adsenanim comments on Kevin T. Kelly's Ockham Efficiency Theorem - Less Wrong

30 Post author: Johnicholas 16 August 2010 04:46AM

You are viewing a comment permalink. View the original post to see all comments and the full post content.

Comments (81)

You are viewing a single comment's thread. Show more comments above.

Comment author: adsenanim 18 August 2010 08:04:39PM *  1 point [-]

I think removing the parentheses can affect the overall pattern formation.

12 == (((1+1)^(1+1))*(1+1+1))

12 != +^+*++

Because we are changing the order of operations:

1+1^1+1*1+1+1

2^2*3

2^6

64

Also, I don't think there is a way to use exponents and make 12 without using parentheses.

So if we simplify the method of notation, then some operations cannot be used for certain numbers. and this changes the pattern, and/or the availability of possible patterns.

This topic is discussed in D. Hofstadter's book Godel, Escher, Bach: an Eternal Golden Braid.

Comment author: thomblake 18 August 2010 08:14:19PM 0 points [-]

Indeed, I hadn't thought far enough ahead to worry about cases where order of operations would matter between exponentiation and multiplication. So parens would have to be acceptable where it's ambiguous. What the default order of operations should be, can be left as an exercise for the reader.

as for your specific concern, if we left the order of operations as "+ then ^ then *", +^+*++ would work fine.

Comment author: DSimon 18 August 2010 09:26:09PM 0 points [-]

We could keep the abbreviated syntax but avoid the necessity of parentheses by using post-op notation, couldn't we?

Comment author: thomblake 18 August 2010 09:32:36PM 0 points [-]

For that, I think we'd need a stack with a line-termination character. In effect, we'd be removing two characters for one, which I suppose would be an improvement.

Comment author: adsenanim 18 August 2010 10:02:25PM *  0 points [-]

So:

  • The order of operations can be variable dependent on the number

  • The notation method effects pattern

  • The pattern can change if primes are always used

  • Etc.

How many ways can we change the rules? How many rules are there?

It is amazing just how much variation in pattern can be achieved just by changing and/or adding/subtracting rules.

There is the general rule of rules:

The fewer the rules, the less variable the pattern, and the inverse, the more rules the more variable the pattern.

  • Sorry for the late edit, but, I would also add that the fewer the rules the less adaptable as well...