How to Universally Identify a Person In Your Database – DNA Hash Tag
Database developers have argued about the best way to uniquely identify an individual for as long as there have been relational databases. Many solutions have been used over the years with differing degrees of success.
In the USA the Social Security Number (SSN) has been popular. A parent cannot claim a child on their tax forms without first getting a Social Security Number for their child from the US Government. That doesn’t mean that every child has to have an SSN; just that you can’t claim them as a dependent on your tax return and receive a reduced tax assessment amount as a result.
Because they are practically universal in the USA, many software systems working in the USA have tried using the SSN as a unique identifier for individuals in their system. What happens when an individual does not have an SSN? Their database won’t work until the individual obtains an SSN. Does it happen often? No. But, when it happens, the work you are trying to accomplish is blocked until a number can be obtained.
The problem that I have with an SSN is that it is not guaranteed unique. It is assigned by an external system which may or may not be correct. It has nothing to do with the individual; the only thing an SSN has in common amongst all individuals is that it may be assigned by a centralized authority, and hopefully unique.
So, what do you do when your software must work outside of the USA? You can use a similar technique from other authorities from other countries, etc. You may even generate your own number system as many health insurance organizations do. They may link you back to an external authority; but they assign their own guaranteed unique number as well.
So, I’m getting out of the box here with a totally different concept. The only thing we can measure that is guaranteed to uniquely identify any individual in history past, present, or future is our DNA. So, why not come up with a system that will generate a hash value based on that DNA, just as unique as the DNA sequence itself, but small enough to fit in a long integer.
I’m not sure how difficult it would be to do. But, we could stop arguing about how to uniquely identify an individual, and have something that would be universal for any database irrespective of any other factor.
I’m picturing a device that will take a small sample and generate a consistent value based on the DNA of an individual that would be guaranteed unique. In this fashion, a person carries their unique id in their very person. No card is required or can be counterfeit. Nothing is implanted, because the uniqueness is part of every cell that makes up an individual.
That’s my dream for tonight. What do you think? Is this virtually impossible? What will you use in the meantime to uniquely identify an individual in your data? Leave your comment below, or drop an email to btaylor@sswug.org.
Cheers,
Ben