I think the cookie pusher method is inherently flawed, with lots of 
problematic edge cases like this.

I simply don't use it.  Yes, that creates other problems of it's own.

The fundamental problem with the whole DOI resolution design is ignoring 
the "appropriate copy" problem, not sure there's any great way around 
it. If there is, I don't think "cookie pusher" is it.

On 3/22/2011 2:04 PM, Tim McGeary wrote:
> Lehigh is currently using the CrossRef cookie pusher to redirect DOIs to
> our OpenURL menu (SFX). We use the methodology as described in the
> "Redirect DOI links ('DOInapping')" section of this document:
>
> http://igelu.org/wp-content/uploads/2010/09/SFX-and-DOI.pdf
>
> We currently have the cookie pusher on our most frequently used access
> points to subscription resources: our database/e-journal lists, library
> catalog, SFX citation linker, SFX menu itself, etc.
>
> However, we are having some valid complaints from non-Lehigh users, who
> have this cookie set for them unintentionally.
>
> One recommendation given in the article cited above, is to includ
> opt-out documentation, which we are not currently doing. Does anyone
> have an example of what this documentation would look like?
>
> Other options we are considering:
> 1)Stop cookie pushing entirely
> 2)Have an explicit link to the cookie pusher page, so users would have
> to opt-in to have it set.
> 3)Limit the display of the cookie-pusher to only users in our IP-Range
> and place cookie-pusher on our EZproxy login page for off-campus Lehigh
> Users.
>
> Any comments/suggestions for these or other solutions?
>
> Thanks,
> Tim
>
> * - credit to Rob Weidman at Lehigh for the original solution, and
> investigation to other solutions
>
>