What is the correct picture of our world? Are noise and errors part of the essence of matters, and the beautiful perfect patterns we see around us, as well as the notions of information and computation, are just derived concepts in a noisy world? Or do noise and errors just express our imperfect perception of otherwise perfect laws of nature? Talking about an inherently noisy reality may well reflect a better understanding across various scales and areas.

### Recent Comments

Gil Kalai on To Cheer You Up in Difficult T… Gil Kalai on To Cheer You Up in Difficult T… Alexander Barvinok on To Cheer You Up in Difficult T… Kevin on To Cheer You Up in Difficult T… Gil Kalai on To Cheer You Up in Difficult T… Arseniy on To Cheer You Up in Difficult T… Alexander Barvinok on To Cheer You Up in Difficult T… uniform on To Cheer You Up in Difficult T… Arseniy on To Cheer You Up in Difficult T… normanstresskopf on To Cheer You Up in Difficult T… To Cheer You Up in D… on Beyond the g-conjecture… To Cheer You Up in D… on Euler’s Formula, Fibonac… -
### Recent Posts

- To Cheer You Up in Difficult Times 31: Federico Ardila’s Four Axioms for Cultivating Diversity
- Dream a Little Dream: Quantum Computer Poetry for the Skeptics (Part I, mainly 2019)
- To Cheer you up in difficult times 30: Irit Dinur, Shai Evra, Ron Livne, Alex Lubotzky, and Shahar Mozes Constructed Locally Testable Codes with Constant Rate, Distance, and Locality
- To cheer you up in difficult times 29: Free will, predictability and quantum computers
- Alef’s corner: Mathematical research
- Let me tell you about three of my recent papers
- Mathematical news to cheer you up
- To Cheer You Up in Difficult Times 28: Math On the Beach (Alef’s Corner)
- To cheer you up in difficult times 27: A major recent “Lean” proof verification

### Top Posts & Pages

- To Cheer You Up in Difficult Times 31: Federico Ardila's Four Axioms for Cultivating Diversity
- The Intermediate Value Theorem Applied to Football
- Extremal Combinatorics III: Some Basic Theorems
- Amazing: Karim Adiprasito proved the g-conjecture for spheres!
- Updates and plans III.
- Answer: Lord Kelvin, The Age of the Earth, and the Age of the Sun
- TYI 30: Expected number of Dice throws
- An interview with Noga Alon
- Joel David Hamkins' 1000th MO Answer is Coming

### RSS

While reading Gregory Chaitin’s MetaMath I jotted down some notes:

“Randomness is irreducible, incompressible.

Pattern stems from a subset of randomness.”

I’d consider noise as a subset of randomness containing compressible pattern. As for errors: some – like the halting problem – are unpredictable.

I note that you and most people use laws (plural) of nature.

For laws to be perfect, predictable and without error, I’ve been wondering if that implies irreducibility hence randomness.

Irreducible laws of nature.

I somehow feel happy with this answer, despite the paradox it implies.

Pingback: Readings (II) 01/23/09 | Venture Capital Bloggers Network

“Are noise and errors part of the essence of matters, and the beautiful perfect patterns we see around us, as well as the notions of information and computation, are just derived concepts in a noisy world? Or do noise and errors just express our imperfect perception of otherwise perfect laws of nature?”

I hate to have to choose. Perhaps it can be both.

Gil,

A bit off topic, but have you seen:

Phys. Rev. A 79, 012332 (2009)

Fibonacci scheme for fault-tolerant quantum computation

by Panos Aliferis and John Preskill?

Given your skepticism regarding error correction in connection with Quantum Computation (I believe my assumption is correct), I was curious what you thought of the article.

Dear Michael, fualt-tolerance quantum computation is based on a remarkable theorem called the “threshold theorem” which was proved by several groups of researchers in the mid 90s. Since then there have been significant progress in extending the scope of the theorem in terms of the type of noise it can handle, and reducing the numerical value of the threshold.

A breakthrough work by Knill uses error-detection codes rather than error-correction codes and massive post-selection. This allows one to raise the value of the threshold (based on numerical simulations) to 0.03 or so. This idea also leads to a substantially higher provable bounds and there are several papers, including I believe the one you cited, that demonstrate it. This is an exciting direction.

Pingback: Noise Sensitivity Lecture and Tales « Combinatorics and more

Pingback: Randomness in Nature « Combinatorics and more