6 ms·

Somehow this seems to be a confusion of categories: markets are real-world mechanisms, while P and NP are mathematical abstractions. Does not compute. To the ex

Somehow this seems to be a confusion of categories: markets are real-world mechanisms, while P and NP are mathematical abstractions. Does not compute. To the extent that it does, it's typical mathematical macroeconomic BS.

adrianN 8y ago
I economics you sometimes define mathematical objects to model real world markets. It's these models that the paper talks about.
talltimtom 8y ago
Markets are a conceptual and mathematica abstaction used to describe the real world.... so is computation.
coldtea 8y ago
>markets are real-world mechanisms, while P and NP are mathematical abstractions We use mathematical abstractions to model real-world mechanism every day for millennia. Not only that, but there are all kinds of mathematically defined limits that no real-world mechanism can bypass, from a purely logical perspective. If you only have 10 dollars and I give you 20 dollars, you'll have 30 dollars, not 500 -- that's a mathematical truth that absolutely holds in the real world too.
jonathanstrange 8y ago
Since when do mathematical limitations not apply to the real world? Have you ever tried to square the circle?
- ernst_klim 8y ago
  Let's start with the fact that circles do not exist in the real world.
  - jonathanstrange 8y ago
    Well, that's a quite insubstantial point, to say the least. Mathematical theorems place upper and lower bounds on what is possible in the real world within almost every domain. Of course, there may often be other reasons why something is impossible, too.
  - Retra 8y ago
    They don't need to exist in mathematics either, but the outcome would be the same. You can't do the real-world analogue of squaring the real-world analogue of a circle.
QML 8y ago
Not really. Computational complexity and markets have more in common that most think. Mechanisms which rely on algorithms, such on auctions, have runtimes. Hence, why we have a field called algorithmic game theory [1]. [1] https://en.wikipedia.org/wiki/Algorithmic_game_theory https://en.wikipedia.org/wiki/Algorithmic_game_theory

adrianN 8y ago

I economics you sometimes define mathematical objects to model real world markets. It's these models that the paper talks about.

talltimtom 8y ago

Markets are a conceptual and mathematica abstaction used to describe the real world.... so is computation.

coldtea 8y ago

>markets are real-world mechanisms, while P and NP are mathematical abstractions We use mathematical abstractions to model real-world mechanism every day for millennia. Not only that, but there are all kinds of mathematically defined limits that no real-world mechanism can bypass, from a purely logical perspective. If you only have 10 dollars and I give you 20 dollars, you'll have 30 dollars, not 500 -- that's a mathematical truth that absolutely holds in the real world too.

jonathanstrange 8y ago

Since when do mathematical limitations not apply to the real world? Have you ever tried to square the circle?

ernst_klim 8y ago
Let's start with the fact that circles do not exist in the real world.
- jonathanstrange 8y ago
  Well, that's a quite insubstantial point, to say the least. Mathematical theorems place upper and lower bounds on what is possible in the real world within almost every domain. Of course, there may often be other reasons why something is impossible, too.
- Retra 8y ago
  They don't need to exist in mathematics either, but the outcome would be the same. You can't do the real-world analogue of squaring the real-world analogue of a circle.

QML 8y ago

Not really. Computational complexity and markets have more in common that most think. Mechanisms which rely on algorithms, such on auctions, have runtimes. Hence, why we have a field called algorithmic game theory [1]. [1] https://en.wikipedia.org/wiki/Algorithmic_game_theory https://en.wikipedia.org/wiki/Algorithmic_game_theory