Josie Ford
Feedback is our weekly column of bizarre stories, implausible advertising claims, confusing instructions and more
Caught on camera
One of the defining features of conspiracy theories is that they are “self-sealing”. If you confront a conspiracy theorist with the lack of evidence for their claim, they will turn this back on you and argue that this very lack of evidence is proof of how powerful the conspiracy is.
This is foreshadowing.
Advertisement
Feedback is interested in watching the new documentary Capturing Bigfoot. We can’t, because we live in the UK and it isn’t available here, but if anyone would care to rectify that, we will stick ourselves in front of it immediately.
The film is about the classic footage that purports to show a Bigfoot loping alongside a river in California. Shot by Roger Patterson and Robert Gimlin in 1967, the clip became one of the canonical pieces of evidence for the existence of Bigfoot: a mythical humanoid that supposedly lives in the forests of North America. Bigfoot is one of the icons of cryptozoology, the search for probably-nonexistent creatures. It has hordes of devotees.
In Capturing Bigfoot, film-maker Marq Evans uncovers what really happened in 1967. He speaks to Patterson’s associate Bob Heironimus, who says he wore an ape costume in the footage.
Crucially, Evans also finds a lost piece of 16-millimetre film, shot in 1966, showing a rehearsal for the final Patterson-Gimlin video. A review of the documentary in Variety calls this footage “the smoking gun”. It is, Feedback would suggest, a case no defence lawyer could wriggle out of.
So, the Patterson-Gimlin film purporting to show Bigfoot was fake. Feedback and our editor worried if we were legally safe saying that, but then, as our editor said, “come on”.
It is perhaps telling that, when we mentioned this to Mrs Feedback, her response was “I could have sworn I already knew that”. Indeed, these accusations have been made before, notably in Greg Long’s 2004 book The Making of Bigfoot. The new thing is that rehearsal footage.
Naturally, then, this footage is what the Bigfoot Field Researchers Organization pushed back on. Its president is quoted in The Times as saying: “My initial reaction is, how could you believe that in the age of AI?… Even if they found some reel, how would you know whether what they’re showing in the documentary is what’s on the reel?” And thus the foreshadowing pays off.
A not-so-helpful acronym
In late October, a lot of websites suddenly went offline. This was because of a vulnerability in the PHP programming language, which had been known since 2024 but was thought to be too difficult to exploit, until AI tools found how to do it.
Reader Chris Hall had never heard of PHP. This seems to have been a common experience: even the BBC’s cyber correspondent Joe Tidy only learned of the existence of PHP when the story broke. Tidy got a lot of online abuse for this, which Feedback thinks is unfair: his job is to explain things clearly, not to be a walking technical manual. In the interests of solidarity, Feedback will say that we still don’t know what distinguishes a baryon from a boson, and at this point we refuse to find out.
Anyway, Chris looked up PHP and was given “the helpful information” that it stands for “PHP: Hypertext Preprocessor”. “I can understand the ‘H’ and second ‘P’, ” he says, “but what does the first ‘P’ mean?”
Feedback is happy to clarify that PHP originally stood for “personal home page”. When it was upgraded in the late 1990s, the programmers kept the letters and had to think of something new for them to stand for. Hence the recursive backronym that is “PHP: Hypertext Preprocessor”.
They just said it
Sometimes, scientists fake their results. The problem is spotting the fakery, because it isn’t always obvious. Often, images have been copied from other papers, necessitating systematic searches. Or you have to perform careful statistical analyses on the raw numbers to identify the trends that reveal they have been made up.
Or, sometimes, the authors might just flat-out tell you – leaving you to decide whether or not they mean it.
Nihar Shah is a computer scientist at Carnegie Mellon University in Pennsylvania, who mostly studies AI scientists: those systems that can do research and write papers with little or no human intervention. However, he is also the author of the blog Research on Research, about “academia, research, and peer review”, where he writes about some of the weird stuff that finds its way into peer-reviewed journals.
In October, Shah looked at a host of recent scientific studies and found that many of them included variations on the same phrases. These remarks showed up over and over again, in at least 60 papers. Phrases like: “The values in this table are illustrative placeholders for formatting purposes and must be replaced with results obtained from actual experiments”. Or “Planning scenario, not measured results”. Or “note: the values shown are placeholders and should be replaced with your actual experimental results”.
What on Earth is going on? On the face of it, the authors are just admitting that they have made up data. However, Shah says (and Feedback agrees) that there is no reason to assume chicanery. Perhaps all of these papers are built on real results, and the authors just forgot to remove these placeholder texts from early drafts. In other words, this could all be the result of really, really sloppy editing on the part of all concerned. Which is, we guess, reassuring? Um.
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.