Sign in

Chris Stokes

@nisaccom.bsky.social
212 followers 350 following 2.9K posts
PostsRepliesMedia
Chris Stokes @nisaccom.bsky.social · 20h
Interesting Polanyi quote! MP continues: 'as contemplated by Marxist scientists and their friends. To what conflicts may not have led their “arrogant pretence” to be sole judges of their own preference!' Polanyi had just been dissing Crowther for 'the writing of history backwards'.
010
Reposted by Chris Stokes
New-Cleckit Dominie @ncdominie.bsky.social · 22h
Insofar as I understand this paper, the moral seems to be that maths is easy and translation is hard, which I think every mathematician who spends quality time with a translator will find easy to believe. arxiv.org/abs/2610.08144
arxiv.org
Navier-Stokes lost in translation: Why Lean verification of AI autoformalisation does not guarantee correct natural language proofs
Autoformalisation is increasingly used to verify mathematical texts, including those generated by AI, as in OpenAI's announced proof of blow-up of solutions to the Navier-Stokes equations. In this pro...
253
Chris Stokes @nisaccom.bsky.social · 21h
Asked a colleague about the Navier-Stokes proof last week. He told me how bulky the proof was, so I naturally asked how we could know it was correct. Lean, he said! (And it took me back to sitting in lectures given by Alan Bundy at U of Edinburgh on automatic theorem proving.)
120
Reposted by Chris Stokes
The English Oak Project @thekentacorn.bsky.social · 06/10/2026
Fibonacci in Nature. Fibonacci in Breakfast
1029735
Chris Stokes @nisaccom.bsky.social · 06/10/2026
As she says, she's written a book or rather a 'pamphlet' explaining why big project go awry, so she's got this bit covered!
000
Chris Stokes @nisaccom.bsky.social · 06/10/2026
Bold to throw in Cyprus and 'even the Falkland Islands' too! Quite some system that'll be.
020
Chris Stokes @nisaccom.bsky.social · 05/10/2026
BMW spotted with indicator on!
111
Chris Stokes @nisaccom.bsky.social · 04/10/2026
Before nodding off, you'd remembered that the word for mouse in some Indo-European languages has the additional sense of muscle, esp. in the upper arm, due to the resemblance between them. But the flexing of abs, you thought, is more like the movements of sheep.
020
Reposted by Chris Stokes
Kevin Thornton @lowcarbonkev.bsky.social · 03/10/2026
No, Kent birders shouldn't be we are 'past main migration period' Oct ain't just 'rarity season'. Our obsession with ticks and lists shouldn't outweigh the common stuff, but it does. We've commercialised it, we sell it. There is plenty of chance to see migration in October (here's Redwing c/o BoE)
131
Reposted by Chris Stokes
Emily @emilydawn.bsky.social · 01/10/2026
@hockeyviz.com I found some beautiful data viz for you
7142331024
Reposted by Chris Stokes
BeijingPalmer @beijingpalmer.bsky.social · 02/10/2026
this would be an incredible film if you started it with all the normal serial killer hunting tropes
782815513
Chris Stokes @nisaccom.bsky.social · 01/10/2026
It's in the research section of Ehud's website. Published in Integr. psych. behav. 45, 154–159 (2011). www.ehudlamm.com/huxley.html
ehudlamm.com
Ehud Lamm - Review of Evolution The Modern Synthesis
110
Chris Stokes @nisaccom.bsky.social · 30/09/2026
'Data... shows a rapid descent... just two hours after take-off, descending more than 17,000 feet... in less than two minutes...' www.bbc.co.uk/news/article...
bbc.co.uk
Israel-bound flight diverted after fight between pilots
Despite a brief emergency alert signalling a hijacking, the Israeli prime minister's office confirmed this had been ruled out.
000
Chris Stokes @nisaccom.bsky.social · 30/09/2026
Not excellent at all! Unless the locals report any erasure of the L the moment it happens and Amey paint it back instantly.
110
Chris Stokes @nisaccom.bsky.social · 30/09/2026
'Taking an average, statistically the "L" is safe for an average of 639 days or 21 months.' So in Sheffield, the locals instantly report the vanishing of the L and the highways maintenance contractor responds within seconds to restore it. They do things properly in South Yorkshire.
000
Chris Stokes @nisaccom.bsky.social · 30/09/2026
Et tu, Brute.
010
Reposted by Chris Stokes
New-Cleckit Dominie @ncdominie.bsky.social · 29/09/2026
Source (credit where it's due etc.): mastodon.scot/@gl33p@masto...
