Feedback is New Scientist’s popular sideways look at the latest science and technology news. You can submit items you believe may amuse readers to Feedback by emailing feedback@newscientist.com
Solved! Or not
Feedback is as fond of true crime as the next morbidly curious ghoul, so we have occasionally dipped our toes into the never-ending well of speculation about the Whitechapel murders of 1888-91 and the near-mythical Jack the Ripper. Although frankly, we didn’t get much further than Alan Moore and Eddie Campbell’s From Hell, which (spoiler!) ties the killings to the British establishment and the Freemasons, who supposedly arranged the murders to create an evil psychic force that would perpetuate the patriarchy. But the field of “Ripperology” extends far beyond one eccentric graphic novel.
So our attention was drawn to recent news stories reporting calls for a fresh inquest – backed by Karen Miller, a distant descendant of Catherine Eddowes, one of the case’s five murdered women.
It all hinges on a shawl that supposedly belonged to Eddowes, which was collected by a police officer at the time and kept in his family for over a century. The shawl came up at auction in 2007 and was bought by Ripperologist Russell Edwards. He arranged for the shawl to undergo DNA testing, the results of which were published in 2019. Geneticists Jari Louhelainen and David Miller obtained mitochondrial DNA (mtDNA) from two people. One had genetic markers in common with Karen Miller, suggesting it came from Eddowes. The other matched a distant relative of Aaron Kosminski, a barber who was a suspect at the time of the killings.
For Edwards, this is proof that Kosminski was the murderer – a scenario he has promoted in his book Naming Jack the Ripper. Other Ripperologists are sceptical: social scientist Katie Charlwood points out there is no reliable chain of custody for the shawl, and no evidence the five murders were all committed by the same person.
Feedback is in no doubt this story will rumble on forever, but we do want to add something – something not one of the recent news stories picked up on. We looked at the 2019 study and discovered that the editors had added an “expression of concern” in August 2024.
It mentions “concerns raised by third parties after publication”, as well as letters to the editor. And then comes the bombshell: “During the investigation, the publisher and Editor-in-Chief made every effort to obtain from the authors the original raw data from the mtDNA analysis. However, the authors stated that the data were no longer available, due to instrument data failure and other complications.”
Yes, you read that right – the crucial mtDNA evidence can never be verified, because the authors have lost it. Maybe Moore was right about the evil psychic force after all.
The equation for love
Here’s a romantic tale. News editor Alexandra Thompson draws our attention to a preprint entitled, “A Formula for Love: Partner merit and appreciation beget actor significance“.
The authors argue that romantic love is “a means to the end of feeling significant and worthy”. Feedback isn’t sure about that, but let’s go with it. This leads them to “a multiplicative tri-factor model” that determines “the likelihood of the actor falling in love with the partner”.
Specifically: “Love for a partner depends on the actor’s perceptions that (1) the partner possesses meritorious characteristics, and (2) that they appreciate the actor and view them as significant. We assume that these two factors multiplicatively combine with the magnitude of actor’s quest for significance to determine the likelihood of the actor becoming enamored with partner.”
In other words, the likelihood of you falling for someone is a combination of how good you think they are, how much you think they appreciate you and how much you care about finding meaning in your life.
Feedback tried to extrapolate this into dating advice. The frequent suggestion that one should play hard to get seems counterproductive, if the amount of appreciation you show your partner is a predictor of whether they fall for you. Instead, it seems like a good idea to seek out a partner who is engaged in a desperate quest for significance in life, because they are more prone to falling in love. However, this might have its own downsides, not least the distinct possibility that such a partner might join a cult.
Good luck out there, folks.
Biting the hand that bit
Via news editor Jacob Aron and the Financial Times, Feedback learns that AI company Anthropic doesn’t want potential employees to use AI when writing job applications. Their job ads say: “While we encourage people to use AI systems during their role… we also want to evaluate your non-AI-assisted communication skills.” But why, Anthropic? Could it be that the AI letters are full of guff that is unbearably tedious to sift through?
By curious coincidence, Feedback learned of this just days after news broke that Chinese AI firm DeepSeek has outperformed US tech giants. OpenAI promptly complained, saying it was “reviewing indications that DeepSeek may have inappropriately distilled our models” – that is, engaged in copyright theft.
To sum up, these AI companies don’t like being bombarded with AI-written slop and they don’t like it if their work is used to train an AI without permission. As a writer whose work has almost certainly been scraped by AI companies, and who has not seen a penny in return, Feedback can only say: “Bwahahaha, sucks to be 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’s and past Feedbacks can be seen on our website.
