As someone who is passionate about science communication, I am all in favour of any means that can get information about science out to the general public. Scientists often criticize science journalists for being too simplistic, but I can testify to the ease with which an over-simplified explanation slips out when you are in a radio studio, being interviewed by a jolly breakfast DJ who hasn't a clue about science and, to be honest, hasn't much interest. It is natural to highlight the sexy aspects of the science and inevitably to distort the picture.
The same temptation occurs when writing about science, but at least then there is the opportunity to take a step back and think about what it is that you are saying. However there is always a difficult balance to be made between making a subject approachable and capturing the science accurately.
There is another sin of science communication, though, which strangely scientists are just as prone to as science writers - and that is drawing unnecessary conclusions.
I found a great example of this in a piece in Chris Smith's Naked Scientist book, which consists of Smith's short write-ups of really interesting scientific papers. The example in question comes in a piece on the discovery of complex planetary system around the star 55 Cancri, about 41 light years from Earth. It was already known this had four gas giant sized planets, mostly occupying the space that rocky planets take up in our solar system (one is so close to the star that it orbits every 2.8 days).
The paper in Astrophysical Journal (back in 2008) described a fifth planet, in a 260 day orbit (so roughly positioned like Venus), at least 45 times the size of Earth - which made 55 Cancri's the largest planetary system that had been identified. So far, so interesting and uncontroversial. But the piece ends with a quote from co-discoverer Geoff Marcy. It's not clear if this is in the journal or whether Chris Smith interviewed Marcy, but either way the remark is a classic of what not to say. Marcy remarked 'We now know that our Sun and its family of planets is not unusual.'
The implications of this statement are a) this system is like ours and b) there are lots of them out there. But what are the facts? Admittedly 55 Cancri is not dissimilar to the Sun, but this is a system with five planets discovered, all of them vastly bigger than Earth, and with the closest three in 2.8, 14 and 44 day orbits - extremely tightly packed around the star. (Compare Mercury's 88 day orbit.) So not at all like the solar system. And discovering one other example doesn't make something is not unusual. We think there are billions of planetary systems in our galaxy. This deduction is made from two of them. Two points, it has been observed, do not provide any information on a graph.
You might argue that everyone has the facts, so will be aware this is a dramatic exaggeration. But when an expert makes a pronouncement like this it is those words that get picked up on, not the detail. There's some pressure at the moment for scientists to make more effort to communicate directly to the public. And that's fine. But they need to be trained to think as carefully about their public pronunciations as they do about their peer reviewed publications.