Repository depositions – what scales? A simple idea

One of the problems of repositories at present is that everything is new. And much of it is complex. And some changes rapidly. So here is a simple idea, motivated by Dorothea's reply to a post of mine…

Dorothea Salo Says:
November 12th, 2007 at 2:51 pm e

[… why repositories need investment …]

And some of the work is automatable over time. Once you know a particular journal's inclinations, pushing anything from that journal out of the bucket becomes a two-second decision instead of a ten-minute slog through SHERPA and publisher websites.

PMR: Now this is an area that is a vast time-sink. Suppose I (as a simple scientific author) want to know if I can archive my Springer article (read also Wiley, Elsevier, ACS, RSC…). What do I have to do? When?
I imagine that hundreds of people struggle through this every year. Frantically hacking through awful, yes awful, pages from publishers. Many of these are not aimed at helping authors self-archive but suggesting how they can pay money to the publisher for re-use of articles. (I could easily rack up a bill of 1000 USD for re-using my own article if I wanted to include it in a book, use it for distance education, use it for training etc.). It is not easy to find out how to self-archive – I wonder why?
So I thought I would try to do this responsibly and find out what Springer actually allows. I have a paper in a Springer journal – what am I allowed to do and when? The following journey may be inexact, and I'd appreciate correction, but it's the one that a fairly intelligent, fairly knowledgeable, scientist who knows something about Open Access followed.
I went to the home page of J. Molecular Modeling and looked in "For authors and editors". Nothing about self-archiving. A fair amount about Open Choice (the option where the author pays Springer 3000 USD to have their full-text article article visible (like all other current non-OpenChoice articles in J.Mol.Mod) and archived in Pubmed and re-useable for teaching without payment but with the copyright retained by Springer. I went to Google and typed "Springer self-archiving". I won't list all in detail but the results in order were:
A report by Peter Suber (2005), Journal of Gambling Studies, a critique by Stevan Harnad, a Springer PPT presentation (2004) on Open choice (which stated:

Springer supports Self-archiving: Authors are allowed to post their own version on their personal website or in their institution's online repository, with a link to the publisher's version. (PMR this is the ONLY page I found from Springer).

… an attack by Richard Zach on the 3000 USD for Open Choice, an attack by Stevan Harnad on why Jan Velterop opposes Green self-archiving, a page from Sherpa-Romeo which gives the conditions:
http://www.sherpa.ac.uk/romeoupdate.php?id=74
and is the most helpful of all.
I immediately take away the fact that Springer is making no effort to help authors find the conditions for self-archiving. I have no idea where they are. I'd hate to do anything that violated the conditions.
So, to follow up Dorothea's post. A LOT of useful human effort is wasted because the publishers make it so difficult to find out how to self-archive. I'd like a fair bargain. If the publisher has agreed that you can self-archive, tell us how. Or we start to see publishers as a difficulty to be overcome.
So a suggestion. Suppose each institutional repositarian spent 1 day a year posting how to self-archive articles from the Journal of Irreproducible Results. (Don't be fooled – it takes a day to get a clear answer from most publishers or journals). And each one took a different journal. And posted it on a communal Wiki. Then we would have a clear up-to-date indication of what was allowed and what wasn't. Including things like "I asked if I could retain copyright and they said yes". Really vital info.
It's not a lot of work per person. It would pay back within a year. Someone has to set up the Wiki. And keep it free of spam. But that's not enormous. But sorry – I'm not volunteering. I'm in a discpline where there is very little chance of self-archiving legally. I've spent enough time trying.