[Esip-preserve] Need for recommended best practice for DOI minting?

[Ruth Duerr]

Dear ESIP'ers;

It looks like DOI's are well on the road to implementation by the data community.  While the ESIP citation guidelines do talk briefly about best practices in minting DOI's - specifically 

"Best practice is that the suffix of the identifier does not include a reference to the archive in case the data are moved from the original location where the persistent identifier was assigned initially."

there is no other guidance provided.  As a result a variety of implementations are being proposed by organizations like NASA, research groups, etc.  The question is whether ESIP should promote more uniform implementation by providing more detailed best practices?  

I note that many of the proposed implementations encode information in the DOI string itself - information which may change over time (e.g., mission or program name, instrument identifier, product name, etc.) leading one to expect that many data sets will end up with multiple DOI's over time.  

The EZID folks actually mint DOI's using an opaque identifier and recommend this as a best practice.  A quote from their documentation follows:

"Opacity

Many of the terms in this document serve semantic opacity, which is useful in creating identifiers that age and travel well ...  Perfect opacity in identifiers is not as important as identifiers' having semantics that are not widely recognizable, even if those semantics may support administrative activity by specialists (cf. ISBNs).  ...  For the purposes of longevity, however, it is critical that attachments to authority and sub-authority names ... be avoided; it is common for political pressures to require sacrificing identifiers created out of short-sighted administrative or branding convenience."

I also note that the DOI's generated by publishers are generally opaque in the sense described by EZID.  The first 5 DOI's from a query of Google Scholar for the phrase "Digital object identifier" were:

doi:10.1215/S0012-7094-90-06119-8
http://dx.doi.org/10.2481/dsj.4.12
doi:10.1214/07-AIHP146
doi:10.2307/2373173
http://dx.doi.org/10.3998/3336451.0003.204

These were from articles from 5 different journals!

What does the community think?  Is a more comprehensive best practices warranted?  If so, what should it say?

Ruth