Archive: January 10, 2008

<<< January 9, 2008

Home

January 11, 2008 >>>


milestone ScanScopes

Thursday,  01/10/08  10:46 PM

And now for a little chest beating...  Just wanted to report a cool Aperio milestone, we recently completed manufacturing our 100th ScanScope XT and our 200th ScanScope CS!  The picture below shows these two units in the manufacturing test area, along with a few of the many people responsible for this achievement. 

 

Not pictured but very much "in the picture" is Marc Cohn, Aperio's Director of Manufacturing, who leads this team.  Congratulations to all Aperions for their hard work and dedication which made this accomplishment possible!

 

Thursday,  01/10/08  10:48 PM

Ronald Baily: Why we should care what candidates think about biological evolution.  "I don't need them to describe their faith; that's their business and not mine.  But I do care about their scientific knowledge and how it will inform their leadership."  Not only their knowledge, but their ability to reason.  Would I vote for a candidate who didn't "believe in evolution"?  No.  [ via Panda's Thumb

Yesterday I posted about the sheep of Saint-Affrique, including the Millau Viaduct.  So I Googled for "Millau", and got a hit on this site, from December of 2004!  Sometimes the small world is even smaller than you thought! 

Wired: The Untold Story: How the iPhone blew up the Wireless industry.  "The demo was not going well.  Again.  It was a late morning in the fall of 2006...  It was clear that the prototype was still a disaster.  It wasn't just buggy, it flat-out didn't work. The phone dropped calls constantly, the battery stopped charging before it was full, data and applications routinely became corrupted and unusable...  At the end of the demo, Jobs fixed the dozen or so people in the room with a level stare and said, 'We don't have a product yet.'"  The rest is [or will be] history... 

Jason Kottke comments: "I wish this story had more direct quotes and explicit references...it's hard to read it and not take the whole thing with a huge grain of salt."  Yeah, but it feels true :)

Zeldman: Everything that can be iPhonelike, will be...  "It’s inevitable.  The iPhone is too great a leap forward in interface design to be confined to, well, the iPhone."  I predict this will seem prescient after Macworld.  Stay tuned :)

Well, if you read Scott Rosenberg's excellent Dreaming in Code you knew this was coming: CNet reports Mitch Kapor bails on the Chandler Project.  "Kapor--the designer of Lotus 1-2-3, the co-founder of the Electronic Frontier Foundation, the founding chairman of the Mozilla Foundation, and the chairman of Linden Labs--certainly can find other things to do.  OSAF wasn't a bad idea, and Chandler wasn't either.  But neither was managed particularly well."  So be it. 

Uncov is merciless: Wikia Search: Weapons Grade Fail.  "Here's the story: Wikia has set up Nutch, a mediocre open source search engine, and laid a wiki on top of it. You can write a mini wiki page for your search result if you don't like what Nutch comes up with. This is even more craptacular than Mahalo."  Fire on 'em! 

OMG!  Halley Suitt is back!!  Yay! 

 

cowboys dilemma analyzed

Thursday,  01/10/08  11:57 PM

Okay, here we go; an analysis of the cowboy's dilemma.

You're a cowboy, and get involved in a three way pistol duel with two other cowboys.  You are a poor shot, with an accuracy of only 33%.  The other two cowboys shoot with accuracies of 50% and 100%, respectively.  The rules of the duel are one shot per cowboy per round.  The shooting order is from worst shooter to best shooter, so you get to shoot first, the 50% guy goes second, and the 100% guy goes third, then repeat.  If a cowboy is shot he's out for good, and his turn is skipped.  Where or who should you shoot first?

Yesterday we noted there are three possibilities:

  1. Shoot at the 100% cowboy.
  2. Shoot at the 50% cowboy.
  3. Shoot anywhere else.

So we want to determine which of these possibilities has the best probability for our survival.  To do this, we take each case and determine all the outcomes, along with the probability of each outcome.  The overall probability for the case is the sum of all the possible outcomes.  Here we go:

n      Shoot at 100% cowboy

o       33%: hit him

§         50% cowboy shoots at you (you’re the only one left)

·         50%: he hits you (he wins – 33*50)

·         50%: he misses, your turn

o       You shoot at 50% cowboy

§         33%: hit him (you win – 33*50*33)

§         66%: you miss, his turn

·         50%: he hits you (he wins – 33*50*66*50)

·         50%: he misses, your turn

o       33%: hit him (you win – 33*50*66*50*33)

o       66% you miss, his turn

§         50%: he hits you (he wins – 33*50*66*50*66*50)

