[eml-dev] obsolete EML tags ?

[Corinna Gries]

Inigo,

You are really in EML bashing mode, aren't you?

I would like to rescue a few of these tags here. I am using the <view>
tag and actually really need it to break down large databases into
reasonable datasets. And when we get into these ODMs and variants
thereof, the view will be the way to define a dataset in EML.
I have all our protocols in <protocol> and if we ever get to some sort
of LTER protocol catalog/registry that should be the way to go. I think
this as well as <software> is not used much because everybody is busy
(struggling) keeping up with the datasets - we'll get there.
I agree that <citation> needs some work and probably should become some
citation collection, i.e. allow many <citation> to be used in one file
to be useful. I think that has been discussed.

My two cents
Corinna

-----Original Message-----
From: eml-dev-bounces at ecoinformatics.org
[mailto:eml-dev-bounces at ecoinformatics.org] On Behalf Of inigo
Sent: Friday, September 19, 2008 9:20 AM
To: eml-dev
Subject: [eml-dev] obsolete EML tags ?



more for the record.

The use and adoption of the EML probably has not peaked yet, but one can

argue that is never to soon to revisit obsolete parts of the standard.

I am sure that long discussions and thorough design plans that were 
approved unanimously ensued when placing atop of the EML hierarchy items

like "software", "citation" and "protocol".  After some time,  people 
has used the EML in telling ways. 

In my view, the following are clear candidates to be deemed obsolete. 
<software> :  /eml/software  - No one at LTER+Pisco is using the 
"software" tag - top of the EML hierarchy-. 
<view>   :   /eml/dataset/view - One LTER site used in one instance the 
<view> tag - what's up guys,  no one willing to describe database views?
<storedProcedure> : /eml/dataset/storedProcedure.   No one is using this

at LTER+ObFS+Pisco.  If you need it, take it home, it's screaming for 
shelter.
<series> : /eml/*/series - This element may have been adopted via the 
FGDC - There is one LTER site that used it 14 times.
numerous elements within "physical". (size, maxRecordLenght, 
encodingMethod, numPhysicalLinesPerRecord,caseSensitive,aunthentication,

characterEncoding). - Some of these, NEVER used, and some sparingly
used.
some elements under <distribution>. there is talk to add <access> under 
here (or was it upstream?). However, how about deprecating those 
dinasour-age elements? compression... offline.. why tease people saying 
that you have data, but not willing to share it ? if you do not want to 
share it, why mention it. too large? no dataset is too large.  even 
Comcast still allows internet traffic of the order of 250Gb/month/user. 
/distribution//connection is not popular either. 2 sites use it 
sparingly, i actually helped promote the use of this particular element 
for one of these sites. but im open to change that..

Other elements may not deserve being deprecated, but certainly will not 
win any EML popularity contest: Here are some.

<protocol> : /eml/protocol/   Three (two of those coordinated from the 
same team) LTER sites use the "protocol" tag.   I know there was some 
discussion here to revamp this tag.  a chaneg now will not affect many, 
it seems.

<citation>  : /eml/citation/   Four sites use this, two of these 
sparingly.  This tag has been discussed, and there was some sound 
silence in eml-dev (surprise) but it would seem that a change (like the 
one pursued by Servilla and OBrien would do some good to some. 

Im sure there are more tags alike. we can discuss also the 
spatialVector/Raster complex groups. or not.

cheers, inigo


_______________________________________________
Eml-dev mailing list
Eml-dev at ecoinformatics.org
http://mercury.nceas.ucsb.edu/ecoinformatics/mailman/listinfo/eml-dev