Just to chip in...
> I think part of the problem is that when I see a date string I don't
> necessarily think of it as a point in time but rather a range. So, when
> I see "2005-08-19" I think of the whole 24 hour period rather than just
> the start of that day.
... and some visualization tools think the same way. I think that if
a time represents an instant it should appear as such - but what
precision is appropriate? Seconds are probably appropriate in most
cases, but (for example) Java "instants" have millisecond precision.
However, in the vast majority of cases the milliseconds will be zero
and there will be a lot of distracting ".000Z"s appearing in ncdump...
Cheers, Jon
On Wed, Aug 20, 2008 at 5:06 PM, Ethan Davis <edavis@xxxxxxxxxxxxxxxx> wrote:
> Hi Russ,
>
> I find the varying precision of the time strings a bit distracting and
> was wondering if it would make sense to set the precision for the entire
> array. Or another option would be to match the precision of the time
> strings to that given in the file. In this case, out to the second:
>
>> time:units = "days since 1858-11-17 00:00:00" ;
>
> I think part of the problem is that when I see a date string I don't
> necessarily think of it as a point in time but rather a range. So, when
> I see "2005-08-19" I think of the whole 24 hour period rather than just
> the start of that day.
>
> Ethan
>
> Russ Rew wrote:
>> [...] The current netCDF snapshot distribution supports a
>> "-t" option that displays times in a human readable ISO-8601 form, so if
>> you run "ncdump -t" on the netCDF file corresponding to the CDL you've
>> given, the output for the time variable is:
>>
>> time = "2005-08-18 18", "2005-08-19", "2005-08-19 06", "2005-08-19 12",
>> "2005-08-19 18", "2005-08-20", "2005-08-20 06", "2005-08-20 12",
>> "2005-08-20 18", "2005-08-21", "2005-08-21 06", "2005-08-21 12",
>> "2005-08-21 18", "2005-08-22", "2005-08-22 06", "2005-08-22 12",
>> "2005-08-22 18", "2005-08-23", "2005-08-23 06", "2005-08-23 12",
>> "2005-08-23 18", "2005-08-24", "2005-08-24 06", "2005-08-24 12",
>> "2005-08-24 18", "2005-08-25", "2005-08-25 06", "2005-08-25 12",
>> "2005-08-25 18", "2005-08-26", "2005-08-26 06", "2005-08-26 12",
>> "2005-08-26 18", "2005-08-27", "2005-08-21 12", "2005-08-21 18",
>> "2005-08-22", "2005-08-22 06", "2005-08-22 12" ;
>>
>> instead of
>>
>>
>>> time = 53601.75, 53602, 53602.25, 53602.5, 53602.75, 53603, 53603.25,
>>> 53603.5, 53603.75, 53604, 53604.25, 53604.5, 53604.75, 53605, 53605.25,
>>> 53605.5, 53605.75, 53606, 53606.25, 53606.5, 53606.75, 53607, 53607.25,
>>> 53607.5, 53607.75, 53608, 53608.25, 53608.5, 53608.75, 53609, 53609.25,
>>> 53609.5, 53609.75, 53610, 53604.5, 53604.75, 53605, 53605.25, 53605.5 ;
>>>
>>
>>
>>> float time(record) ;
>>> time:long_name = "Modified Julian Day" ;
>>> time:units = "days since 1858-11-17 00:00:00" ;
>>>
>
> --
> Ethan R. Davis Telephone: (303) 497-8155
> Software Engineer Fax: (303) 497-8690
> UCAR Unidata Program Center E-mail: edavis@xxxxxxxx
> P.O. Box 3000
> Boulder, CO 80307-3000 http://www.unidata.ucar.edu/
> ---------------------------------------------------------------------------
>
> _______________________________________________
> netcdf-java mailing list
> netcdf-java@xxxxxxxxxxxxxxxx
> For list information or to unsubscribe, visit:
> http://www.unidata.ucar.edu/mailing_lists/
>
--
--------------------------------------------------------------
Dr Jon Blower Tel: +44 118 378 5213 (direct line)
Technical Director Tel: +44 118 378 8741 (ESSC)
Reading e-Science Centre Fax: +44 118 378 6413
ESSC Email: j.d.blower@xxxxxxxxxxxxx
University of Reading
3 Earley Gate
Reading RG6 6AL, UK
--------------------------------------------------------------
