This is an important topic and past history for me has been mixed. For instance, just because one _can_ turn on access to 100s or 1,000s of open access resources in a link resolver or other knowledgebase at the click of a mouse does not mean one _should_. That’s
my opinion. Past practice at my institution seemed to favor turning on access to everything and initially I thought that was fine and continued doing it. But over the last few years I’ve drastically pared back due to a number of factors. Primary among those
is unreliability of so many OA titles (will a title level link even work consistently? an article-level link is even less likely in many cases), which is nothing new, as well as a reality that the things we pay for are still not adequately accounted for. By
this I’m referring to things like JSTOR, subscription titles, etc.
If there’s one thing I’d advocate OA publishers do it is to ensure and enable consistent linking to the article level or book section level, e.g. by supporting DOIs. This is very often overlooked. Users expect every link to work every time and
when that doesn’t happen, they get understandably frustrated. The oaDOI SFX integration mentioned by Chris is a great tool and we use it, but it is far from foolproof.
In addition, the ease with which access can be turned on can hide the reality that OA resources, in my view, should fit within a collection scope. Just because it’s “free” doesn’t mean it belongs.
Just a few thoughts or reactions to this thread.
Steve
Steve Oberg
Assistant
Professor and Group Leader for Resource Description and Digital Initiatives
Buswell Library, Wheaton
College (IL)
+1
(630) 752-5852
