIDAI is going to serve the poor. “Aadhaar will empower poor and underprivileged residents in accessing services such as the formal banking system and give them the opportunity to easily avail various other services provided by the Government and the private sector.”
Sounds very innocuous, and not very attractive to a would-be thief.
Till one reads between the lines. Bank accounts, mobile phones, payment systems – all are, already or in the pipeline, features of the modern economy that will be ‘facilitated’ by Aadhaar. There’ll be more, but this is quite enough, eh?
Now that’s a treasure!
And to close, let me leave my Patient Reader with this last thought from Dinesh’s post:
“Hape is inevitable.”
