5:21 AM
@Qmechanic did you delete my comment? If so, thanks :-) I should read the question properly before commenting!

5:40 AM
@JohnRennie : Below which post?

1

How is it proven that the magnetic lines are closed lines? Because the divergence of $\mathbf{B}$ is zero does not sound convincing, since fields like $\mathbf{F} = a\mathbf i +b\mathbf j+c\mathbf k$ (with $a$, $b$, and $c$ being constant) do have zero divergence; and yet the field lines in these...

@Qmechanic That one

@JohnRennie : Nah, that was another mod :)

Ah, it might have been Buzz.
Anyway, I am (once again) grateful to the moderating team! :-)

2 hours later…
7:41 AM
Hi @JohnRennie can you come to our room?

3 hours later…
10:38 AM
@Slereah The first page of Weibel is great, the rest...

4 hours later…
2:14 PM
0

I wanted to know if this question can be asked on Physics SE and if not, can it be discussed here? In almost all mechanics textbooks the quantity $$\int F.\,ds$$ is defined as the work done by the force F. However, since thermodynamics tells us that work is actually a means/method of transferrin...

9 hours later…
11:42 PM
Dumb terminology question...the linear in linear regression is about the model itself while the linear in linear least squares is about the parameters, correct?
So linear regression is specifically about a model $a_0 + a_1 x$ while linear least squares can handle the more general $a_n x^n$?
I always grouped linear regression into the latter case, but it sounds like that may be incorrect (or at least against the common terminology)

11:53 PM
@DanielUnderwood I think terminology in that area is just a mess :P
you can see this confusion in the answers and comments here where people are just really confused about whether "linear regression" means fitting a linear function or a model linear in its parameters
The Wiki article is similarily confused, at one point claiming that fitting a polynomial to the data is linear regression because it's linear in the coefficients, but then talking about a "linear regression line" which only makes sense if you restrict linear regression to fitting a linear function to the data
I think most people will think of fitting a linear function when you say "linear regression", but if you want to be unambiguous you should probably just avoid the term :P