begin  quoting Todd Walton as of Tue, Jul 17, 2007 at 11:48:00PM -0500:
[snip]
> "'The problem was that all the different pricing rules interact in
> ways that not even those who designed the pricing systems could begin
> to fully understand. Mathematically, this made the (idealized) problem
> of finding an optimal fare between two given locations undecidable,
> which means that it is impossible to write a computer program to solve
> the problem."

Baloney. I don't buy it.

> The crazy math of airline ticket pricing:
> http://www.maa.org/devlin/devlin_09_02.html

  "Admittedly, in order to study the airline pricing problem mathematically
  and obtain these results, de Marcken had to make some simplifying
  assumptions that don't apply to real airline travel, such as allowing an
  unlimited number of destinations, flights of unlimited length, or
  arbitrarily long lists of rules. But the implications for real airline
  pricing are unavoidable."

Okay, once you have an unlimited number of destinations and allow for
flights of unlimited length, yeah, I buy *that* is undecidable.

In the Real World, however, we have a finite number of airplanes and
pilots, each with a finite distance they can fly (think refueling),
between finite points, using a finite set of rules for pricing.

That's a finite problem, and /that/ means it's decidable.

-- 
It doesn't mean it's _tractable_.
Stewart Stremler


-- 
[email protected]
http://www.kernel-panic.org/cgi-bin/mailman/listinfo/kplug-list

Reply via email to