Science editorial on who takes responsibility for the research in a paper

In its January 1st editorial Science laid out ideas for “Promoting Scientific Standards” including dealing with the issue of who takes responsibility for parts of a research project. The editorial states that “Science will require that the senior author for each laboratory or group confirm that he or she has personally reviewed the original data generated by that unit, ascertaining that the data selected for publication in specific figures and tables have been appropriately presented.”

The journal feels that requiring a single individual to accept this responsibility “has become increasingly unrealistic, considering that a large fraction of publications now contain contributions from groups with very different expertise—and that half of the papers published in 2009 by Science had authors from more than one nation.”

Is this the way to go, or should journals still try to have one person act as the guarantor? Thisis what the GPP guidelines continue to recommend. Are there example fromdisciplines outside of biomedicine who have other  practices?"