TODO-MM.html

<!doctype html public "-//w3c//dtd html 4.0 transitional//en">
<html>
<head>
   <meta http-equiv="Content-Type" content="text/html; charset=utf-8">
   <title>MM's TODO list for R</title>
</head>

<h1 align=center>Martin Maechler's TODO list for R</h1>
<p>
<i>This page contains (some of) my plans for enhancements to
   <A HREF="www.R-project.org">R</A> ... with absolutely no warranty ...
</i>
<BR> You might also be interested in <A HREF="TODO-MM-nolonger.html">things
 <em>no longer</em></A> on the current list below.

<p>Martin Maechler
<tt><a href="mailto:maechler@r-project.org">maechler@r-project.org</a></tt>
</p>

<hr>

<h3>Shorter term </h3>
<!-- ----------------------------------- -->

<ul>
<li> Improve  pair()  to allow spreading over a few pages
     ("wall paper" :-)  {a bit as <tt>lag.plot()</tt> (from the
     <tt>ts</tt> package) does}.

     Improve it along the lines of Catherine Hurley  (even much easier)

<li> plot.factor() should  *not* do boxplots by default when
     the boxplot contains less than or only about 5-7 observations!

<li> Extend  box() to also accept 'bg' and draw backgrounds using GRect().
     box(bg="green3") will then be a higher level alternative to
     (code from <tt>demo(graphics)</tt>!)
     <pre>
     usr <- par("usr")
     rect(usr[1], usr[3], usr[2], usr[4], col="green3")
     box()
     </pre>

<li> The dendrogram class {package "stats"} needs better and more versatile
     methods,  particularly plot.*().  <br>
     One application is for the  <A
     HREF="http://cran.r-project.org/src/contrib/PACKAGES.html#VLMC">
     VLMC </A> package available from <A
     HREF="http://cran.r-project.org">CRAN</A>.  Here, do
     as.dendrogram.VLMC() properly and implement plot.VLMC() basically
     doing plot.dendrogram( as.dendrogram.VLMC( * ) ).
     <br> <em>(done in parts for R 1.9.0; more todo)</em>

<li> Minimum Spanning Tree: two versions (maybe in one function):<br>
  <ol> <!--- `Enumerate' : -->
  <li> mstree() using <em>Euclidean</em> distance and the usual n x p data
       matrix.  Robert has a simple <tt>mstree()</tt> in his unofficial
       Genecluster package in bioconductor {though lacking Friedman/Rafsky's
       planing option}  which should be easy to put into 'stats' as is.
  <li> I really want to compute the general MST problem (i.e. given an
       arbitrary graph with weighted edges, find a subgraph which is a tree
       with minimal edge weight sum).  The typical application I have in mind
       would still be graphs with many edges, i.e. <em>almost complete</em> ones.
  </ol>


<li> procrustes() is wanted, but not just the two sample case,
     and (eventually at least) also with case and variable weighting.

<li> Extend the <A HREF="http:Rds.html">Guidelines for Rd files</A>,
     (say more about <tt>\item</tt> in <tt>\arguments</tt>,
      also recommendation/examples about the usage of <tt>\describe</tt>)
     and think about merging (or moving) parts into the
     <em> Writing R Extensions </em> manual which is part of R, or
     available from the
      <A HREF="http:///CRAN.R-project.org/manuals.html">CRAN Manuals</A>.

<li>  cov.wt() should also work with NA's, with a `use' argument identical
      to  cov(), cor();
      Using a notion of "generalized ranks", maybe could also grow  "spearman"
      and "kendall" methods.

<li>  <tt>str()</tt> for lists (and data.frames) should get an option to allow
      <em>numbering</em> the list components (maybe via a smart
      specification of the <tt>indent.str</tt> argument).

</ul>


<h3> Of less priority </h3>
<!-- ----------------------------------- -->
<ul>

<li> Need C API for the Median() and more generally  quantile(x, alpha).
    <em><font=-1> (not sure anymore; we have a C API to partial sorting)
        </font>
     </em>
<li> <tt>stem()</tt> should exclude outliers and list them separately
  (as S does).

<li> New print option for printing exact "0"s:<br>
    Consider using <tt>"."</tt> instead of <tt>"0"</tt> in numerical
    matrices. This would make sparse matrices much more readable.
    <small>[I have been doing this with S-plus for years.]</small>

<li> scat1d() {improved rug(); the thing I've co-authored and has then
	       been put into Hmisc}.

</ul>

<h3> Even Longer Term </h3>
<!-- ----------------------------------- -->
<ul>

<li> Differential Equation Solvers; I need a boundary value problem solver,
    additional to the LSODA functionality of the very good
  <A HREF="http://cran.r-project.org/src/contrib/PACKAGES.html#odesolve">
  odesolve </A> package available from
  <A HREF="http://cran.r-project.org">CRAN</A>.
</ul>

<HR><!---------------------------------------------------------------->
<P>
Return to the <A HREF="http://developer.R-project.org">R Developer Page</A>.
</P>

Last update at $Date: 2004/06/04 17:42:50 $ UTC (aka GMT)
<address>
 <A HREF="http://stat.ethz.ch/~maechler/"> Martin Mächler </A>,
 Seminar für Statistik, <A HREF="http://www.ethz.ch/"> ETH Zurich</A>,
 Switzerland.

</body>
</html>