麻豆传媒

Heightened interest

Our hats, of which we have very many, are firmly off this week to Fr茅d茅ric Bouquet and fellow inhabitants of the 15-metre-high Laboratory of Solid State Physics in Orsay, France, for their paper 鈥�61 ways to measure the height of a building with a smartphone鈥� ().

We are reminded of an urban legend about a physics student who later turned out to be Niels Bohr. When challenged during an oral examination to describe how he might do something similar with a barometer, he came up with several indubitably correct answers that contributed to getting him failed.

These included lowering the barometer from the roof to the ground on a long piece of string and measuring the length of string plus barometer, and throwing the barometer off the roof and timing how long before it smashed on the ground. The traditionalist within us is pleased to record that both of these methods appear on the team鈥檚 list, although this does suggest that measuring the height of a building with smartphones, plural, would be a more accurate description.

A mere cavil. We can鈥檛 fault the comprehensiveness of the research, which employs methods involving free fall, giant pendulums and various acoustic and optical techniques, as well as a section of the most interest to physicists, 鈥渕ethods that only work in theory鈥�. Nor can we fail to be impressed by the range of values recorded for the building鈥檚 height, from zero (鈥淕iant pendulum, using the accelerometer鈥�), to 800,000 metres +/- 100,000 m (鈥淰ariation of the Earth magnetic field between the top and the ground鈥�).

The method 鈥淰ariation of gravity between the top and the ground, determined by general relativity time dilation鈥� also returned a value of zero, although with a margin of error of 3 million kilometres either way, leading the researchers to conclude, perhaps ruefully, that 鈥渙n average, the simpler the method, the more precise the results鈥�.

Not consistently, however. Some methods were unable to deliver an answer at all, among them number 61: 鈥淧hone call to the building鈥檚 architect.鈥�

Lower!

鈥溾��, reports a headline on Sky News. Feedback is disappointed on further reading to find that this wasn鈥檛 a result of the common doctor鈥檚 injunction of people presenting with problems down below 鈥� 鈥渏ust cough for me鈥� 鈥� one that always puzzled Feedback until we mindfully noted the muscular spasm that passes through those areas when we cough. It seems it was actually positron emission tomography that led to the breakthrough, which is real science, but far less fun.

Bone of contention

A delightful exchange ensued on Twitter after reported that his wife had fallen foul of software being used to censor language in submissions to a question and answer session at a conference. 鈥淥ne of the words it is censoring is 鈥榖one鈥�,鈥� he writes. 鈥淚t鈥檚 a palaeontology conference.鈥�

Happily, the organisers of this year鈥檚 Society for Vertebrate Paleontology (SVP) meeting , tweeting: 鈥淗i all, you鈥檒l be pleased to know that 鈥楬ell鈥� and 鈥榖one鈥� are now permitted on the Q&A function #2020svp Please use responsibly!鈥� The tweet included a smiling face and three bone emojis.

Courtesy of our man with the old bones, Jeff Hecht, Feedback has been able to view a full list of words that the software blocked 鈥� all, we presume, included in participants鈥� submissions to Q&A sessions.

Largely of a putatively sexual nature, they provide insight into the palaeontological mind. We can only assume that anyone enquiring about 鈥渆rections鈥� in that context would have unimpeachable intentions, but we are less sure about 鈥渆nlargement鈥�. Also appearing on the list, eliciting varying degrees of polite nodding and smiling on Feedback鈥檚 behalf, are 鈥渟troke鈥�, 鈥渟tream鈥�, 鈥渒nob鈥� and 鈥渇lange鈥�.

That last one causes us to wander down a byway of wondering how words can assume sexual connotations simply by sounding as if they should have them. Perhaps some expert in acoustolinguistics, a research field that we have just made up in hope, can enlighten us as to the quality of these words that makes it so.

But this is by the by. The whole thing reminds us of the recent story reported by the BBC of the Canadian seed and garden supply company whose picture of onions was rejected by Facebook鈥檚 algorithms for being .

AIs clearly aren鈥檛 going to make it far in the saucy seaside postcard industry, but more generally, following recent discussions about artificial intelligence taking our jobs, we are glad of further evidence that, actually, our future jobs will be correcting the job that the AI did.

Not (yet) in the bag

Allen Young writes in with an advert from our own esteemed organ for the 麻豆传媒 shop, and a special offer of a free tote bag on all purchases over 拢20. (Good offer, that, and the bags are very fetching.)

Too late, however 鈥� possibly. 鈥淥ffer valid until 31st October鈥�, the advert trills, 鈥渙r until stocks last鈥�. Well, it is past 31 October now, Allen, so we can only suggest you enquire whether stocks have begun to last yet, and if they haven鈥檛, claim your free bag. Just don鈥檛 tell anyone we sent you.

Got a story for Feedback?

You can send stories to Feedback by email at feedback@newscientist.com. Please include your home address. This week鈥檚 and past Feedbacks can be seen on our website.