« UK Daily Telegraph does OpenID | Main | OAI ORE specification roll-out meetings »

January 22, 2008

Bunfight at the Athens/Shibboleth gateway

Those of you on the UK academic jisc-shibboleth@jiscmail.ac.uk list will have seen by now that the competing press releases, concerning the non-funding of the Athens/Shibboleth gateways, from us (Eduserv) and JISC have been released.

What's all the noise about?  Well UK academia is in transition from Athens (the proprietary access management service run by Athens on behalf of JISC for the last 10 years or so) to Shibboleth (the open, standards-based identity and access management framework, currently popular in academia).  This is absolutely the right transition to make but, of course, there are implementation issues that come with the details of the transition.  One such detail surrounds the provision of the so-called Athens/Shibboleth gateways.

What are the gateways?  Essentially, they provide a bridge between the old and new worlds.  Why are they necessary?  Because at this stage it looks highly likely that not all service providers will have switched to Shibboleth by the time the Athens funding ceases (later this summer).  Any institution that has made the switch to Shibboleth but that needs to continue to access resources from service providers that remain Athens-only will need to do so via an Athens/Shibboleth gateway.

Unfortunately, discussions between the JISC and Eduserv about how much money is available for Eduserv to run the those gateways have broken down - leading to a situation where both parties have issued press releases explaining their view of the situation.  This has led one commentator on the mailing list to ask:

Am I the only one that feels like they are witnessing a school playground argument?

I can certainly sympathise with that point of view!

People are clearly witnessing a disagreement.  Whether it is of the school playground variety is a different matter.  I tend to disagree at this stage, though I'm sure something can be arranged if necessary :-)

Also clearly, the disagreement has to do with cost vs. value.  In short, the cost at which we felt it was viable to offer the gateways was in excess of the value the JISC chose to put on them - there was a disagreement about price, pure and simple.  Who was right and who was wrong in that disagreement is another matter of course, as is the issue of whether some middle ground could have been reached.  I'm not aware that we refused to negotiate (I could be wrong) but in any case, from what I've seen, the gap between the two sides was such that I doubt any negotiated agreement could have been reached even if a longer period of negotiation had been allowed.

Clearly, this is unfortunate for the community.  We did not take our side of the decision making lightly, at least as far as I understand it.  I'm sure the JISC would say the same.  I'm equally sure that both sides of the argument feel like they are taking the 'best' decision in the circumstances.  Sometimes things just don't work out.

The bottom line, from our perspective as an educational charity, is that we have to take not-for-profit business decisions around those services that we believe to be of value to the community in order to ensure the ongoing viability of both our services and the charity overall - we can't simply provide services at well below our own costs.

Note: I tried to post part of this response directly to the mailing list but got the following response:

Your message  dated Tue,  22 Jan  2008 17:19:14 -0000  with subject  "Re: LA (mixed messages)" has been submitted to the moderator of the JISC-SHIBBOLETH list.

Call me paranoid, but I don't recall the list being moderated in the past.  Perhaps it always has been?


TrackBack URL for this entry:

Listed below are links to weblogs that reference Bunfight at the Athens/Shibboleth gateway:


It's such a shame that a step in the right direction like the move to Shibboleth could be derailed by a dispute like this. I can sympathise with both points of view - I do hope some resolution can be arranged.

"This is absolutely the right transition to make"

I become less and less convinced of this as time passes - so far all I can see is a whole load of hassle with few, or no, tanglible benefits.

My main gripe is that implementing Shibboleth to allow access to 'library' type resources is putting in a technical solution to a problem that didn't seem to exist...


You might find the answer to some of your questions on my last post on my blog Owen: http://access.jiscinvolve.org/2008/01/23/why-did-we-go-down-this-shibboleth-route-at-all/ The trouble is, everyone has become fixated on the idea that we are trying to replace Athens with Shibboleth. This has never been JISC's intention.

The comments to this entry are closed.



eFoundations is powered by TypePad