David Shukman's "how to report climate change" video over at the BBC academy website is actually good in parts. But in parts it's not.
He first sets out what is agreed - carbon dioxide is a greenhouse gas, its concentration in the atmosphere has risen as have global temperatures. So far so good, although I think it would be fair to point out that the temperature rise is indistinguishable (in statistical terms) from business as normal and also that temperatures haven't risen for more than fifteen years.
He then correctly notes that the reasons for the warming are disputed, noting in the process that natural cycles can cause warming and that "the vast majority" of scientists stand by the IPCC and it's claim that "most" of the observed warming is "very likely" to have been caused by changes in greenhouse gases. He precedes the next discussion - on climate projections - by noting that what has gone before is based largely on observations. This is of course not actually true of the IPCC's attribution statement, which depends very much on climate models, and climate models moreover that have no proven ability to describe the climate. This being the case use of the word "very likely" seems troublesome, particularly when placed in close proximity to the word "most".
The section on predictions is quite good, with a short and to-the-point summary of the difficulties.
At this point one is somewhat in awe of somebody at the BBC who is saying something sensible about climate, but unfortunately Shukman is not able to hold it together, descending into normal BBC wiffle waffle about "libertarians who think it's an anti-free market conspiracy".
I suppose good in parts is better than consistently awful though.
A closing thought. If "where somebody is coming from" is, as Shukman suggests, so important for viewers, will he demand that BBC science staff discuss Paul Nurse's left-wing politics every time he appears on TV? We need to know if "where somebody is coming from" is an approach that is consistently applied.