There was a minor media flap the other day when former CIA official Paul Pillar went public (again) with accusations that the Bush Administration "cherry picked" intelligence to justify invading Iraq. Thomas Joscelyn and Stephen Hayes do an excellent job of de-constructing Mr. Pillar's charges, reminding us that Pillar has been a long-time critic of the White House, and was guilty of his own "cherry picking" in the past.
On a related note, the accusation of cherry picking is something of a red herring in the intel business, and Mr. Pillar knows it. Any major report, assessment, or National Intelligence Estimate (NIE) is cobbled from dozens, sometimes hundreds of bits of data, drawn from a variety of sources. Determining the material that is used (or ignored) in the final report is left up to the analyst and his supervisors. And, for better or worse, most analysts in 2002-2003 believed that Saddam had WMD. Mr. Pillar may call it cherry picking, but that assessment reflected the consensus of the intel community at that time, based on information extracted from literally thousands of reports.
As Hayes and Joscelyn point out, Pillar's performance as National Intelligence Officer on the Near East/South Asia desk left much to be desired. That performance, coupled with his outspoken criticism of the Bush Administration, apparently hastened his departure from Langley. So there's more than a element of sour grapes in Pillar's charges--something the WaPo left out (surprise, surprise) in its original report.