§         50%: he misses, your turn (33*50*66*50*66*50)

§         Probability of winning approx .06%, ignore further…

o       66%: you miss

§         50% cowboy shoots at 100% cowboy (best chance for him)

·         50%: he hits him (100% cowboy dead)

o       You shoot at 50% cowboy

§         33%: hit him (you win – 66*50*33)

§         66% you miss, his turn

·         50%: he hits you (he wins – 66*50*66*50)

·         50%: he misses, your turn

o       33%: hit him (you win – 66*50*66*50*33)

o       66% you miss, his turn

§         50%: he hits you (he wins – 66*50*66*50*66*50)

§         50%: he misses, your turn (66*50*66*50*66*50)

§         Probability of winning approx 1.2%, ignore further…

·         50%: he misses, 100% cowboy's turn

o       100% cowboy shoots at 50% cowboy (best chance for him)

§         100%: kills him, your turn

§         You shoot at 100% cowboy

·         33%: hit him (you win – 66*50*33)

·         66%: you miss, his turn

o       100%: he hits you (he wins – 66*50*66)

The total probability of survival is the sum of each outcome where you win, highlighted above:
(33*50*33 +  33*50*66*50*33 + .06 + 66*50*33 + 66*50*66*50*33 + 1.2 + 66*50*33 = 34.4

If we shoot at the 100% cowboy, our probability of winning is 34%.

n      Shoot at 50% cowboy

o       33%: hit him

§         100% cowboy shoots at you (you’re the only one left)

·         100%: he hits you (he wins – 33)

o       66%: you miss

§         50% cowboy shoots at 100% cowboy (best chance for him)

·         50%: he hits him (100% cowboy dead)

o       You shoot at 50% cowboy

§         33%: hit him (you win – 66*50*33)

§         66% you miss, his turn

·         50%: he hits you (he wins – 66*50*66*50)

·         50%: he misses, your turn

o       33%: hit him (you win – 66*50*66*50*33)

o       66% you miss, his turn

§         50%: he hits you (he wins – 66*50*66*50*66*50)

§         50%: he misses, your turn (66*50*66*50*66*50)

§         Probability of winning approx 1.2%, ignore further…

·         50%: he misses, 100% shooter’s turn

o       100% cowboy shoots at 50% cowboy (best chance for him)

§         100%: kills him, your turn

§         You shoot at 100% cowboy

·         33%: hit him (you win – 66*50*33)

·         66%: you miss, his turn

o       100%: he hits you (he wins – 66*50*66)

The total probability of survival is the sum of each outcome where you win, highlighted above:
(66*50*33 + 66*50*66*50*33 + 1.2 + 66*50*33 = 26.6

If we shoot at the 50% cowboy, our probability of winning is 27%.

n      Shoot at the sky

o       50% shooter’s turn, he shoots at 100% shooter (best chance for him)

§         50%: he hits him (100% shooter dead)

·         You shoot at 50% shooter

o       33%: hit him (you win – 50*33)

o       66% you miss, his turn

§         50%: he hits you (he wins – 50*66*50)

§         50%: he misses, your turn

·         33%: hit him (you win – 50*66*50*33)

·         66% you miss, his turn

o       50%: he hits you (he wins – 50*66*50*66*50)

o       50%: he misses, your turn (50*66*50*66*50)

o       Probability of winning approx 1.8%, ignore further…

§         50%: he misses, 100% shooter’s turn

·         100% shooter shoots at 50% shooter (best chance for him)

o       100%: kills him, your turn

o       You shoot at 100% shooter

§         33%: hit him (you win – 50*33)

§         66%: you miss, his turn

·         100%: he hits you (he wins – 50*66)

The total probability of survival is the sum of each outcome where you win, highlighted above:
50*33 + 50*66*50*33 + 1.8 + 50*33 = 40.2

If we shoot at the sky, our probability of winning is 40%.

 

There you have it.  The best option is door #3, shoot at the sky.  We actually have a 40% chance of winning if we do so, despite the fact that both of the other cowboys are better shots than we are.

This is a pretty interesting result.  By not shooting at either of the "better" cowboys, we let them shoot at each other first, doing our dirty work for us.  Then later we engage and shoot at whoever's left.  This same effect is responsible for the complexity of reality TV shows.  Many times a "weaker" competitor can survive by being unaggressive in the early going, enabling the "stronger" competitors to take each other out.  Then later the weaker competitor can become more aggresive and [possibly] win.  Pretty cool!



 
 

Return to the archive.