On 3/15/11 3:20 PM, Bob Simons wrote:
> * Isn't part of the problem related to udunits having single values for
> the length of a month and a year,
no, the problem is that there is no such thing as a "month" as a
standard length of time, so it shouldn't be used that way.
> In that spirit, I encourage
> support for something like "calendar_months since" and "calendar_years
> since" for data related to monthly or yearly data.
To me, the primary reason to use a "time-delta since" structure is that
you get an array or values that is continuous and meaningful -- you can
directly use it to do things that Steve suggests:
Calculate integrals, derivatives, smoothing
use as the X-axis on a plot
etc.
All that requires that the time unit be something consistent. a calendar
month is not a consistent unit of time.
if you have data that is associated with a calendar month, then I think
it should be labeled with:
January, Feb, etc.
It's very common to have tables of data that may be irregularly spaced
have a field(s) for the date-time -- why not use that approach?
-Chris
--
Christopher Barker, Ph.D.
Oceanographer
Emergency Response Division
NOAA/NOS/OR&R (206) 526-6959 voice
7600 Sand Point Way NE (206) 526-6329 fax
Seattle, WA 98115 (206) 526-6317 main reception
Chris.Barker at noaa.gov
Received on Tue Mar 15 2011 - 16:47:58 GMT
