We need to call for significant reform.
Some possibilities we might consider:
1) decisions to reject over significance should be made within a couple months. Later decisions to reject should only be due to error.
2) obviously biased or condescending reports should not be permitted
3) referees should be publicly named
4) editors should be serve limited renewable terms
5) solicitations for new editors should be conducted without bias like any other job search: openly advertised with due consideration given to all applicants
6) journals that lack diversity among the editors should be reviewed for bias in the selection of editors
I am not yet sure how we should call for reform. Perhaps first should be to request that the mathematics organizations like the AMS establish certain standards. We might also consider contacting the publishers.
Please post your ideas here in this google group. If you fear the fact that this group is visible to the public you may privately email your ideas to me and I will post them.