81–90 of 169 posts
This reminds me of Akinator: http://en.akinator.com It's a little spammy nowadays, but it's had enough input that it seems pretty amazingly accurate at "guessing" what / who you are thinking of in ~ 25 questions.
Earlier quoted context omitted.
Here's my admittedly naive Sunday afternoon spitball on an elegant solution: I like the idea of a human UUID/GUID type identifier. I would also like to think that this is solvable using strictly biological and physical properties, sampled at birth. Otherwise, time and culture factors would seem make it difficult to produce a static set of "apples to apples" questions and answers. I wonder if the right maths applied t…
DNA is your UUID/GUID
Just use: - Birthday (19~ bits) - Rough Location (remaining bits) And base the questions around those two, for example, where you born on a 1-15th, does the city you were born in start with the letter's a-k. This part would be an exercise in statistics, I would think. edit: And one bit for if you were the first to be born of two identical twins =p
I don't think this problem is solvable in any elegant form, but it is solvable. You'll just end up with massively conjunctive questions that you can't even hold in your head at once, like "27: Are you a non-practicing Catholic with exactly three children, or an asian owner of a minivan produced between 1998 and 2004 that isn't green, or a licensed boat mechanic with astigmatism, or..." and so on for the next 6 pages.…
Here's my admittedly naive Sunday afternoon spitball on an elegant solution: I like the idea of a human UUID/GUID type identifier. I would also like to think that this is solvable using strictly biological and physical properties, sampled at birth. Otherwise, time and culture factors would seem make it difficult to produce a static set of "apples to apples" questions and answers. I wonder if the right maths applied t…
I'd imagine that genetic markers would be the best way to do to (Disclaimer: I'm no biologist and might have made completely wrong assumptions here). They're less likely to change than, say, someone's political or religious beliefs; one could get a nasty hit on their head and forget.
The thing with genetics is that they can change over time. Some gene's turn on and off. Attributes like your face and fingerprints change over time. They're not constants.
If you could identify a set of 33 lifetime constants, you'd end up with a life-long UUID. If you expanded beyond 33 bits and included genetic markers which change over time, such as gene's which flip on/off, you could end up with a point-in-time (PIT) UUID.
UUID = Constant throughout life.
PIT+UUID = UUID plus markers identifying you at a particular point in time.
Also, you'd probably need more than 33 bits. 33 will encompass all living humans today in 2013 ADE, but would have to be expanded as the living population grows, and to include the billions of deceased humans.
In the end, a "true" unique identifier, encompassing any human, would be their UUID plus a list of all PIT+UUID's they generated during their lifetime. Or in english, an entire record of their genetics from start to end:
struct LIFETIME_UUID {
void * uuid;
void * pit_uuid_TIMESTAMP1;
void * pit_uuid_TIMESTAMP2;
void * pit_uuid_TIMESTAMP3;
...
};
Are there 33 genetic markers that each has no correlation on the presence of the others?
If the goal is to have questions which can be answered only with yes or no. I don't think asking for location of the person is good thing, because there would be so many questions as there is locations. "Do you live in China, India, The United States, Indonesia, Brazil or Pakistan?" is not good question.
I don't think this problem is solvable in any elegant form, but it is solvable. You'll just end up with massively conjunctive questions that you can't even hold in your head at once, like "27: Are you a non-practicing Catholic with exactly three children, or an asian owner of a minivan produced between 1998 and 2004 that isn't green, or a licensed boat mechanic with astigmatism, or..." and so on for the next 6 pages.…
So maybe we get to assume we have some oracle that helps us simplify the hard questions.
At that stage, it's easy. Begin with, "Assume we build a list of people sorted by time of birth (with some arbitrary tiebreakers, like proximity of birthplace to Barbados, or darkness of hair color...)."
Question 1: Are you on the top half or bottom half of this list?
Question 2: Are you on the top quarter or bottom quarter of the half?
Question 3: ...