KMLM List

[Cliff Spencer <c.spencer@xxxxxxxxxxxxxxx>]

Hi Chris - Interesting.

IMO no current identifier is perfect for post-collection data analysis. I've 
been using PRINT issn for years as a key field to manipulate the data.

Title field has about a 7% error in a typical report of ~2,000 titles - so is 
that: Journal of Environment; Environment journal The; Journal & Environment 
etc.

Print issn has about a 3% error rate, often omitted or just plain wrong. Online 
issn is sometimes not given at all so I don't use.

I know little of the DOI standard other than it to locates articles rather than 
titles (but I guess it's configurable so that it could tag jnl titles too).

The next big move forward is for COUNTER to set a standard for article level 
metrics (see http://article-level-metrics.plos.org/ and ;
http://www.jisc.ac.uk/whatwedo/programmes/pals3/pirus.aspx)  where the DOI 
would certainly be a contender?

E-books might report either 13 or 10 digit isbn, (sometimes both!) and with 
older books they may not have an isbn at all. At the moment I use the title 
field for data processing which seems to be more accurate than journal titles.

Cliff.

===

From: sushidevelopers@xxxxxxxxxxxxx [mailto:sushidevelopers@xxxxxxxxxxxxx] On ;
Behalf Of BAKER, Chris
Sent: 09 July 2010 11:00
To: sushidevelopers@xxxxxxxxxxxxx
Subject: [sushidevelopers] ISBN field in COUNTER reports

This is not, exclusivelya SUSHI question  but I feel certain that people on 
this list are among the right ones to discuss it, so I hope you don't mind me 
raising it.

Many COUNTER reports require an ISBN and ISSN field.
Publishers often have many ISBNs -including ones that do not affect the content 
that the customer sees  e.g. paperback and hardback ISBNs or ISBNs for various 
sales offerings (such as bundles of titles).#

Which ISBNs do you use?

Some of us at OUP have been musing about suggesting to COUNTER that the ISSN or 
ISBN fields ought to be replaced by a DOI (Digital Object Identifier) field. 
Our thinking is this -  the idea of quoting an ISSN or ISBN is to give the 
librarian a reference with which she can compare usage of the same content from 
two suppliers (we think). If that is so, then the modern way of unambiguously 
identifying a piece of content is the DOI.

Your thoughts please!

We have thought of one caveat abotu thsi idea (maybe you can think of more)-
Do many people make format assumptions abotu the ISBN and ISSN fields when they 
process SUSHI or manually downloaded COUNTER reports? (E.g. you might assume 
that the ISBN field will always be a 13-digit numeral - This assumption would 
become wrong if DOIs were used instead)?