[Fwd: VOTimeSeries]

[Francois Ochsenbein]

Rob,

My point was not to say that the document had values outside some
boudaries, but that the metadata look inconsistent:

==> in the TimeObs, the value is a number of days since TimeZero.
    The comment says that the value is an MJD value . 

==> in the VOTable, the unit part of the Time field specifies that
    the units are 's' (seconds). But the data values (1305.65557 etc)
    are most likely days elapsed since TimeZero.

Sorry to insist on this, but the reliability of the metadata is the
most important part of the VO. Unreliable data are of little use. 
Reliable data with unreliable metadata can obviously lead to crazy
results in the context of the VO.

--Francois

>
>Francois replies to Roy:
>
>>> QUESTION: What semantic information is MISSING from this time series?
>>> QUESTION: Is there TOO MUCH metadata here, is it too COMPLEX?
>>> QUESTION: What other syntax could be used besides VOTable?
>>
>> Just take care of the comments -- TimeObs tells it's the "Modified
>> Heliocentric Julian date of mid-exposure" but the value indicates
>> 1305, corresponding to 14 June 1862, 145 years ago...
>
>Is bounds checking to be part of the DM?  There is nothing
>intrinsically incorrect about conveying a nineteenth century
>dataset.  The magnitude scale (and hence the potentiality of a light
>curve) originated with Hipparchus.  De Stella Nova expressed in 1573
>that stars vary in brightness.  Anybody have a candidate for the
>earliest published light curve or time series?
>
>> There seems also to be some contradictions in the time units --
>> TimeStart and TimeDate are indicated in 'd' (days),
>
>The clock is a subdivision of the calendar, so time can be given in
>units of days.  (This is the crux of the ongoing UTC leap second
>debate.)  The bigger issue is that date and time must form an atomic
>whole.  Perhaps the best way to eliminate this issue is to avoid
>"time" fields entirely in favor of "date" analogs as with FITS DATE-OBS.
>
>> TimeObs has no unit (but a wrong comment); the time series itself
>> indicates 's' as units for Time (top table), but the values given
>> for the time in the examples (middle table) seem to be days elapsed
>> since the TimeZero.
>
>For some reason, perhaps because clocks and calendars seem "obvious"
>while telescope setting circles do not, self-consistent time usage
>continues to elude us even as esoteric WCS usage proves tractable.
>
>My own two cents is to second Doug's comment:
>
>>>> Since the data model is defined independently of the
>>>> serialization, there is no need to pick a single serialization.
>
>Which I understand to imply that VOEvent - an XML publication
>standard - might settle on a distinct time series representation from
>other VO applications relying on CSV, FITS, native VOTables, or
>"graphical rendition" (e.g., IRAF GKI metacode, I suppose).
>
>Rob
>seaman at noao.edu
================================================================================
Francois Ochsenbein       ------       Observatoire Astronomique de Strasbourg
   11, rue de l'Universite F-67000 STRASBOURG       Phone: +33-(0)390 24 24 29
Email: francois at astro.u-strasbg.fr   (France)         Fax: +33-(0)390 24 24 32
================================================================================