071
Chris Stokes @nisaccom.bsky.social · 29/09/2026
An academic, he says, told him how irked he is to get essays to mark from students who've obviously used AI to write them in spite of having signed an agreement to say they will not. Sathi is astonished. How could an academic believe this would be effective?!
010
Chris Stokes @nisaccom.bsky.social · 29/09/2026
A few days after the essay was published, Sathi was a guest on the TLS podcast. Once more, he said nothing of being an AI user himself, let alone of having used it in the production of the essay. Remarkable. shows.acast.com/tlsvoices/ep...
shows.acast.com
Who's in Charge | The TLS Podcast
This week, Anchit Sathi asks what an "author" is in the age of AI; and Keggie Carew on the world of symbiotic slime and friendly fungi.
112
Reposted by Chris Stokes
David Edgerton @davidedgerton.bsky.social · 29/09/2026
Burnham claims to be providing a more serious and thought- through prospectus for Britain, but believing Britain is the world's innovation nation; trusted world leader in AI; leading the next industrial revolution, is neither serious nor new.
062
Chris Stokes @nisaccom.bsky.social · 29/09/2026
That's late too. 'Across the year, we also advanced Research England’s Strategic Institutional Research Funding review (SIRF), completing its SIRF impact evaluation (to be published in early summer 2026).' www.ukri.org/publications...
ukri.org
UKRI annual report and accounts 2025 to 2026
000
Chris Stokes @nisaccom.bsky.social · 29/09/2026
will be announced alongside a summary of the findings from the pilot year. www.ukri.org/councils/res... 2/2
ukri.org
Transparency Programme
The Transparency Programme will gather evidence and provide insights into how higher education providers use formula-based research funding.
100
Chris Stokes @nisaccom.bsky.social · 29/09/2026
The deadline for pilot submissions wasn't all that long ago. I can find two versions—1 & 30 June. But it is late: 'Subject to the feedback from the pilot, the process is likely to be implemented from academic year 2026 to 2027 and 1/2
110
Chris Stokes @nisaccom.bsky.social · 28/09/2026
Global warming.
100
Chris Stokes @nisaccom.bsky.social · 27/09/2026
Bought that Wallace Stevens collected from a secondhand bookshop earlier this month!
110
Chris Stokes @nisaccom.bsky.social · 26/09/2026
I say *for* now in light of what UKRI said about grant assessment modernization a couple of weeks ago, esp. the bit about 'exploring the safe use of AI-assisted assessment'. www.ukri.org/news/ukri-mo...
ukri.org
UKRI modernises grant assessment for the age of AI
UKRI is updating its grant assessment approach to speed up grant decisions, respond to generative AI and ensure we continue backing the best talent and ideas.
000
Chris Stokes @nisaccom.bsky.social · 26/09/2026
There is now (and for now!). RGC 3.1.2 requires you to follow 'Our policy on the governance of good research practice', which includes: 'Use of generative AI tools as part of assessment by reviewers and panelists...is a breach of confidentiality...' Not sure tho when this was introduced.
200
Reposted by Chris Stokes
Colin Carlson @colincarlson.bsky.social · 25/09/2026
David Quammen's essay in The New York Times today closes with this paragraph. It's both a scientific falsehood, and an incredibly dangerous narrative. If you're willing to give me five minutes of your time, I'd like to explain why.
The Ebola virus doesn’t need to spill back. It abides in its reservoir host, and awaits its next opportunity. A virus can’t climb out of a bat or a monkey and fly off in search of human prey. It can only spill over. If humans left the reservoir host alone — not hunting it, not capturing it, not eating it, not exporting it live, not disturbing its forest habitat — the virus wouldn’t re-emerge in humans. It would remain where it is, harming no one. But of course that’s not what happens. And when viruses spill, they often spill into us, because the human population is such a big, busy target.
10529224
Reposted by Chris Stokes
Colin Carlson @colincarlson.bsky.social · 25/09/2026
It's time to stop letting a small number of white American journalists who write airport books about scary diseases set the policy frame for pandemics. And it's unbelievably galling to do it - and this is always how it happens - in the middle of an ongoing Ebola virus epidemic.
315923
Reposted by Chris Stokes
Mic Wright @brokenbottleboy.bsky.social · 25/09/2026
What’s your best sleeping story? Mine is probably falling asleep under a piano at Gare Du Nord after reporting on the Bataclan attacks and being woken up by armed police.
541014
Chris Stokes @nisaccom.bsky.social · 26/09/2026
Woke up in my bedroom at home. At my bedside were three large birds lined up in order of height—an ostrich, an emu and, I don't know, a rhea. I reached out and one of them bit me. Fell asleep. In the morning my parents told me I'd dreamt the whole thing, red finger notwithstanding. Adults are blind.
030
Chris Stokes @nisaccom.bsky.social · 25/09/2026
*Alfred* Koestler?
000
Chris Stokes @nisaccom.bsky.social · 25/09/2026
If they do the research and publish the results as quickly as they did when they looked at UK participation in Horizon 2020, it'll be a touch late for the government's decision making on associating to FP10. The H2020 report came out just after Christmas last year. www.gov.uk/government/p...
gov.uk
Evaluation of UK participation in H2020
010
Chris Stokes @nisaccom.bsky.social · 25/09/2026
AI triage of grant applications! Appears to have been the work of one of EPSRC's Place Based Impact Acceleration Account programmes. www.ukri.org/news/researc...
ukri.org
Research and innovation clusters to boost UK regional growth
From digital healthcare and rehabilitation technology to offshore wind energy, the seven new projects will kickstart economic growth and address regional needs.
010
Chris Stokes @nisaccom.bsky.social · 25/09/2026
Ah, no. A place-based IAA makes much more sense.
020
Chris Stokes @nisaccom.bsky.social · 24/09/2026
Interesting. A network pooling EPSRC IAA contributions from the institutions in the network?
100
Chris Stokes @nisaccom.bsky.social · 24/09/2026
This was not UKRI's work. bsky.app/profile/jpsa...
030
Chris Stokes @nisaccom.bsky.social · 24/09/2026
So organized and executed within your own institution?
100
Chris Stokes @nisaccom.bsky.social · 23/09/2026
There's another call with distributed peer review and it's already open. This one's from EPSRC. Deadline less than a month off. www.ukri.org/opportunity/...
ukri.org
Next generation artificial intelligence (AI): explainable AI
Apply for funding for projects focused on speculative and high-risk fundamental research with the potential to deliver high reward and a step change in the explainability of future AI systems.
000
Chris Stokes @nisaccom.bsky.social · 22/09/2026
It's possible, I suppose, that the applications UKRI sends for review to a team of applicants could be delivered over several weeks. Then they may get more than six weeks to return their reviews.
010
Chris Stokes @nisaccom.bsky.social · 22/09/2026
It's been a long time since my last encounter with a stats textbook. Thankfully I have a swamp of conceptual muddle in my head as a lasting memento.
020
Chris Stokes @nisaccom.bsky.social · 22/09/2026
'Reviewers will have three weeks to complete each assigned review. Reviewers will be assigned eight or four applications to review depending on whether only the the project lead or the project lead and project co-lead are contributing reviews.'
110
Chris Stokes @nisaccom.bsky.social · 22/09/2026
There is a hint. 'Please check that nominated reviewers do not have existing commitments (such as a long period of annual leave) which would prevent them from acting as a reviewer throughout January and February 2027. 1/2
110
Chris Stokes @nisaccom.bsky.social · 22/09/2026
From the how-to-apply section: 'Nominate reviewers 'Word limit: 200 'Please identify up to two members of the team involved who will participate in the distributed peer review process. This should include the project lead and a co-lead, if appropriate.'
100
Chris Stokes @nisaccom.bsky.social · 22/09/2026
From the call text: 'All applications for the ‘Enabling AI adoption across science and engineering’ funding opportunity will undergo distributed peer review (DPR).'
100
Chris Stokes @nisaccom.bsky.social · 22/09/2026
Distributed peer review has now appeared in a UKRI research funding call. It opens in a week. www.ukri.org/opportunity/...
ukri.org
UKRI Enabling AI adoption across science and engineering
Apply for funding for ambitious and exploratory projects bringing together AI and domain expertise. Projects will have the potential to unlock new opportunities for applying AI to research challenges ...
100
Chris Stokes @nisaccom.bsky.social · 22/09/2026
Alice Sara Ott & the City of Birmingham Symphony Orchestra performing Grieg's piano concerto in A minor live on BBC Radio 3 competing with the loud ringing of a phone for about 15 seconds there.
000
Reposted by Chris Stokes
Tony Jackson @jacksont0ny.bsky.social · 19/09/2026
When I was around nine years old I saw The Wombles at Preston Guildhall. One of them fell off the stage (and allegedly broke their arm). It was the talk of our village school for years ;) #rocknroll
1222
Chris Stokes @nisaccom.bsky.social · 18/09/2026
Isn't 2 an empirical question? And isn't the history of the concept, its name and associated research practices of some interest in answering that question?
100
Chris Stokes @nisaccom.bsky.social · 16/09/2026
The tea is English, is it? Must be Yorkshire tea.
010