On Thu, Oct 06, 2011 at 04:14:16PM -0400, Gordana Vitez wrote: > Hi Brian, > > We came across the same problem back in the summer and I think the only thing > to do is ask the materials provider to switch the order. I did get a good > explanation from Galen and I'll include it here: > > "For what it's worth, arranging the 020s so that the ISBN-13 comes first, > followed by the ISBN-10, is actually supposed to be the standard > behavior. For example, see: > > http://www.loc.gov/catdir/cpso/020-13%20plan.pdf > http://www.ingrambook.com/isbn13/libraries/default.asp#4
For what it's worth, at least in the context of BibTemplate, it wouldn't be too hard to teach the code to iterate through all of the possible ISBNs and find the 10-digit ones. The issue is that rdetail.js is using the record.isbn() value, which comes from the old slim MODS representation of the bibliographic data. Ripping out the pertinent code from rdetail.js and dropping it into the rdetail_summary.xml / result_table.xml display would be how I would go about doing such a thing if I were to do such a thing as quickly as possible with minimal disruption. We're already doing part of that at Conifer, just to show all of the ISBNs for a given record: see http://ur1.ca/5bjbu for example; extending that to say "okay, just give me the 10-digit ISBN" wouldn't be much more work. Of course, a better way to handle this might be to push the logic down into the added content layer, such that each added content provider would be responsible for pulling out the preferred ISBN - or ISSN, or UPC, or other identifiers of interest beyond the currently limited ISBN-only approach - and sending that to the added content service. Jeff Godin had started towards this approach at the Evergreen hackfest in 2011, it would be great to move along that path further.
