iiinova — where your digital self keeps evolving

Jarvist Moore Frost @jarvist.bsky.social · 12/10/2026

I had to upload a mugshot for a bio for a conference abstract submission, and realised with slight horror that my I-use-it-everywhere red-jumper photo is dated Feb2017. Pre kids! I look so rested. It's finally, inevitably, happened... phdcomics.com/comics.php?f...

phdcomics.com

Academic Homepage

Link to phdcomics.com

010

Jarvist Moore Frost @jarvist.bsky.social · 12/10/2026

Julia, and Simon Peyton Jones, are my two favourite things in programming, therefore my joy of this will be at least O(n^2).

000

Reposted by Jarvist Moore Frost

JuliaCon 2026 @juliacon.org · 12/10/2026

We're starting our JuliaCon Global '26 video posting with our first keynote by Simon Peyton Jones! youtu.be/uHlT5sFejXM

044

Jarvist Moore Frost @jarvist.bsky.social · 02/10/2026

Still secret after all these years! Quite amazing.

000

Jarvist Moore Frost @jarvist.bsky.social · 02/10/2026

Oh, and here is probably the best original detail: Report of Experiments on Uranium Oxide, 1940-05 FileReference Code: GBR/0014/TMSN 2 This is a typed draft of work carried out at Imperial College... TMSN 5 is CLOSED until cleared by UKAEA at Harwell. archivesearch.lib.cam.ac.uk/repositories...

archivesearch.lib.cam.ac.uk

Report of Experiments on Uranium Oxide, 1940-05 | ArchiveSearch

100

Jarvist Moore Frost @jarvist.bsky.social · 02/10/2026

& p.39 Britain and Atomic Energy 1939-1945 by Margaret Gowing - archive.org/details/brit... Just casually trying to self-sustain fission in South Kensington, as you do!

130

Jarvist Moore Frost @jarvist.bsky.social · 02/10/2026

Ah! "Thomson 'borrowed' Moon in the summer of 1939 for a quick test of the possibility of a nuclear chain reaction in one ton of (natural) uranium oxide mixed with water and/or paraffin wax. The result was negative - fortunately for the experimenters." www.independent.co.uk/news/people/...

independent.co.uk

Obituary: Professor P. B. Moon

Philip Burton Moon, atomic and nuclear physicist: born Lewisham 17 May 1907; Research Student, Cambridge University 1928-31; Lecturer in Physics, Imperial College, London 1932-37; Lecturer in Physics,...

120

Jarvist Moore Frost @jarvist.bsky.social · 01/10/2026

AHHH, do you know anything about the reactor they tried to build in the basement of the V&A in WW2? (I don't think it ever went critical.) An old Prof at Imperial told me about it over dinner (and said there were some things in the IC archive), but I've never been able to find any definite info.

100

Jarvist Moore Frost @jarvist.bsky.social · 30/09/2026

> One experiment involving humans given 140–250g heavy water, corresponding to 0.5% of total body water, showed only temporary symptoms such as vertigo 10/10 mad-up experiment.

030

Jarvist Moore Frost @jarvist.bsky.social · 30/09/2026

Didn't they give heavy water to dogs back in the 1950s to find out what fraction is the lethal dose (about 30% I recall? Presumably messing up enzyme kinetics enough?)

121

Reposted by Jarvist Moore Frost

Robert Wright @reporterrwright.ft.com · 25/09/2026

I see a lot of people complaining about the undoubtedly stressful conditions at present on London's roads. But I see few people complaining about motor vehicles. Yet in the last 24 hours three vulnerable road users have died after being hit by people driving motor vehicles.

311960

Jarvist Moore Frost @jarvist.bsky.social · 17/09/2026

> Academically, our state schools are getting better and better. Children are being let down by junk food lunches Did you see the subtitle? I actually thought it was really interesting, including the comparison with France (£6 government spend per dinner, versus £2.60-3.45 in UK).

100

Reposted by Jarvist Moore Frost

Alex Hern @hern.bsky.social · 14/09/2026

continually maddening that people blame "the water companies" for hosepipe bans when "the water companies" have been begging to build new reservoirs but a few retired QCs in Oxfordshire have used the entire weight of the british planning system to prevent it www.theguardian.com/business/202...

theguardian.com

Thames Water redraws mega-reservoir plan amid local opposition

A new public consultation has opened on the 150bn-litre above-ground White Horse project in Abingdon, Oxfordshire

35385108

Reposted by Jarvist Moore Frost

Moby Dick @mobydickatsea.bsky.social · 12/09/2026

Science! Curse thee, thou vain toy; and cursed be all the things that cast man’s eyes aloft to that heaven

38923

Jarvist Moore Frost @jarvist.bsky.social · 12/09/2026

Now it is a 'mastery' approach, which seems a bit repetitive and boring, but does aim to get everyone to have core arithmetic, estimation etc. skills. The secondary stuff they work on is more enriching. All the national syllabus are online: www.gov.uk/government/p...

gov.uk

National curriculum in England: mathematics programmes of study

The statutory programmes of study and attainment targets for mathematics at key stages 1 to 4.

