[Esip-ejournal] FW: ESIP Electronic Data Journal

[Howard Burrows]

Menas and all,
 
Is it useful to talk about two different "journals"?  One is the
twice-a-year commentary on datasets as discussed by Bob Downs and Bruce
Barkstrom.  The other is a persistent data citation that would allow
academic credit for datasets.
 
It was in reference to this latter that I had mentioned the idea of a
registry.  Gene Majors alerted me to the DoI literature and the existing
efforts within GCMD in this direction.  I particularly recommend the article
by Paskins listed at http://www.handle.net/papers.html .  In the context of
Digital Rights Management, Paskins reviews the relation between persistent
citable online references and essential metadata that makes them useable
across interests.  
 
I was thinking that the citation requirement could be met by registry
efforts like UDDI or NAIC as cited under the NASA Taxonomies project at: 
https://partners-lib.jpl.nasa.gov/partners-lib/
 
Howard
603-868-3221
email: ghburrows at attbi.com
web: http://www.ausi.org <http://www.ausi.org/>  
 
-----Original Message-----
From: Howard Burrows [mailto:ghburrows at attbi.com] 
Sent: Tuesday, March 25, 2003 3:34 PM
To: 'Menas Kafatos'; 'Vanessa Griffin'
Cc: 'Robert Benada'
Subject: ESIP Electronic Data Journal
 
Dear Menas and Vanessa,
 
The question I raised concerning the relation of this "journal" to a
registry or a clearinghouse came up to satisfy the role for a persistent
data citation.  There seems to be a hierarchy of services that we could
provide.  The simplest, requiring only a persistent server, would be to
establish and maintain a fixed reference link to allow people a permanent
citation for data.  This would establish a URI for data and could be
flexible enough to handle combination and segmentation of datasets.  At this
level, there need be no quality control other than frequent confirmation
that the dataset has not relocated.
 
A clearinghouse would provide more, and could be built on top of the
registry.  This is in the realm of a metadata repository.  In our case, the
metadata would concern requirements that any URI in the registry has met.
We could provide categories and even "registries" of different requirements,
including review by some blue ribbon panel of experts.  It would also be
good to have less social metrics for assessing data.  It is important that
the URI not be "overloaded" with semantics such as its science domain.
 
There are many additional levels to this hierarchy.  For example, we could
set up a registry for tools to process data or a clearinghouse to link
specific data to science questions, concepts, and theories, as in the NASA
roadmaps.  Depending on the domain, many new technologies exist to help.  As
noted in the discussion, this might have the form of a traditional review
article, but much more dynamic and lively forms could be envisioned using
such tools as the wiki or methods of topic or concept mapping.
 
I look forward to more detailed discussions of this project.  Let us look
for the simplest first step, cost it out for persistence, raise a proper
endowment or commitment from a stable organization to provide credibility
for that persistence, and then promote it in those developer and user arenas
that can make it effective.
 
Howard
603-868-3221
email: ghburrows at attbi.com
web: http://www.ausi.org <http://www.ausi.org/>