h a l f b a k e r ySuperficial Intelligence
add, search, annotate, link, view, overview, recent, by name, random
news, help, about, links, report a problem
browse anonymously,
or get an account
and write.
register,
|
|
|
This would be a (necessarily) massive database, probably powering a website, containing the probabilities for everything humankind has calculated/estimated one for.
Each event would have a description, the likelihood of it occurring _given no other assumptions_, and the option to add a number of
clauses (because the likelihood of being mauled by a polar bear is higher at the North Pole than in the Marianas Trench).
The fun, if this kind of thing rattles yr bucket, begins when you start querying the database to find out just how likely something is to happen to you - or just how unlikely something that happened to you was. The ability to string together swathes of events, selecting clauses where necessary, will give us the power to verify what that bloke told us very authoritatively down the pub last night.
There'd have to be some kind of system for resolving disputes - or simply the option to choose a favoured probability, and various logical & mathematical operators (e.g. every, 4 occurrences of), so that just about any event can be described in as fine detail as required.
Whilst such a thing could never claim to be comprehensive (or, at least, the probability of it would be exceedingly low), for producing rough guidelines and putting practically unlimited statistics at our fingertips, I think it'd be pretty cool.
Personally I keep wondering what the chances are that every single girlfriend I've had studied for a psychology degree.
Point a MySQL engine at this
http://www.halfbake...e_20of_20everything
[angel, Oct 04 2004, last modified Oct 05 2004]
Long Bets
http://www.longbets.org/
really a different thing entirely, but this reminded me of it [krelnik, Oct 04 2004, last modified Oct 21 2004]
Anti-Lemon
http://www.halfbakery.com/idea/Anti-Lemon
Mayhap it'll help you find this. [phoenix, Oct 04 2004, last modified Oct 05 2004]
Odds of Death Due to Injury, United States, 1999
http://www.nsc.org/lrs/statinfo/odds.htm
Odds of dying from being bitten or stung by nonvenomous insect: 1 in 355,698. Some people are just unlucky, I guess. [dalek, Oct 04 2004, last modified Oct 21 2004]
Probabilities of asteroids hitting Earth
http://192.211.16.1...steroid/impacts.htm
Altogether too high for comfort, but these still look like wild guesses. [dalek, Oct 04 2004, last modified Oct 21 2004]
The Elements of Style
http://www.bartleby.com/141/
[thumbwax, Oct 04 2004, last modified Oct 21 2004]
Please log in.
If you're not logged in,
you can see what this page
looks like, but you will
not be able to add anything.
Annotation:
|
| |
Brilliant. One small step for the future omnipotent life form in the universe, one giant leap for an army of programmers. |
|
| |
What are the chances of me liking this? {hint: (+)} |
|
| |
Makin' bacon spontaneously. |
|
| |
pretty sure I read that there was more chance of being struck by a meteorite than winning the lottery..
or was that winning it twice? or was it hit by a bus? |
|
| |
Was'nt the ships computer capable of this in the Douglas Adams book The Hitchhiker's Guide To The Galaxy? |
|
| |
This would be the online Bible for insurance adjusters. |
|
| |
Interestingly, [Unabubba], while the odds of a meteorite
hitting an individual are very low, the odds of someone
being killed by the kind of huge meteor impact such as is
thought to have
caused the extinction of the dinosaurs is not that low.
Suppose a meteor impact strikes the Earth every
500,000 years which wipes out half the Earth's population.
Assume the human lifespan is 100 years - there is then a
1/5000 chance of it happening in your lifetime, and a
1/10,000 chance of you dying. |
|
| |
...and being fatally injured by a lottery ball? |
|
| |
The "Heart of Gold" was the name of the ship in H2G2. It was based on the probability generators, which reach the probability of something happening (like an apple appearing) so that it does happen. But, the Heart of Gold used an IMprobability drive, as it was improbable to be in all points of the universe at once, as it was capable of doing. That device had the same name, and was created by a regular probability machine. |
|
| |
I had actually been working on something almost like this, or at least something that could be used to build this database and to query it. It was a piece of software called Psychic, which given patterns of numbers and statistics, would produce the must probable future values. |
|
| |
This idea is good. It is even bakable. I give my flaky bit of goodness to it. |
|
| |
What happens if something with an extremely low probability of occurrence does come to pass? Is it logged as a bug? |
|
| |
I'm thinking of the high school girl in Washington State who got struck twice by lightning, a few years ago. She was fine; they interviewed her on a national morning program afterward. This girl had a mouth so big you'd swear the top of her head was about to come off when she opened it; and when she started speaking, you understood why a righteous God had tried twice to kill her. That voice was like fingernails on a chalkboard. |
|
| |
It was her umbrella that both acted as a lightning rod (big, with one of those long spikes on top) and saved her life. Something about a shock-proof handle. |
|
| |
On the morning this girl was interviewed, millions of Americans sat in front of their television sets and cursed the vagaries of probability for this egregious failure. |
|
| |
Then they turned the sound down. |
|
| |
1percent: My understanding is that things with low probability of occurrance happen all the time and are perfectly normal. Why would their occurance indicate a bug? |
|
| |
I guess it's the staggering notion of the same improbable thing happening twice, to the same person, in the same way, in as many weeks, that inspires me to invoke the term "bug". |
|
| |
Lots of QA work in a database environment will cause a person to see bugs everywhere. if my business were faith I'd call it a "miracle" ... and if my business were Microsoft, I'd call it a "feature". |
|
| |
If you're business was Microsoft, you could call it pretty much anything you want. |
|
| |
dos fritos, please see link I've provided |
|
| |
[/pedant_on] You say [alx] _given no other assumptions_ but in fact you would need to make the most enormous of assumptions every time you looked at the probability of something happening. This would be that nothing related to the conditions that allowed for the probability of the event to transpire would change in the meantime, and affect its probability of transpiring. 'Ceteris paribus' as my old economics teacher used to say with an infuriating expression on his face. [/pedant_off] |
|
| |
Grand idea though, definitely worth some pastry. |
|
| |