Let's take a break from the usual flood of "open access, open access!" content to step back and look at institutional repositories more in general.
From Marisa Ramirez and Ann Hanlon's Asking for Permission: A Survey of Copyright Workflows for Institutional Repositories , I learned that one can contact publishers not just about individual articles, but about retaining your publication rights for all articles you send them in the future -- or even all articles you entire institution sends them in the future - and yes, you can use form letters for all of this; no need to agonize over your phrasing. (Amy and I have therefore launched an effort to get blanket permission from Purdue from one of our department's major publication outlets. More on this story later as it develops.)
I also noted the following passage with a tone of righteous indignation:
"...libraries should not pay to provide open-access to articles authored by their faculty if they are already paying content licensing fees, and paying salaries to faculty who are not compensated by publishers for their contribution."
It's like a strange parallel universe. Usually, when you write something for someone else to sell, they pay you. But in academia, you (or your university) pay them. And then you pay them for a subscription. To your own work. What the hey?
Figuring out copyright is hard, and it's made harder because almost nobody (20% of the responding repositories) shares the responses they get from publishers - and even those who do are very careful because of fears (founded or unfounded) regarding legal liability, sharing only individual policies on general sites like SHERPA/RoMEO. Which is incredibly useful, by the way. Check it out if you're curious about an individual publisher's policies - some repository managers didn't even know this sort of resource existed.
The survey (of 121 repositories, mostly US/UK ones due to a paucity of institutional repositories elsewhere in the world) confirmed that librarians (not authors) are doing the bulk of the work needed to get content in a repository, that everyone was scared of copyright, and that "educating authors on copyright" was the most common challenge on the open access repository front, followed closely by "obtaining publisher copyright policies," both of which can be tedious and thankless jobs.
Summary: learn more about copyright so you don't have to be afraid, and realize that those who want to see the work done (in this case, institutional repositories populated -- and in my case, open access) are going to have to do the work, because even if faculty are supportive of the effort, they have no time.
Next week I'll begin working with my first faculty participant - the goal is to make one of my former professors the first person to have their complete scholarly works up on Olin's institutional repository (or at least the ones they published while at Olin). Stay tuned.