VOunits draft

[Anita M. S. Richards]

Dear Steve,

Thanks very much for your careful reading of the document.

On Tue, 19 May 2009, Steve Allen wrote:
>
> I am mystified by the references to SLALIB.  That library does not
> include the algorithms which are currently approved by the IAU.
> As far as I know the relevant libraries which do implement the
> IAU 2000 recommendations are
>
> IAU SOFA
> http://www.iau-sofa.rl.ac.uk/
>
> USNO NOVAS
> http://aa.usno.navy.mil/software/novas/novas_info.php

Dear Steve,

I have quickly consulted (including with Mark Taylor, who has written some 
of the currently most widely-use VO applications which support coordinate 
conversion).  I think that SLALIB is widely used and regarded as a highly 
trustworthy package containing astronomical coordinate conversions amongst 
other things. SLALIB exists in FORTRAN (public use permitted) and C (use 
only by agreement with Pat Wallace); a java version called PAL exists, 
which seems to work pretty well, but it has not been tested anything like 
as rigorously as the FORTRAN/C versions.  It is possible that it might 
need minor updates (recent leap seconds?) - if anyone is aware of specific 
issues, they should inform us.

The SOFA libraries don't appear to cover quite as much ground as SLALIB 
(e.g. no FK5<->Galactic conversions), although they may have other 
features.

The main intention in referring to existing libraries was to illustrate 
that the VO can draw on current high-standard work and does not have to 
produce everything from scratch.

My feeling is that we should add your links to IAU SOFA and USNO NOVAS but 
Mireille has made what I think is an excellet suggestion, that the 
particular libraries should not be embedded and named in the document, but 
that the document should point to a web page where we can maintain 
up-to-date links to libraries which support aspects of the conversions 
required.

>
> I may be stealing the words of Arnold Rots when I write
> MJD is not a unit.  It is a measure of time which has
> an origin distinction akin to that of K or degC for temperature.

I apologise if the document is worded clumsily, as I remember Sebastien 
making a related point although the sutlety was lost on me.  However, MJD 
(like some aspects covered by ISO-8601) is a label for a unit of elapsed 
time.  We need to make sure that we can handle common use cases and that 
involves knowing the origin of coordinate systems.  Perhaps someone can 
come up with a more semantically correct term for unit labels like MJD.

> Does the VO effort intend to address the difference between
> meters and seconds in "TT units" and "TCB units"
> and "TCG units"?
> I.e., as seen on page 12 of IERS conventions 2003 (tech note 32)
> a quantity of length or time has a different numerical value "x"
> depending on the reference frame in which its value is expressed
> x_{TDB} = x_{TCB} * (1 - L_B)
> x_{TT}  = x_{TCG} * (1 - L_G)
> where L_G is defined as 6.969290134e-10
> and L_B is approximately 1.55051976772e-8
> and whereas the IAU recommends measurements be in TCG or TCB units,
> a common practice is to express them in TDB/TT units.

I think that if there is a use case and a suggested solution, that could 
be included, although the conversion rules are not for us to define (if 
possible!), but for us to point to libraries or quote where they are 
defined.

Best wishes

Anita

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

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)

On Tue, 19 May 2009, Steve Allen wrote: