Rewrite units section based on comment from review

Problem remains: can't get the acute accent over the
e to work (apparently; its not working on my desktop)
Need help from Latex experts!

acfits.bib

@@ -187,6 +187,14 @@ @INPROCEEDINGS{1995ASPC...77..229J
   adsnote =      {Provided by the SAO/NASA Astrophysics Data System}
 }
 
+@Misc{1995OGIPUnits,
+  author =       {{George}, I.~M. and {Angelini}, L.},
+  title =        "{Specification of Physical Units within OGIP FITS files}",
+  howpublished =
+                  {\url{http://heasarc.gsfc.nasa.gov/docs/heasarc/ofwg/docs/general/ogip\_93\_001/ogip\_93\_001.html}},
+  year =         1995
+}
+
 @INPROCEEDINGS{1997ASPC..125..257W,
   author =       {{Wells}, D.~C.},
   title =        "{Speculations on the Future of FITS}",
@@ -252,6 +260,14 @@ @INPROCEEDINGS{2000ASPC..216..243G
   adsnote =      {Provided by the SAO/NASA Astrophysics Data System}
 }
 
+@Misc{2000CDSUnits,
+  author =       {{Ochsenbein}, F.},
+  title =        "{Standard for Documentation of Astronomical Catalogues}",
+  howpublished =
+                  {\url{http://vizier.u-strasbg.fr/doc/catstd.htx}},
+  year =         2000
+}
+
 @ARTICLE{2001A&A...376..359H,
   author =       {{Hanisch}, R.~J. and {Farris}, A. and {Greisen},
                   E.~W. and {Pence}, W.~D. and {Schlesinger},

acfits.tex

@@ -994,33 +994,49 @@ \subsection{Data Quality}
 
 
 \subsection{Units}
-
-
-The FITS support for units is syntactically flexible (albeit with
-a few specification ambiguities which would be resolved by an explicit
-grammar), but the model does not accommodate the full range of
-contemporary astronomical data.  This is evident from the significant
-fraction of unit strings in current databases which do not conform to
-the FITS prescriptions.  The solution is not simply to expand the list
-of recommended units since, as well as being slow, this fails to
-distinguish between, for example, different definitions of the second,
-or to communicate places where the distinction does or does not
-matter.
-
-There is also a provenance issue to defining new units, since -- to
-pick the example of the unit of `Jupiter radius' -- two different
-groups may prefer the mean or equatorial values for the radius, or in contrast
-may regard it, not as simply an abbreviation for a certain number of
+\label{sec_units}
+
+A strength of FITS is that includes support for units within its core
+standard. There are, however, limitations in the utility of the provided
+specification.
+
+First, while it syntactically flexible, there are a few
+specification ambiguities which could be resolved by an explicit grammar.  
+This limitation has perhaps been one reason that others
+have felt the need to publish more explicit prescriptions for units
+\citep{1995OGIPUnits}.  Another limitation is that the model does   
+not accommodate the full range of contemporary astronomical data. This
+is evident from the adoption of other units systems by some major
+archives such as the CDS (Centre de Données astronomiques de Strasbourg,
+\url{http://cds.u-strasbg.fr/}) which contains a large number of published 
+astronomical tables (see the units section within \citealt{2000CDSUnits}).  
+Finally, there is also a provenance issue to defining new units, since 
+-- to pick the example of
+the unit of `Jupiter radius' -- two different groups may prefer the
+mean or equatorial values for the radius, or in contrast may regard it,
+not as simply an abbreviation for a certain number of
 kilometers, but instead as a distance whose value is determined at a
 certain atmospheric pressure level.
 
+The solution to these limitations is not simply to expand the list of
+recommended units since, as well as being slow, this fails to distinguish
+between, for example, different definitions of the second, or to communicate 
+places where the distinction does or does not matter.
+
 The purely syntactical issues surrounding unit strings are
 being addressed by the IVOA's `VOUnits' work \citep{VOUnits}, but the higher
 level questions -- of communicating and defining new units, of
 indicating documentation, and of converting between them in a
 scientifically meaningful manner -- are out of scope for that work by
 design, since experience has shown them to be more contentious than
-one might expect.  These questions should perhaps be taken up by the FITS community.
+one might expect. These questions should be taken up by the FITS community.
+
+Solutions then may involve cherry-picking VOUnits syntactical fixes and 
+alteration of the units model to use some namespacing mechanism which 
+would help to disambiguate sources of extended units models found within a file.
+Finally, we believe that an analysis of other more recent work, such as that
+found in \citeauthor{VOUnits} and \citeauthor{2000CDSUnits} can help to quickly
+round out the roster of standard units.
 
 
 \section{Metadata and data representation}