I think the problem is that there isn't necessarily an infrastructure in place, you just need to reach out to people. You probably know the people you would like to contribute, it wouldn't hurt to test the waters, point them to the American Yawp as a successful example, & see if anyone would be willing to collaborate. This kind of project disrupts the status quo of academic publishing, and we're just indoctrinated with an 'it is what it is' viewpoint that we don't break it.