On Monday 24 July 2006 11:16 am, Abhishek Daga wrote:
--- jtd jtd@mtnl.net.in wrote:
On Monday 24 July 2006 10:42 am, jtd wrote:
Even dna fingerprinting is far from fool proof with a substantial portion of the population having chimeric genes - genes that do not belong to either parent. No mommy wasn't having a ball on the sidelines. And besides that does not explain how dady's and mr. X's genes both got into knottty baby.
Ok so what are are claiming? that we are back to square one? or all the steps we took forward werent really forward?
Yes. Mostly. For simple cases where physical prescence is essential and can be counter verified by a person or where id verification need not be 100.000000 % it works well. If the It dept wants to weed out false identity they are being stupid to think that existing biometric systems will work. In fact it would be the perfect play ground for scamsters. If it is for the convienence of tax payers AES and plainold passwords work well.
And what FOSS options are available that work with handheld or fingerprint scanners?
None as far as algorithims go (see earlier links). So far foss are still investigating algorithms and ideas and abandoning them because of the above said flaws or costs or impracticality - the foss community does not paper over problems with fancy presenatations. However there is sufficient gpld software to get things done with all the serious flaws that exist with biometric tech.