Re: [XPOST] A unique number for every "person" - can it be done?

From: Ioannis (morpheus_at_olympus.mons)
Date: 02/27/05 

Date: Sun, 27 Feb 2005 21:38:19 +0200

TGOS wrote:

> Hello,
>
> First please excuse the xpost, but the topic of this post does not fit
> into any single NG and I didn't want to miss the right people by posting
> to the wrong NG. Maybe we can find out where it really belongs to in the
> discussion process and move the thread to the right group.
>
> For halve a year now I'm thinking about creating an algorithm, this
> sounds like math, but without some form of hashing used in cryptography
> it will most likely get nowhere, and the topic is program related. You
> see, it's very hard to categorize the topic.
>
> The problem can be summarized in one sentence:
>
> Calculate a number for every human being, company and organization on
> earth, that is guaranteed to be unique till the end of time.
>
> The rules in detail:
[snip]
> Be creative, try to find data useful for the purpose. Things you may
> want to use:
>
> - Date of birth / Year of foundation
> - Place of birth (consider not always known, names can change over time,
> better go for coordinates)
> - Name (First, Last / Name of company/organization)
> - Name of parents (consider orphans / companies)
> - Blood Type (consider companies have no blood type)
> - Gender (consider companies)
> - Eye color (should be constant, consider companies)
>
> and so on.
> Just because some people may not know something or something may not
> apply to certain people doesn't mean you can not use it. But explain
> what to do in the case it does not exist or is unknown.

The simplest idea I can come up with would be to find a way to
mathematically encode one's fingerprints, since they are unique. One
could for example take the decimal equivalent of the file contents of a
standard JPEG or GIF's consecutive bytes of one's fingerprint, if one
standardizes the process of scanning it.

A more complicated solution is to use Godel codons. A possible number
would be:

2^{data_1}*3^{data_2}*5^{data_3}*...*p^{data_n},
where p is a prime number and data_i, i in N is taken from a suitable
pool such as the one you present above. So for example, for the data set
you've given above,

N=2^{Date of Birth}*3^{Place of Birth}*5^{Name}*7^{Name of
Parents}*11^{Blood Type}*13^{Gender}*17^{Eye Color}.

The larger your data set, the less the likelihood of collision. For
practical purposes, if you also added a factor for fingerprints or a
factor for DNA genes this number would be unique, although I have no
idea how one would go about encoding DNA.

Since every number is factorizable, you can recover your information,
although with some effort if the data set is large.

> I came up with plenty of ideas, but they were either too complicated or
> creating collisions was too likely, that are not easily resolved.
>
> Simply writing down some data and hashing it creates a decent number,
> but how long will this be collision free? How big must the hash be to be
> secure for thousands of years and 6 billions of people and millions of
> companies/organizations? 

-- 
I. N. Galidakis
http://users.forthnet.gr/ath/jgal/
------------------------------------------
Eventually, _everything_ is understandable

Relevant Pages