A long time ago, in what seemed like another life, I attended the Access 2006 conference as a relatively new systems librarian at Laurentian University. The subject of the preconference was this totally new-to-me thing called "institutional repositories", which I eventually worked out were basically web applications oriented towards content preservation and dissemination. One of the presenters was Mark Jordan, who talked about his efforts to build a harvester and search engine for the institutional repositories for all Canadian academic libraries. And an outcome of that effort was another effort, to build a Dublin Core application profile that academic libraries could adopt to improve the consistency (and thus the usefulness) of the metadata that was harvested.
Fast-forward to a discussion I participated in today as part of the ScholarsPortal Technical Advisory Group, in which the subject of working towards a standard set of metadata for institutional repositories in Ontario was raised. "Wait", I thought, as my hardening synapses attempted to fire, "isn't this a solved problem?" And indeed, I was able to dig up my notes from Access 2006, quickly refreshed my poor biological memory, and tried to track down the CARLCore Metadata Application Profile guidelines that I had thought all Canadian academic institutions had agreed to adopt. And then... 404 galore. The document had vanished from the net.
However, through the power of social networking, the amazing Mark Jordan was able to produce a copy and sent it to me via email, with the okay to redistribute it freely. So... here we go. Let the hard work of the members of that group find a new life, as the guidelines are now once again available at: CARLCore Metadata Application Profile (PDF)
Aside: Holy heck, when I was a new systems librarian I sure did stay quiet and listen a lot. I should go back to doing more of that. Also: extensive notes published on blog can pay dividends years down the road!
I guess this would suggest the problem is not a technical one (deciding what the standard should be) but a social business one.... an agreed upon standard is so ignored that not only was it never adopted, it's not even hosted in a canonical location anymore and it's very existence has been mostly adopted.
Unless, I guess, someone wants to suggest that the reason it was so abandoned was because of it's technical content.
But the story does suggest that entities ought not to spend precious time inventing or ammending the technical content, until they've understood and attempted to address the social/business/organizational issues. Or they may be wasting their time on another thing that will be ignored and forgotten.
Jonathan: You're probably right about it being primarily an organizational problem, but I would be wary about over-simplifying lessons that you or the community in general should draw from this one experience.
The reality is that both the technical and the organizational dimensions of the problem were rather large at that point in time. In 2006, the open access movement was a thing, but not a broadly understood thing; and institutional repositories were also not such a broadly understood thing. (Probably not even now).
I would, however, happily submit that the work that Mark Jordan and the committee did to parse through the actual aggregated data from all of the IRs in Canada at that time, in the attempt to derive and drive a common set of standards, was a valuable exercise.
In 2006, the Access preconference brought together most of the players that could have made it happen (so that the CARL harvester could have been more valuable); however, when the CARLCore 1.0 guidelines were published in 2008, I'm not sure there was a similar coalescing of the right people & sufficient will & resources to make it happen.
Now, in 2013, given the interest expressed during the ScholarsPortal TAG call, maybe the people & interest are coming together again to at least make it happen for Ontario. At the very least, there's a well though-out set of guidelines to start from, rather than going back to scratch and delaying for another year or two.