[Re: VOunits draft] comments

[Anita M. S. Richards]

Dear all,

Here are some responses from the editors to the comments on the Units 
draft.  Many thanks to everyone who has read it so carefully.  We will try 
and produce an updated draft including the points where there is 
agreement.

1) cgs

Please note that section 4 is concerned with IVOA conventions for
internal unit handling. If we adopt SI internally, that does not have
any impact on data providers who publish in cgs, except that it will
become easier for astronomers who use SI units to access cgs data
(and v.v.).  Insofar as conversions are unambiguous, unconditional and
straighforward to implement, we envisage that tools provided by the VO
(using existing libraries as far as possible) will be used to convert
units as part of a seamless chain of operations.  For example, I have
a data set in cgs.  I publish it using VO tools which convert the
coarse-grained coverage information to SI for registration purposes (I
can check this or take it on trust).  A user can then search for data
in any recognised system.  When they have chosen some data sets and
want to make a detailed request, in some cases they may have to use
the native units and get data back therein; in other cases conversion
may be possible at either or both stages (see example 5.3 in the
document).

So, please note Section 6 - the model is not seeking to 'redefine or
forbid' current usage of units, but to support services which will be
more effective if they can perform unit conversions (and to provide
standards for those which can already).

(although, personally, I agree with Francois O. that we should encourage 
people to adhere to IAU standards in adopting SI - but not force them!)

2) MJD
Sorry, I did not make myself clear enough.  Data providers *do* use
MJD as a unit _label_.  This illustrates why the VO needs to be able to
understand and translate native usage of units into a standard.  Of
course Arnold is right, but I think that he has missed the
point. Section 4.1 example 6 was designed to illustrate a possible
ambiguity which a unit parser might encounter whilst looking for SI
prefixes.

3) distinguishing between meter for wavelength and meter for distances
Point 3 of Section 2.1

This may well be covered by Quantity, but Units needs to be able to do
this.  Perhaps in Section 3.2 we could be more explicit about what
will be reused from Quantity, if it is suitable.  Certainly, we should
not do the same thing in two different ways and thanks very much to
Herve for pointing this out.

4) Casacore
Yes, that could be added to the list of libraries - although I am not
quite sure whether that or the aips++ toolkit equivalent is the most
appropriate reference. (as a CASA user, my limited experience with
TAQL is not very happy, but the units part seems OK).

5) Suggested unit syntax

Thanks to everyone who has pointed out deficiencies.  The proposals
are certainly open to modification.  However, we may end up with rules
which can cope with 99% of cases, but will need individual rules for
the remaining few.

6) FITS WCS Paper I

Apologies Arnold, that reference was in an earlier internal draft but
was inadvertantly omitted, sorry I did not spot that in proofreading.
We will certainly to check for the FITS convention for units chaining 
and look at using those conventions.

7) Translation e.g. for HEASARC units

If necessary, customised translation layers can be prepared for
(preferably by :-) ) major data centres who have evolved their own
vocabularies (this might overcome the problems Mireille has noted
otherwise).  Indeed, if any data provider has an idiosyncratic use of
units, we can prvide a tool for them to establish a customised
translation, just as with other data publication tools.

best wishes

Anita and Mireille

- - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - -

Dr. A.M.S. Richards, UK ARC Node/AstroGrid,
Jodrell Bank Centre for Astrophysics, Alan Turing Building, University of 
Manchester, M13 9PL
+44 (0)161 275 4124
and
MERLIN/VLBI National Facility, Jodrell Bank Observatory, Cheshire SK11 9DL, 
U.K. +44 (0)1477 571321 (tel) 571618 (fax)

"Socialism or barbarism?" Rosa Luxemburg (1915)