According to a post at Nature by Richard Van Noorden, the rate of retracted scientific articles is growing far faster than the rate of published or posted articles. No one is sure why, but it is exposing inconsistencies in policies for dealing with retracted articles.
Suggested reforms include better systems for linking papers to their retraction notices or revisions, more responsibility on the part of journal editors and, most of all, greater transparency and clarity about mistakes in research.
It’s encouraging that it’s taken as obvious that the proper response is links and transparency. Gotta love science.