Re: [netcdf-java] [thredds] GRIB variable name changes in 4.3

I was wondering about the how these proposed changes impact not just
variable descriptions but other parts of the encoding that rely on tables?
In particular, how will the grid description be handled? In edition 2 this
requires code templates in addition to lookup tables, how will these be
managed with the same issues in mind?


On Thu, Mar 1, 2012 at 6:31 AM, Don Murray <don.murray@xxxxxxxx> wrote:

> All-
> On 2/29/12 1:27 PM, John Caron wrote:
>  I think you have to keep in mind that the proposal is to use the long
>> name for human selection, and the variable name for machine selection.
>> If you look at this:
>> float Temperature(time=1, lat=361, lon=720);
>> vs
>> float VAR_0-0-0_L6_I6_Hour_S194(**time=1, lat=361, lon=720);
>> :long_name = "Temperature (6_Hour Average) @ Maximum wind level";
>> I think you see that the long name is better at conveying all the info a
>> human needs.
> I agree that the long_name should contain the most descriptive form of the
> variable.  However, I think that a shortened form of this which has been
> available in 4.2 and the earlier version of 4.3 can convey just as much
> information.  Doesn't the attached list that you sent me convey all the
> information in VAR_0-0-0_L6_I6_Hour_S194 + attributes?
>  So the subset service, the dods html page, and all other UIs will want
>> to change to that. So the answer is that Hal must serve both we the
>> humans, and our overlords, the machines, who, I for one, welcome ;^)
> And, therein lies the problem. ;-)
>  PS: The VAR names might change, but only when we spectacularly
>> misunderstand the GRIB spec, eg like we did for time intervals. They
>> will be impervious to the minor changes that the tables go through.
> How often do the tables change? Daily, monthly, yearly?  I share Glenn's
> perspective that the rate of change is slow and mostly in local variations
> of the tables.  Do the CFSR tables change continually?
>  PPS: We are going to move this discussion to a new list or something, to
>> minimize the spam.
> Isn't it best to get input from all the users?
> Don
> --
> Don Murray
> 303-497-3596
> _______________________________________________
> thredds mailing list
> thredds@xxxxxxxxxxxxxxxx
> For list information or to unsubscribe,  visit:

Paul Hamer

NOAA Earth System Research Laboratory (ESRL)
Global Systems Division (GSD)
325 Broadway R/GS5
Boulder, CO 80305-3328

phone : 303.497.6342
  • 2012 messages navigation, sorted by:
    1. Thread
    2. Subject
    3. Author
    4. Date
    5. ↑ Table Of Contents
  • Search the netcdf-java archives: