🆘 Need some urgent help on this Rejoin EU poll.

Vote and RT ✅

www.dailymail.co.uk/news/article...

David Squires on … the TikTok of the clock as Arsenal’s title charge falters Our cartoonist on the Gunners’ latest wobble and who could be brought in to get final push back on track

Here’s this week’s cartoon for @sport.theguardian.com

www.theguardian.com/football/pic...

WIth regard to hallucinated references, I suspect that, for many journals, the big publishers do only cursory copy-editing. I've seen errors in paper titles (not just references) that should never have passed the first stage of the submission process. But they place profit over quality. 4/4

CERCA Download CERCA for free. CERCA – Citation Extraction & Reference Checking Assistant. CERCA is an open-source research tool that supports the verification of bibliographic references in scientific manu...

To lighten the load, I use the open source tool Cerca (sourceforge.net/projects/cer...), by @lidianycs.bsky.social, which checks reference lists for validity. I can highly recommend it. 3/4

Hallucinated references now mean that every reference has to be checked – just in case. I have yet to see an example in the journal that I work on (not published by one of the Big 5), probably because they have robust measures in place to screen out any problems before they reach me. 2/4

I'm a freelance copy-editor. In the pre-internet era, reference lists had to be taken on trust, with any issues referred back to the author(s). With the internet, it became possible for copy-editors to look up any missing or inaccurate information, but other references could still be trusted. 1/4

I'm delighted to say that Jenny Chambers (@icpjenny.bsky.social) last night won the 2026 Mama Academy (@mamaacademy.bsky.social) Lifetime Achievement Award!

Over more than 30 years, her work in research and as founder and CEO of ICP Support has contributed to preventing countless stillbirths.

Another in my ongoing series on publishers' failings: a paper in which Web of Science is variously given as 'web-of-science' and 'Web-of-Science'. The same paper also has a lot of missing spaces in the text.

#ScientificPublishing #AcademicPublishing #CopyEditing #Proofreading

I watched the first 10 minutes. Once I realised that the monster was impervious to bullets and could tip a ship weighing hundreds of tons, and that one gunshot could shatter thick ice, I gave up.

If it was a printing error, then the party's lack of attention to detail when proofreading is totally on-brand.

In my spam today:

'I am glad to learn from the internet that you are the main publisher in your country'

Which is news to me.

And here's another one:

doi.org/10.1016/j.jo...

Different journal, same publisher. Apparently, indefinite articles are optional.

Given how much profit journal publishers make, is it asking too much to check that the title of a paper is correct? Here's the link for anyone interested:

doi.org/10.1016/j.ij...

It's the misplaced apostrophe that's annoyed me.

#ScientificPublishing #CopyEditing

Anyone in the same business as me (publishing), take note: you cannot use AI for anything – anything at all! – when facts matter.

#BookSky #ScientificPublishing

Too late – installed Waterfox, made sure there were no problems, removed Firefox. Easy switch.

This is a house style that I have to follow for one client. I agree with you, but they're paying.

Why 2026 Is a Mathematically Special Number Neither a square nor prime number, 2026 is still intriguing

The fascinating math of the year 2026

How I Built a Tool to Detect AI-Generated Fake References Large Language Models(LLMs) have become part of everyday academic and technical writing. But there is...

How I Built a Tool to Detect AI-Generated Fake References
dev.to/lidianycs/ho...

Sharing a small open-source project I’ve been working on over the past weeks: CERCA – Citation Extraction & Reference Checking Assistant.

#research #openscience #softwaredevelopment

... the remaining copy-editors are under increasing pressure as the last line of defence against AI-generated rubbish, and no LLM will fix your reference list for you – they will hallucinate and make things worse.

3/3

... you used to be able to rely on copy-editors and proofreaders to spot and fix that kind of thing. But with publishers increasingly seeming to be willing to dispense with that element of quality control, you risk looking sloppy in your work.

And... 2/3

Another whinge about reference lists!

If you're using some kind of reference management software, please read the reference list after you've generated it and fill in any missing info (including in your reference database.

Why? Well... 1/3

#ScientificPublishing #AcademicPublishing

Very useful - thanks.

I've never run into the ghost reference problem in the journal I copy-edit, but I've had to be much more proactive about checking references since this issue arose. Copy-editors are the last line of defence, but guess who journals drop when they want to save money?

... I answered a bonus question about fictional games on University Challenge because I'd just finished the marathon task of reading Infinite Jest. (I got the other two bonuses as well, which just shows how impressively well-read and pretentious I am.)

High and low culture in quizzing in 2025: I watched KPop Demon Hunters and consequently scored a point in a pub quiz because none of my teammates had lowered themselves to watch the film. At the other end of the scale...

A cappuccino with a snowman sprinkled in chocolate on the foam.

At the Auction Café, Lichfield.

A minor disagreement: 'A total of x' gets around the 'Don't start a sentence with a number' rule in a lot of style guides.

Lots of people somehow get Star Trek and Star Wars mixed up. I wear my 'Nerd' credentials proudly.

Just watched the first 10 minutes of the new Frankenstein movie. I don't care how strong the creature is supposed to be - it's just not possible for it move hundreds of tons of ship trapped in ice. Hard pass.

This week's BibTeX gripe: authors who don't cite references correctly in the text: using 'Smith et al. (Smith et al., 2024)' just shows that they don't know what they're doing.

And this particular paper has no mathematics in it – no justification for LaTeX at all.

#ScientificPublishing