CCC comments on Rationality Quotes Thread September 2015 - Less Wrong

3 Post author: elharo 02 September 2015 09:25AM

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

Comments (482)

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

Comment author: CCC 15 October 2015 09:49:54AM 0 points [-]

...it would be a bit like plotting 0, 1 and i colinearly. (I assume you're familiar with complex numbers?)

Comment author: gjm 15 October 2015 01:43:14PM 0 points [-]

Yes, very familiar with complex numbers, thanks. But, I repeat, you can plot what you want however you want; the question is whether it's helpful, and that will depend on the application. (Suppose the values taken by your dependent variable are all on the circle of radius 1/sqrt(2) centred at (1+i)/2. Then plotting 0, 1, and i collinearly may make a whole lot of sense, though you might actually want to call them -3pi/4, -pi/4 and pi/4 respectively.)

Comment author: CCC 16 October 2015 10:29:19AM 0 points [-]

(Suppose the values taken by your dependent variable are all on the circle of radius 1/sqrt(2) centred at (1+i)/2. Then plotting 0, 1, and i collinearly may make a whole lot of sense, though you might actually want to call them -3pi/4, -pi/4 and pi/4 respectively.)

I reluctantly concede the point, but firmly maintain that calling them -3pi/4, -pi/4 and pi/4 respectively would make a lot more sense.