String representations of numeric values
Anita Richards
amsr at jb.man.ac.uk
Fri Apr 21 03:32:20 PDT 2006
We have to live with VOTables containing sexagesimal - sorry - the best
that we can hope for is that new data contains decimal degrees as well.
However, vast amounts of vital archive data are in sexagesimal (and/or
B1950).
I thnk that Vizier cope with this by more or less automatically adding
columns in decimal degrees, J2000.
The problem is also of course that the delimiter can be space or : and the
+ is optional but - isn't.
Moreover, I believe that I have seen formats like
12 34 56 +78 9.1
etc. - but maybe those are rare enough not to worry
But part of the reason why e.g. I use TopCat and not excel (apart from
being allergic to windows) is that the former understands astronomical
coordinates. Fundamentally, the format is a limited problem; the more
important issue is dealing with transormations with the right amount of
accuracy - or indeed any data; I ahve frequently run into problems with
milli-arcsec precision data, that conversion to decimal radians inside a
program introduces errors because someone has made an inadequate
assumption about the number of sig. fig. Thre is also the quesion of what
algoritms to use, wrapping etc.
Sorry to go off the plot a bit, but the bottom line is that we are stuck
with sexagessimal and the standards for the display format are trivial
compared with the trig!
cheers
a
- - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - -
Dr. Anita M. S. Richards, AstroGrid Astronomer
MERLIN/VLBI National Facility, University of Manchester,
Jodrell Bank Observatory, Macclesfield, Cheshire SK11 9DL, U.K.
tel +44 (0)1477 572683 (direct); 571321 (switchboard); 571618 (fax).
On Fri, 21 Apr 2006, Alasdair Allan wrote:
>
> Roy Williams wrote:
>> Who is writing VOTables with sexagesimal in them?
>
> I frequently see these, and most of my software has had to be hacked to deal
> with them. They're quite common, although I can't remember the worst
> offenders off the top of my head...
>
> Al.
>
>
More information about the votable
mailing list