Error column in VOResource

Ed Shaya Edward.J.Shaya.1 at gsfc.nasa.gov
Wed May 19 06:13:24 PDT 2004



Roy Williams wrote:

>>If you attach all the metadata to every field element, it expands from
>>just a few characters "<td> </td>" to a hundred or more.
>>    
>>
>
>Its not just that, there is also a structural problem.
>
>VOTable as it stands is a *generic* container. A single tool can handle any
>table (think of VOPlot or Excel). But if every element is part of a specific
>markup, then we must have a different software implementation for each
>different table header.
>
>
>  
>
I would rather not think of Excel, thank you.
VOplot would be more powerful if/when it  understands an SED and 
instantly plots it
with both vertical and horizontal error bars and can call up metadata  
on any datapoint when you click on it (including filter name and 
response function, telescope, observer, date, and aperture size) and can 
provide either frequency or wavelength axes.  Maybe it will handle 
addition of the users data and write out a new SED.  Likewise for each 
of the half dozen or so DM models that are envisioned.

For dumb tools that only know VOTable there can always be a generic XSLT 
to transform a DM object into VOTable.  There would usually be some loss 
of information so it is not reversable but this is good enough for the 
purpose.

Ed





More information about the registry mailing list