010

Jarvist Moore Frost @jarvist.bsky.social · 12/09/2026

England has so far held off the decline: secondary schools moved to a 'knowledge rich' curriculum ~10 years ago (i.e. more traditional), and there is an incredible emphasis on reading in primary schools.

120

Jarvist Moore Frost @jarvist.bsky.social · 11/09/2026

2 GW --> 5 million A100s @ 400W each, which is still $50 bn (at $10'000 each) of GPU hardware lit up? Isn't that... about right?

001

Reposted by Jarvist Moore Frost

Dulwich Quantum Computing @dulwichquantum.bsky.social · 10/09/2026

Latest rumors...

Grim reaper meme with IMO gold, My PhD, Unit distance, Riemann zeta 67%, Navier-Stokes, and now knocking on Hodge conjecture, and Birch and Swinnerton-Dyer door.

191

Reposted by Jarvist Moore Frost

Philip Ball @philipcball.bsky.social · 09/09/2026

In other news, thousands of climate scientists say for decades that climate change could kill huge numbers of people and destabilize the world, and experience shows they were right all along, but it's deemed less thrillingly doomy to tackle seriously, so here we are.

413834

Jarvist Moore Frost @jarvist.bsky.social · 09/09/2026

Also I know the token costs aren't real money, but gosh, the science we could get done and the post grads we could train if that money was turned into PhD studentships!

020

Jarvist Moore Frost @jarvist.bsky.social · 09/09/2026

Ah, I hadn't realised that, but makes perfect sense. But I do reckon I could crush 10 tons of basalt a year if I had to. And similarly, it must be 1:1 with the amount of fossil fuels burnt, so if we can do that and transport them to point of use...

100

Jarvist Moore Frost @jarvist.bsky.social · 08/09/2026

Those are quite different though, they are forms of cooling geoengineering, not carbon sequestration.

101

Jarvist Moore Frost @jarvist.bsky.social · 08/09/2026

What is your take on enhanced rock weathering, and other techniques that don't require the preposterous idea of directly filtering the air?

320

Jarvist Moore Frost @jarvist.bsky.social · 06/09/2026

Really? I thought that the study of 'pupil premium' in the UK showed that the extra cash was spent on additional staff (TAs mind), but that it had absolutely no discernable impact on outcomes?

000

Jarvist Moore Frost @jarvist.bsky.social · 05/09/2026

(( It is pitch black. You are likely to be eaten by Ethan Mollick. ))

030

Jarvist Moore Frost @jarvist.bsky.social · 26/08/2026

As someone who spends a lot of time mangling data on the command line, this seems super useful!

000

Jarvist Moore Frost @jarvist.bsky.social · 24/08/2026

That 1/Delta-E bit in the super-exchange effective transfer integral? Yes, as you might guess it is from perturbation theory, and it does blow up at resonance. However, I was shocked that the full non-singular non-perturbative expression just involves taking the determinant of a 3x3 matrix! #_#

$Equation, $V_{1p} = - \frac{V_e V_h}{\Delta E}$ Two button Meme:
Button 1: Diagonalise a 3x3 matrix
Button 2: Perturbative and singular expression
Sweating test-pilot: "THEORETICAL CHEMISTS"

Two button Meme:
Button 1: Diagonalise a 3x3 matrix
Button 2: Perturbative and singular expression
Sweating test-pilot: "THEORETICAL CHEMISTS"

030

Jarvist Moore Frost @jarvist.bsky.social · 13/08/2026

But what is the alternative? Abandon the Oxbridge tutorial obsession / more recent A-level style classroom teaching, and deliver in a more continental style with fewer contact hours? (Traditional lectures / problem sheets / exam.) Or state subsidy? Accept a rationing of places via Teaching Grant?

000

Reposted by Jarvist Moore Frost

David Higham @oldtrotter.bsky.social · 13/08/2026

As the heatwave continues let’s all take a moment to remember another reason why Osborne should never be indulged as a serious commentator. www.theguardian.com/environment/...

theguardian.com

UK scraps zero carbon homes plan

Treasury axes plans to make new homes carbon neutral from 2016, drawing widespread criticism from housebuilders and environmentalists

0207

Jarvist Moore Frost @jarvist.bsky.social · 13/08/2026

I was at a kids 'softplay' at the weekend was horrified to realise that custom AI generated songs now play nonstop through the park, with inane LLM generated lyrics which make no sense: "going on a walk... pack my map and my pencil... and my capstan [!?] and my magnifying ray.[!?]"

110

Jarvist Moore Frost @jarvist.bsky.social · 12/08/2026

Look at the bottom here - all need is a bit bit of cardboard which you cut so you can shove one lens through (making a shadow behind the binoculars), and then project onto a sheet of white paper. skyandtelescope.org/2024-total-s...

skyandtelescope.org

Low-Tech Eclipse Viewing

What to do if you're caught without optics on eclipse day. Low-tech eclipse viewing options from pinhole to binocular projection.

010

Jarvist Moore Frost @jarvist.bsky.social · 12/08/2026

It's literally the same size as the sun itself, when viewed from the colander. That's how pinholes work. Eclipse glasses aren't that impressive either. The next step up is to project an image onto a piece of card, e.g. with one side of a pair of binoculars.

