Was speaking to Rebecca Lawrence of F1000 lately, and she pointed me to the responses of journals to F1000's publishing of posters:  http://f1000.com/posters/journalresponses.  Most had no problem, i.e., they did not consider it pre-publication, but others did.  Was surprised at this, as statements to the effect of "a preliminary version of this work was published in abstract form" used to be an accepted practice for most journals.  See also a blog in Embargo Watch on this subject.

I have heard authors express concern that depositing data sets in a public repository prior to publication would cause some journals not to consider their work.  I have always said that this was untrue, as new analyses of public data is novel work.  But will try to gather some support for this for the FORCE11 website.  If anyone knows of any documentation, please let me know.