[Esip-documentation] ACDD questions

[John Graybeal]

On Apr 19, 2013, at 13:33, Derrick Snowden - NOAA Federal <derrick.snowden at noaa.gov> wrote:

> I also think we need another page that serves as the versioned list of attributes (if it's not on the list, it's not part of the convention) and that each attribute contain a well written clear definition.  I think the definitions we have now need to be improved. 

As I had a fair number of comments on the last set of definitions, I volunteer to produce a first revision (for discussion of course!) of any term definitions you want me to.

+1 on multiple keyword vocabularies. The nice thing about being able to use a recognizably unique code (e.g., a URI) is that the provider can choose what level to provide the keywords, and for what topics, and with what vocabularies. I think this will result in more, and more accurate, keywords than if we constrain it.

John

On Apr 19, 2013, at 13:33, Derrick Snowden - NOAA Federal <derrick.snowden at noaa.gov> wrote:

> ---------- Forwarded message ----------
> From: Derrick Snowden - NOAA Federal <derrick.snowden at noaa.gov>
> Date: Fri, Apr 19, 2013 at 4:02 PM
> Subject: Re: [Esip-documentation] ACDD questions
> To: ngalbraith at whoi.edu
> 
> 
> Hi Nan,  
> 
> I hope this is the place to discuss ACDD because it looks like the conversation is beginning.  Some comments interspersed below.
> 
> On Fri, Apr 19, 2013 at 3:39 PM, Nan Galbraith <ngalbraith at whoi.edu> wrote:
> Hi all - 
> 
> Is this now the right list to discuss the ACDD? I'm highly motivated to
> confirm some details and definitions, since I'm working on an update 
> to the oceansites format specification, and we're trying to adopt ACDD.
> 
> And, what is now the definitive page describing ACDD? The geo-ide page 
> includes a field that's not on the unidata page, geospatial_bounds. 
> I'm not sure if that's intended to be in addition to the min/max 
> lat/long attributes, or an alternative.  The geo-ide page is at
> https://geo-ide.noaa.gov/wiki/index.php?title=NetCDF_Attribute_Convention_for_Dataset_Discovery
> and I guess I'd like to know if that has been (or is going to be) replaced 
> by something on the ESIP site?
> 
> I would like to develop a new landing page that includes some introductory material discussing the purpose of ACDD and perhaps some discussion of how things work.  The current page on the NOAA wiki immediately launches into mappings which I agree is important, but shouldn't be the first page people see.  I also think we need another page that serves as the versioned list of attributes (if it's not on the list, it's not part of the convention) and that each attribute contain a well written clear definition.  I think the definitions we have now need to be improved.  By committing to using ESIP for this conversation, I think we're committing to using the wiki as the definitive source of ACDD information.  
> 
> Ted, is there any reason we cannot begin creating pages now?  Anything we should know before we start about ESIP wiki conventions?  I wouldn't want to come in to our new environment and start messing up a well laid out plan for the wiki.
> 
>  
> 
> Has anyone who's using a keywords attribute figured out a decent scheme? 
> I'm going to assume that  we're talking about describing the 'science' data 
> with these, not data sources or feature types.
> 
> I can't say we've found a decent scheme but I will say that I don't want to limit ourselves too much. I'd like to be able to include keywords for things like platform type and feature type since I feel like it's an element of discovery.  I think the challenge is to find a way to associate multiple "keyword_vocabulary"s.  I think that ncISO currently pulls out CF standard names from the variable attributes and includes them as keywords in the transformed ISO record.  Including them again as keywords seems like unnecessary duplication.  I can see a need for including multiple keyword vocabularies.  IOOS has been publishing vocabularies and maps between existing vocabularies at MMI.  I'd like to be able to reference more than one of those.  It's not clear how to do that yet.
>  
> My question is whether these should be listing all the data variables in a file, 
> or the general field of the data - e.g.
>  ''air_temperature, relative_humidity, rainfall_rate..."
> vs
> "OCEANOGRAPHY: GENERAL: Marine meteorology" (or just "Marine meteorology")
> 
> Last point - I notice that the NetCDF-U people are identifying files that follow their 
> spec (for uncertainty)  using a more standard approach than ACDD; they add 
> UW-1.0 as a term in the Conventions attribute.  Should we consider this too, in 
> place of the Metadata_Conventions attribute?
> Conventions = "CF-1.5 UW-1.0 ACDD-1.0"
> 
> I like the clarity of this approach.   
> 
> ACDD is certainly not the only metadata convention, and NUG recommends putting
> all the conventions in one attribute.
> 
> Thanks - Nan
> 
> 
>> Our discussion today about mailing list names has borne fruit!
>> 
>> Great work and thanks to our ESIP colleagues!
>> 
> 
> -- 
> *******************************************************
> * Nan Galbraith        Information Systems Specialist *
> * Upper Ocean Processes Group            Mail Stop 29 *
> * Woods Hole Oceanographic Institution                *
> * Woods Hole, MA 02543                 (508) 289-2444 *
> *******************************************************
> 
> 
> 
> _______________________________________________
> Esip-documentation mailing list
> Esip-documentation at lists.esipfed.org
> http://www.lists.esipfed.org/mailman/listinfo/esip-documentation
> 
> 
> 
> 
> -- 
> Derrick Snowden
> System Architect, US IOOS 
> 1100 Wayne Ave, Suite 1225
> Silver Spring, MD 20912
> +1 301 427 2464 (o), +1 301 427 2073 (f)
> 
> Find us on Facebook
> 
> 
> 
> -- 
> Derrick Snowden
> System Architect, US IOOS 
> 1100 Wayne Ave, Suite 1225
> Silver Spring, MD 20912
> +1 301 427 2464 (o), +1 301 427 2073 (f)
> 
> Find us on Facebook
> _______________________________________________
> Esip-documentation mailing list
> Esip-documentation at lists.esipfed.org
> http://www.lists.esipfed.org/mailman/listinfo/esip-documentation


---------------
John Graybeal
Marine Metadata Interoperability Project: http://marinemetadata.org
graybeal at marinemetadata.org




-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://www.lists.esipfed.org/pipermail/esip-documentation/attachments/20130419/d50484dc/attachment-0001.html>