House of ID Cards collapses
Chris Mellor writes in Techworld:
The government has made a U-turn on controversial plans for a single ID Card database.
Instead of one huge, new National Identity Register (NIR) database, it will now be spread across three existing systems. John Reid, the Home Secretary, said it was “a sensible decision” and denied it was a U-turn.
…
This has not fostered much confidence in the project. Of the three systems, one is known to be inadequate except as a stop-gap, the technical feasibility of the second is unknown and the third will need enlarged capabilities to cope with the amount of ID card information. NIR data will also have to be seperated by digital Chinese walls from other data on these systems.
The need to construct a complete ID card from three separate systems not designed to co-operate in the first place, and with sufficient security checks, will add to the scheme’s complexity. The speed of ID card checking will be affected by the performance characteristics of these sub-systems. And there appears to be an initial likelihood that the biometric component is likely to be slow given the stop-gap nature of the chosen system. The future addition of a new biometric IT system adds further uncertainty.
The Home Office stated: “Biometric technology identifies small percentages of what are known as ‘false matches’ or ‘false non-matches’. These need expert human assessment to ensure that matches are being made correctly.” With the numbers of ID cards to be issued such “small numbers” will, in fact, number thousands of cases a year. Human assessments require resources, meaning money, and there is never enough of that.
January 6th, 2007 at 16:34
On this one occasion, the Home Secretary is right.
Big old databases are full of errors. You get duplicates. There are omissions. And the remaining records haven’t had their data updated for years and the data may have been wrong in the first place. Terrible.
And with new databases? You get duplicates. There are omissions. And the data on the remaining records may be wrong. Also terrible. But cheaper and quicker.
Rather than attack this decision, mocking it as a U-turn, I would welcome it as the first sign of a rational response to reality.
And I would hope that this dawn would grow into full daylight.
The results of the UKPS biometrics trials and the experience of US-VISIT in the US both indicate that the biometrics chosen for the new passports and the ID cards scheme cannot identify more than 80% of people. I don’t know where Mr Mellor gets his “thousands of cases a year” figure from but, with 50m ID cards in circulation, that evidence indicates that false non-matches would affect 10m people.
We will know that daylight is upon us and him when the Home Secretary admits that biometrics are not up to it, the technology is flawed and deploying it is an international charade, please see http://www.europeanbiometrics.info/resources/details_s.php?Id_resources=110 and http://dematerialisedid.com/Evidence/Biometrics.html.
January 6th, 2007 at 16:47
CORRECTION
On this one occasion, the Home Secretary is right.
Big old databases are full of errors. You get duplicates. There are omissions. And the remaining records haven’t had their data updated for years and the data may have been wrong in the first place. Terrible.
And with new databases? You get duplicates. There are omissions. And the data on the remaining records may be wrong. Also terrible.
Cheaper and quicker to use the old databases.
Rather than attack this decision, mocking it as a U-turn, I would welcome it as the first sign of a rational response to reality.
And I would hope that this dawn would grow into full daylight.
The results of the UKPS biometrics trials and the experience of US-VISIT in the US both indicate that the biometrics chosen for the new passports and the ID cards scheme cannot identify more than 80% of people. I don’t know where Mr Mellor gets his “thousands of cases a year” figure from but, with 50m ID cards in circulation, that evidence indicates that false non-matches would affect 10m people.
We will know that daylight is upon us and him when the Home Secretary admits that biometrics are not up to it, the technology is flawed and deploying it is an international charade, please see http://www.europeanbiometrics.info/resources/details_s.php?Id_resources=110 and http://dematerialisedid.com/Evidence/Biometrics.html.