Re: "Accepted Manuscript"

[richards1000@comcast.net]

I agree with the excellent points made by Sally Morris, Greg 
Tananbaum, and Peter Hirtle.  To follow up on Sally Morris's 
important point respecting standards, perhaps it's worth 
considering how additional standards, namely OAI-PMH, FRBR, and 
rights management standards, could assist the scholar respecting 
these issues.

We've already discussed standards for identifying the different 
versions of a scholarly article.  Another relevant standard may 
be FRBR, the Functional Requirements for Bibliographic Records, 
http://www.ifla.org/VII/s13/frbr/ , and its machine-readable 
version, FRBR Object-Oriented, or FRBRoo, 
http://cidoc.ics.forth.gr/frbr_inro.html .

FRBR is a standard that permits machines to recognize that 
metadata describing all of the different versions of a piece of 
scholarship relate to a single "work."  FRBR then enables the 
reader-friendly organization of all of that descriptive metadata. 
Specifically, FRBR calls for the creation of a single "master" 
record to represent the work as a whole, with subordinate 
metadata records describing each version.

The single "master" record is the first record that the user 
sees; when the user is ready, he or she can then display the 
subordinate records representing each version, and the details 
respecting each version (including rights and access information) 
can be displayed in a user-friendly manner.  The subordinate 
records for the different versions can be sorted in different 
ways, e.g., by date or format.

For an example of this, see, e.g., the WorldCat.org record for 
Judge Posner's textbook, Economic Analysis of Law, which has 
approximately 22 different editions or versions: 
http://www.worldcat.org/oclc/37920128&referer=brief_results . 
Rights information can be recorded in descriptive metadata for 
each version, using machine-readable standards such as Open 
Digital Rights Language, http://odrl.net/ , or the MPEG Rights 
Expression Language, 
http://www.chiariglione.org/mpeg/standards/mpeg-21/mpeg-21.htm#_Toc23297977 .

So: if an international, machine-readable standard designating 
each version of a work is adopted and incorporated into metadata 
describing each version of an article; and if rights terms are 
described in a standard such as ODRL or MPEG REL and incorporated 
into the descriptive metadata for the versions under copyright; 
and if all that metadata and its repositories conform to 
international retrieval standards such as OAI-PMH; and if "smart" 
information management tools for scholars conform to all of these 
standards; then: a scholar's information management tool could be 
programmed to automatically gather metadata describing all of the 
available versions of an article; organize that metadata into a 
user-friendly display; show the scholar descriptions of each 
version of the article in whatever order he or she wishes (most 
likely in chronological order from most recent to oldest); and 
then offer access options for those versions.

Note that only the descriptive metadata, not the full-text, need 
be retrievable free-of-charge for this process to work, even 
where text-mining is permitted in order to enhance discovery (as 
with the commercial databases indexed in Google Scholar, for 
example).

If standards are used, all of this is possible, with most of it 
occurring automatically, and resulting in tremendous time-savings 
for the scholar.

Robert C. Richards, Jr., J.D.*, M.S.L.I.S., M.A.
Law Librarian & Legal Information Consultant
Philadelphia, PA
richards1000@comcast.net
* Member New York bar, retired status.


----- Original Message -----
From: "Greg Tananbaum" <gtananbaum@gmail.com>
To: liblicense-l@lists.yale.edu
Sent: Monday, March 30, 2009 5:20:31 PM GMT -05:00 US/Canada Eastern
Subject: Re: "Accepted Manuscript"

Sandy, I would file this objection under the "queasiness in the
face of authority erosion"  category.  In your example it is not
as if the citing author has referenced an entirely different
source with different conclusions, methodology, and so forth.
In most cases the repository version will be a reasonable proxy
for the "version of record".  Reasonable people could conclude
that some sloppiness is a price well worth paying in return for
the increased accessibility.

Note that I am not making an Information Wants to be Free
argument here. Rather, I think it is simply worth acknowledging
that in many cases, information *is* free, or at least is certain
versions of it are.  Scholarly communication is trending away
from its tradition of order.  I am not sure that chaos looms, but
certainly at least a little messiness does. Repositories sit
alongside established journals.  Google is used as a proxy for
catalogued databases.  Perhaps soon concepts that were once
presented formally at annual conferences will be twittered out in
140-character bursts.  Can we as scholarly communication
professionals stop this trend? Should we even try?  To me, these
are among the most fascinating questions our field faces.  In a
world trending toward Oscar Madison, how does Felix Unger find
his place?

Best, Greg

-- 
Greg Tananbaum
Consulting Services at the Intersection of Technology, Content, & Academia
(510) 295-7504
greg@scholarnext.com
http://www.scholarnext.com