Like all popular science authors, I get my fair share of communications from people sharing with me their new theories, whether they be an alternate theory of gravity or time, or disproving quantum physics. In the old days, these used to come as letters via my publisher - I particularly treasure examples with impressive diagrams such as the one illustrated. These days, of course, they tend to be emails, but usually my response has to be that I'm a science writer, not a working physicist (or mathematician) and I'm simply not qualified to comment on their theories. Sometimes, though the email is rather different. The other day, I got one asking me if I was intentionally filling one of my books with 'people of many diverse backgrounds'. It is true that some publishers will request more quotes from women or scientists, with diverse ethnic backgrounds, though when writing a primarily historical book, like my Ten Days in Physics that Shook the World , there is a limited pool

My new book,  Biomimetics, is now on sale. It looks at the ways in which technology and engineering have been inspired by nature - from the idea for Velcro that came from a burdock burr, to self-healing concrete. As I comment in the book: This book has turned out to be very different from the one I originally intended to write. The reality of what has happened in this field is at odds with the glossy marketing that refers to it (or, for that matter, practically every book that has been written about it). This lack of following expectation is not a bad thing. Talk to any scientist, and they will tell you that science gets interesting when the world does not behave the way we expect it to. Based on this metric, biomimetics and its implications for our understanding of the use of naturally inspired science and technology in new products and design is very interesting indeed. Rather than tell you why I think it's wonderful (perhaps with a touch of bias), here's what Publishers Wee

Sigh. For reasons I don't entirely understand, I receive emails from Travel Neutral, a company that specialises, according to its website, in 'carbon offset and planet-aware holiday deals' like the one advertised in the image. To be honest, they aren't my kind of holiday firm, because I don't consider going on holiday a good enough reason to fly, but out of interest I followed a link to the company they use for offsetting .  The good news is that they offer a good range of sensible green schemes from tree planting to renewable energy projects. The bad news is that using these for offsetting is mostly greenwash. The idea of offsetting is that at the moment we all have to do things that generate greenhouse gasses. So you pay a little money to a scheme that will reduce greenhouse emissions, and this will balance out your contribution to climate change. But unless the amount of money you contribute will reduce carbon emissions (or equivalent) by at least the amount you

Two writers shine when it comes to Tudorbethan murder mysteries: C. J. Sansom for his Shardlake books and S. J. Parris (Stephanie Merritt) for her novels with the unlikely figure of Giordano Bruno as detective. For popular science writers, the historical Bruno is a bit of a problem, as he is often portrayed as a martyr for science, but in reality was a mystic whose ideas were unoriginal and whose execution was for common-or-garden heresy, rather than being ahead of his time on cosmology. But as a detective he makes a great character in the loveable rogue with a conscience tradition. Think a sixteenth century version of Lovejoy (the books, not the TV series), but with less of tendency to kill people. Parris makes great use of this in her series of novels. This latest, Alchemy , is set in Prague in 1588. The setting, with its contrasts of the Emperor's palace and the conditions of the poor is handled excellently. There's a particular opportunity here to explore some of the oddit

One lesson anyone who has played with AI and big data should learn is that you can't trust it without having some checks in place. The technology can do the grunt work wonderfully - but then it needs verifying. This came to light when ChatGPT was caught making up references - but it can happen in a more subtle way too that is reminiscent to me of a flaw most of us never thought of when we saw Apple's 1987  Knowledge Navigator video. For those of us not ancient enough to remember it, Knowledge Navigator was a 'view of the future' video from Apple that predated iPads, folding screens and AI. In it, a distinctly smug academic uses a combination of AI, big data and a touch interface to link information about different events in the world to get the big picture. What said smug academic never does is actually check that what he is being shown is correct. Of course Knowledge Navigator was fantasy - and for those of us who saw it when it came out, it was a genuinely wonder