Coordinates model - Working draft.

Rob Seaman seaman at lpl.arizona.edu
Fri Jan 11 20:50:04 CET 2019


You guys squabbling make me feel young again ;-)

Note that commercial reference clocks are themselves pretty far from
ISO-8601, with settings for IEEE-1344, etc.

Is there a persuasive argument for why IVOA shouldn't mandate a binary
timestamp standard? For that matter, the FITS language is mostly driven
by Y2K era ASCII keywords. Nothing about FITS forbids encoding metadata
into a binary table structure (let alone VOtables). What would minimal
best usage be for timing metadata in a binary format? Could consensus be
reached if IVOA just ditched the grotty ASCII / UTF-8?

Rob

--

On 1/11/19 12:30 PM, Arnold Rots wrote:
> And this is precisely the part of ISO 8601 that does not make sense in
> astronomy: the exclusive choice between a time zone and UTC. Hence,
> the community has accepted a limited version of the ISO 8601 value
> string which is required to be associated with a specification of the
> time coordinate frame (including a time scale). 
> See FITS WCS Paper IV.
>
> Cheers, 
>
>   - Arnold 
>
> On Thu, Jan 10, 2019, 17:11 Graham, Matthew J. <mjg at caltech.edu
> <mailto:mjg at caltech.edu> wrote:
>
>     The ISO 8601 spec is very clear: "The   zone    designator     
>     is      empty   if      use     is      made    of      local 
>      time    in      accordance      with    4.2.2.2 through 4.2.2.4, 
>           it      is      the     UTC     designator      [Z]     if
>     use  is      made    of      UTC     of      day     in     
>     accordance      with    4.2.4   and     it      is      the   
>      difference-component    if      use     is      made    of     
>     local   time    and     the     difference      from    UTC   
>      in      accordance      with 4.2.5.2.” If you don’t use the zone
>     designator as in the standard then it’s not ISO 8601.
>
>     — Matthew
>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://mail.ivoa.net/pipermail/dm/attachments/20190111/417021f6/attachment.html>


More information about the dm mailing list