http://www.jisc.ac.uk/project/<call code>/<project tag>

You could ask projects to ’suggest’ a tag at proposal stage (I say suggest because projects have no way of knowing whether their suggested tag will be unique within the chosen funding call). You would then have to de-duplicate any duplicate suggested tags in some way - presumably after you’ve agreed what is going to be funded.

At that point the ‘tag’ can be turned into a URI using the following structure:

http://www.jisc.ac.uk/project//

or some such?

Note: in line with Linked Data this would be the URI for the project… not for a web page about the project, so your website would have to be configured to issue a 303 redirect to an ‘information resource’ about the project.

It seems to me that we see this in pretty much every arena of the legislature. The fact that a new law is pointless (or worse still a further erosion of our fundamental rights) or that new plan has no value is irrelevant; the politician or the bureaucrat can point out that “something was done” and move on, while the rest of the world carries the cost. Think of airport screening.

On the specific issue of the records companies blindness, this is standard when technology catches up with a company with a hugely profitable revenue stream. These companies are frightened of anything which affects this cash cow and so will not invest in the future (I have heard the phrase phrase “eating your own babies” used to describe investing in something which may affect your cash cow - that gives an idea of how well the it goes down), rather they will invest in protecting that revenue stream.

The whole tech (and non tech) space is littered with such casualties. The record industry is interesting in that it has significant political clout and probably has more control of the media than we’d like.

I think you are spot on with reference to the need for more work on managing attributes based on groups. My institution is currently grappling with the problem of how to restrict access to resources which are licensed only for use by sub-sets of our members. Logically, we should be able to use additional scoped affiliations but the service providers show no interest in cooperating and seem to think that all they have to do is look for the “member@” and it is our problem to enforce compliance with the license conditions. Use of the federation is now starting to seem to some like a risk and I am am having to field comments along the lines of “it was so much easier when we had athens” …

The problem is not technical because generating additional scoped affiliation strings from group memberships is relatively trivial. It is about the problem of engaging service providers with the need to become more granular in granting access. How do we make this into a benefit for the providers?

The other thing that struck me was the purchasing power of CDL was quite considerable and not comparable to any single UK institution. The only way we can get changes made is to act in concert.

There is also the issue that libraries are a service, in the business of providing access to information for their users. There is a danger to the reputation of libraries within their institutions if they don’t give access to certain resources - no matter how much of a pain it is administratively. The value of a specific resource can vary widely across libraries, depending on their users needs, so agreeing a universal ‘deal breaker’ is tricky, if not impossible.

Finally (and I don’t think you are going to agree with me here…) I’d argue that implementing IP authentication is much cheaper and simpler than implementing SAML, and it is more realistic that small publishers could implement this, so if I was going to put a ‘deal breaker’ requirement on this, I’d be arguing for IP authentication being the ‘minimum’ not SAML.