# CFD-Wiki talk:Format and style guide

(Difference between revisions)
 Revision as of 00:18, 18 December 2005 (view source)Jola (Talk | contribs) (added a bit more to the notation confusion)← Older edit Revision as of 07:04, 18 December 2005 (view source)Tsaad (Talk | contribs) (update)Newer edit → Line 25: Line 25: A third variant on the matrix notation might be to use roman font for it like this: A third variant on the matrix notation might be to use roman font for it like this: * $\mathrm{A}\,\mathbf{x} = \mathbf{b}$  --[[User:Jola|Jola]] 17:18, 17 December 2005 (MST) * $\mathrm{A}\,\mathbf{x} = \mathbf{b}$  --[[User:Jola|Jola]] 17:18, 17 December 2005 (MST) + + --------------------------------------------------------- + I would go with
+ $A\,\mathbf{x} = \vec{b}$.
+ Just kidding. although it seems a good combination of the above variations :P (Jonas, can we have some smileys in the forums?)
+ Okey, I see that the explicit vector notation is pretty good because the boldface doesn't appear really bold on the webpage (or something is messed up with my screen), in addition to the fact that we will most likely use this form of the equation two or three times (in the introduction, in the introduction again, and in the conclusion). So there will be no harm in presenting it in vector form. I always have a newbie in my mind, and we don't want them to get confused between vectors, matrices, and real numbers. + *$A\,\vec{x} = \vec{b}$
+ Is there still a hope for using $\phi$ instead of x ? Both views are good. As Jason mentioned, it will be easier to check against some algorithmic implementations when using x, while on the other hand, phi is pretty well known in CFD circles denoting a generic scalar (to the extent that a professor once called me phi cos he didn't know my name! In this case, phi denoted a generic human!)

## Vector/Matrix formating

I (JasonD) propose that we edit a paragraph/blurb here on this until it comes to some steady state and then either go the forum with it or just move it to the style guide.

Here's what we seem to agree on:

• n-tuples (nx1 arrays) need to differentiated from matrices
• Elements of arrays/vectors are lower case with subscripts

Here are some open questions:

• Exact way to differentiate n-tuples:
• \vec
• \mathbf - this will avoid confusion with vectors in $R^3$
• "Best" (for sake of consistency) letter to use as solution n-tuple
• x is used in most of the linear algebra literature, so I (JasonD) think this would be best for the linear systems section.

### Samples

• $Ax = b$
• $A\,x = b$
• $A\vec{x} = \vec{b}$
• $A\,\vec{x} = \vec{b}$
• $A\mathbf{x} = \mathbf{b}$
• $A\,\mathbf{x} = \mathbf{b}$

A third variant on the matrix notation might be to use roman font for it like this:

• $\mathrm{A}\,\mathbf{x} = \mathbf{b}$ --Jola 17:18, 17 December 2005 (MST)

I would go with
$A\,\mathbf{x} = \vec{b}$.
Just kidding. although it seems a good combination of the above variations :P (Jonas, can we have some smileys in the forums?)
Okey, I see that the explicit vector notation is pretty good because the boldface doesn't appear really bold on the webpage (or something is messed up with my screen), in addition to the fact that we will most likely use this form of the equation two or three times (in the introduction, in the introduction again, and in the conclusion). So there will be no harm in presenting it in vector form. I always have a newbie in my mind, and we don't want them to get confused between vectors, matrices, and real numbers.

• $A\,\vec{x} = \vec{b}$

Is there still a hope for using $\phi$ instead of x ? Both views are good. As Jason mentioned, it will be easier to check against some algorithmic implementations when using x, while on the other hand, phi is pretty well known in CFD circles denoting a generic scalar (to the extent that a professor once called me phi cos he didn't know my name! In this case, phi denoted a generic human!)
-- Tony