EML2 for NTL-LTER meteorological data

[Matt Jones]

Hi Barbara,

Thanks as usual for being on top of things...

I looked over your document, and it looks great.  The only issues that 
came up were the ones we knew would be problematic because we have 
struggled with them.  I once again have a few comments:

1) Under abstract, you need not use a "section" element if you're only 
going to have one paragraph, although what you did is prefectly fine too

2) Your custom unit "wattsPerSquareMeter" uses a unitType 
(energyArealRate) which is undefined.  You need to define it before you 
can use it if it isn't in the eml-unitDictionary.  Plus, we would have 
probably named it differently if we put it in the unitDictionary -- 
rates generally have time in the denominator, so I probably would have 
called the unitType arealPowerDensity. See the definition in 
eml-unitDictionary of power, watt, and arealMassDensity for examples.

3) Your unit definition fro quantum is missing its base SI Unit and the 
conversion factors to SI.  The SI unit for energy is joule, so you need 
to provide the conversion information to calculate joules from quantum. 
See the example for 'calorie' in the eml-unitDictionary for an example.

4) Your unitType timeHHMM is problematic.  Its unitType is undefined, 
and would need to be.  In RC3, for datetime values, we had intended for 
you to use a unit "formattedDateTimeUnit" to provide the format, rather 
than a standard or custom unit.  We've struggled with date-time values, 
and thus have revised this part of the spec for the next release so that 
date-time values are no longer under "interval", but are treated as 
their own measurement scale called "datetime". Thus, datetime scales in 
the new system don't have units, but allow you to specify the format of 
the date or time value, and its precision and domain.  Maybe you could 
look over the new schema and tell me if it would work?  You can get it 
from CVS on the web: 
http://cvs.ecoinformatics.org/cvs/cvsweb.cgi/~checkout~/eml/eml-attribute.xsd?rev=1.96
Note that this is a development copy and differs from any released 
version of EML.  There is also a sample document that show's the use of 
this new schema (including datetime examples) if that would be of use to 
you.  It is at: 
http://cvs.ecoinformatics.org/cvs/cvsweb.cgi/~checkout~/eml/lib/sample/eml-sample.xml?rev=1.15

5) Your custom unit 'percent' is undefined. We've struggled a bit with 
dimensionless quantities and would like to hear your perspectives on how 
to best handle things like percent, ppm, ppb, etc.

6) Your customUnit 'date' is undefined.  Same issues as in (4) above.

7) Your attribute named 'YEAR4' is really a datetime value and should be 
described as such, not as 'number'.

That's about it.  I'm happy to say that all of the things that came up 
reviewing your document were the thorny issues that we think we've fixed 
for the final release. Thanks for all of your hard work on this.  Its 
valuable to us to have someone outside of the group developing EML to 
create a document like this to sanity check things.  Thank you,

Matt

Barbara Benson wrote:
> Hi Matt,
> 
> You wanted to know when we had EML for a dataset that passed the EML2 
> parser.  You can view the EML at
> http://lterquery.limnology.wisc.edu/eml2.jsp?dataset_id=NTLME01
> 
> I am going to talk with our PIs here about changing our data access 
> policy so that at least the metadata can be redistributed if we give 
> someone permission to do that.
> 
> We still have some issues with units (percent, number) but I'll deal 
> with those in a later email.
> 
> Let me know what you think of our EML document.
> 
> Barbara
> 
> Barbara Benson
> Center for Limnology, University of Wisconsin-Madison
> 680 N. Park St.
> Madison, WI 53706
> 
> (608)262-2573
> fax: (608)265-2340


-- 
*******************************************************************
Matt Jones                                    jones at nceas.ucsb.edu
http://www.nceas.ucsb.edu/    Fax: 425-920-2439   Ph: 907-789-0496
National Center for Ecological Analysis and Synthesis (NCEAS)

Interested in ecological informatics? http://www.ecoinformatics.org
*******************************************************************