The American Astronomical Society recently made all its journals
open-access (past, present, and future content):
Not a huge change though since, at least for the past few decades,
they've had page charges (in addition to subscription charges) and the
new article-processing fees are comparable to the old page charges. So
not a big change for authors. Libraries might save a bit, but in most
cases that budget will be redirected to page charges. All of the old
stuff is at ADS anyway, so also not a huge change for many readers. The
biggest change is that the official version is online and freely
accessible without embargo. Not everything is on arXiv; if so, it isn't
always the definitive version; if so, that is not always obvious. Maybe
some people will now notice that there is some good stuff not on arXiv.
Monthly Notices of the Royal Astronomical Society is now also online
only (like the AAS journals have been for a while), but still free for
authors. One can pay a fee to have it open access immediately
(otherwise it is funded by subscription fees and also subsidy from the
RAS), but they actively encourage one to post an author-postprint
identical to the definitive version with no embargo, and authors are
given a link so that anyone can access the definitive version right
away. So in terms of "true open access" that seems the better bet.
Although their astronomy journal is just getting underway, these folks
seem to be doing open-access scientific publishing in all the right
. At least if they cover your field, I
can't think of a better place to publish.