120

Reposted by Jarvist Moore Frost

Philip Ball @philipcball.bsky.social · 08/08/2026

I thought I'd repost this piece quoting the first sentence. Then I thought, no, the first paragraph. No, the first three. And I went on doing that until the final word. www.theguardian.com/books/ng-int...

theguardian.com

‘I hate what AI is doing to the minds and happiness of the young’: Katherine Rundell on the view from the classroom

Education is at a crossroads, argues the author and academic. Should we embrace new technology in the name of efficiency, or is it time to fight back?

411952

Reposted by Jarvist Moore Frost

Dulwich Quantum Computing @dulwichquantum.bsky.social · 07/08/2026

Microsoft is too busy with delivering a useful topological quantum computer by 2029 to worry whether their Majorana qubits actually work. "This notion of having peer reviews and publications doesn't work in this fast-moving world right now. This paper thing is kind of boring now." archive.ph/xGlb7

1252

Reposted by Jarvist Moore Frost

Nature Portfolio @natureportfolio.nature.com · 06/08/2026

I had a problem posting the video associated with this research paper yesterday, but here it is now. 🔭🧪

25311

Jarvist Moore Frost @jarvist.bsky.social · 06/08/2026

& extended to try and model the Feynman-Cohen fractal nodal structure suggested by Zaanen-Kruger, by flow matching to a self-attention (PsiFormer inspired) continuous normalising flow Slater determinant www.youtube.com/shorts/xkxkI...

youtube.com

Zaanen-Kruger Feynman-Cohen nodal surface, CNF, animated

YouTube video by Jarvist Moore Frost

000

Jarvist Moore Frost @jarvist.bsky.social · 05/08/2026

OK, here we go, "2025-010D has been in a highly elliptical (rp≈9,400 km, ra≈377,000 km) geocentric orbit since launch" So could only _possibly_ intersect satellites in geostationary orbits, but with a 32 degree (i)nclination (Table 1), actually ~10,000 km away. arxiv.org/html/2608.00...

arxiv.org

Physical Characterization of Moon Impactor 2025-010D

000

Jarvist Moore Frost @jarvist.bsky.social · 05/08/2026

I'm sorry, but that sounds like the usual Guardian nonsense due to a lack of scientific training. There are no satellites that far out, I'm fairly certain they would have intentionally avoided a 'free return' in the trans lunar trajectory. Apollo stages also impacted, one may still be out there.

100

Jarvist Moore Frost @jarvist.bsky.social · 04/08/2026

N=3, YMMV, room-temperature superconductor by Christmas, etc. It's been a lot of fun though, and there seem ever more ways to develop it further! Probably next I will try and pin down some more of the fractal aspects following Kruger and Zaanen. HEG is a major goal but the PBCs cause headaches.

100

Jarvist Moore Frost @jarvist.bsky.social · 04/08/2026

By following the Bloch equation along imaginary time, the nodal structure can learn itself. Grew out of the fantastic Sign Problems of Fermions workshop at ECT* (Trento, Italy),April2026. I spent the day after in the local mountains, and came back with broken walking boots but a head full of ideas!

View from a peak looking back down the valley.

An alpine meadow, with chain of peaks beyond, and a footpath marker.

Broken walking boots held together with hair ties, standing in the snow in bright sunshine.

110

Jarvist Moore Frost @jarvist.bsky.social · 04/08/2026

We parameterise the density matrix with an equivariant Continuous Normalising Flow (CNF) + invariant Jastrow, built in Julia on Halcyon.jl. The learnt density matrix is valuable in itself: you can directly interrogate it for off-diagonal observables, and effective low-energy models.

100

Jarvist Moore Frost @jarvist.bsky.social · 04/08/2026

How it works: We start at β = 0 with the exact free-particle heat kernel. A regularised Bloch equation residual, gives us pointwise (!) gradients on where the nodal structure is incorrect. Running restricted PIMC with the Worm algorithm, open Worms hop over nodal restrictions to speed ergodicity.

100

Jarvist Moore Frost @jarvist.bsky.social · 04/08/2026

New preprint: SPINDRIFT Rather than infer the maximally entangled ground state wavefunction from nothing, why not watch the correlations condense from the quantum mechanical 'purity' of infinite temperature as we project along continuous imaginary time? arxiv.org/abs/2607.29590

arxiv.org

Spindrift: Learning quantum degeneracy from thermal purity in restricted path integral Monte Carlo

Restricted path integral Monte Carlo (RPIMC) sidesteps the fluctuating Fermion sign problem by confining paths within nodal regions of a trial density matrix, thereby recovering polynomial scaling. Ho...

100

Jarvist Moore Frost @jarvist.bsky.social · 02/08/2026

I've heard! I think in higher ed we are going to have to continue adapting each year to every different year group. They each really had a very different lived experience and set of challenges during COVID / lockdown.

110

Jarvist Moore Frost @jarvist.bsky.social · 02/08/2026

How are you assessing the academic impact though? It is only anecdata, but I very much think it was only the 2024 university intake who had maths and group working skills back to a pre COVID standard.

100