Definitions for equations

Question

I am trying to add the definitions for my equation in my document. I know this has been asked before and I found a beautiful solution. But it is just not working for me.

This is what I found

    \documentclass{article}
    \usepackage{amsmath}% http://ctan.org/pkg/amsmath
    \usepackage{array}% http://ctan.org/pkg/array
    \begin{document}
        \begin{gather}
      P_{xi}=\overline{U}_x+\sigma_x
             \frac{\sum_k^{Nu}D_{kx}\times\left(\frac{S_{ki}-\overline{U}_k}{\sigma_k}\right)}
                  {\sum_k^{Nu}D_{kx}},
    \intertext{Where:}
      \begin{tabular}{>{$}r<{$}@{\ :\ }l}
        P_{xi} & is the predicted rate for user~$x$ on item~$i$ \        S_{ki} & is the rate of song~$i$ given by user~$k$ \        D_{kx} & the correlation between user~$x$ and user~$k$ \        \overline{U}_x & the average rate over user~$x$ \        \overline{U}_k & the average rate over user~$k$ \        \sigma_x & is the standard deviation of all the rates of user~$x$
      \end{tabular}\nonumber
    \end{gather}
\end{document}

and this is what I tried to make of it:

\begin{gather}
    dN/d(d_{p}) = kd_{p}^{\delta},
\intertext{Where:}
\begin{tabular}{>{$}r<{$}@{\ :\ }l}
 dN  : number of particles per unit water volume in the size range d p  to  [d p +d(d p )] \                                       
 k   :  (= constant) depending on the particle concentration \
 \delta   :   (< 0) descriptor of the distributions' spectral slope.  
\end{tabular}\nonumber
\end{gather}

If I put the "&", latex tells me, that I don´t have enough columns. Also the "\" is not showing up in red. I think the problem lies in the {>{$}r<{$}@{\ :\ }l} part. But I don´t understand what is going on in this.

Help will be very much appreciated.

Thanks

Answer 1

You have been bitten by the infamous \\ bug of the site, see Double backslashes disappear from code

You copied the code from

https://tex.stackexchange.com/a/53483/4427

which was affected by the bug. Here is the fixed code.

\documentclass{article}
\usepackage{amsmath}
\usepackage{array}

\begin{document}

\begin{gather}
      P_{xi}=\overline{U}_x+\sigma_x
             \frac{\sum_k^{Nu}D_{kx}\times\left(\frac{S_{ki}-\overline{U}_k}{\sigma_k}\right)}
                  {\sum_k^{Nu}D_{kx}},
    \intertext{Where:}
      \begin{tabular}{>{$}r<{$}@{\ :\ }l}
        P_{xi} & is the predicted rate for user~$x$ on item~$i$ \\
        S_{ki} & is the rate of song~$i$ given by user~$k$ \\
        D_{kx} & the correlation between user~$x$ and user~$k$ \\
        \overline{U}_x & the average rate over user~$x$ \\
        \overline{U}_k & the average rate over user~$k$ \\
        \sigma_x & is the standard deviation of all the rates of user~$x$
      \end{tabular}\nonumber
    \end{gather}
\end{document}

The fixed code for your attempt should be

\begin{gather}
    dN/d(d_{p}) = kd_{p}^{\delta},
\intertext{Where:}
\begin{tabular}{>{$}r<{$}@{\ :\ }l}
 dN     & number of particles per unit water volume in the \\
 \multicolumn{1}{c}{} & size range $dp$  to  $[dp+d(dp)]$ \\
 k      & (constant) depending on the particle concentration \\
 \delta & ($<0$) descriptor of the distributions' spectral slope.
\end{tabular}\nonumber
\end{gather}

Answer 2

Here is one way that might be useful:

Using the [t] option on tabular yields better spacing above the tabular as per campa's comment. However, according to egreg there is a potential problem with this affecting paragraph spacing but I was unable to see this problem so have posted: Unable to duplicate problem: Using [t] option on tabular affect paragraph spacing.

Code:

\documentclass{article}
\usepackage{amsmath}% http://ctan.org/pkg/amsmath
\usepackage{array}% http://ctan.org/pkg/array
\begin{document}
\[
    dN/d(d_{p}) = kd_{p}^{\delta},
\]
where:
\begin{tabular}[t]{r@{}l}
 $dN$     &: number of particles per unit water volume in the size range $d p$  to  $[d p +d(d p )]$ \

 $k$      &: (= constant) depending on the particle concentration \
 $\delta$ &: $(< 0)$  descriptor of the distributions' spectral slope. 
\end{tabular}
\end{document}

Answer 3

Another solution, based on framed and a description environment (more satisfying, semantically). Also, I think the d for the differential and the d in d_p are a somewhat confusing notation, so I decided to have the differential ‘d’ upright, with a macro found on this site

\documentclass{article}

\renewcommand*{\d}{\mathop{\kern0pt\mathrm{d}}\!{}}

\usepackage{enumitem, framed}
\newenvironment{myleftbar}{%
      \def\FrameCommand{{\hspace{0.8cm}\vrule width 0.5pt}\hspace{6pt}}%\
      \MakeFramed {\advance\hsize-2\width \FrameRestore}\noindent\hspace{- 0.4em}}%
     {\endMakeFramed}
\newenvironment{eqdescription}{%
\myleftbar
\llap{\makebox[1.4cm][l]{where: }}\vspace*{-\dimexpr\baselineskip + \topsep}\strut
\description[font=\normalfont, labelwidth=0.5cm, align=right, leftmargin=!, itemsep=0pt]
}
{%
\enddescription\vspace*{-\topsep}
\endmyleftbar}

\begin{document}
\[
    \d N/\d(d_{p}) = kd_{p}^{\delta},
\]

\begin{eqdescription}
\item[$\d N$:] number of particles per unit water volume\\ in the size range $d_p$ to $[d_p +\d(d_p )]$,
\item[$k$:] (= constant) depending on the particle concentration,
\item[$\delta$:]$(< 0)$ descriptor of the distributions' spectral slope.
\end{eqdescription}

\end{document}