Albert Cardona

How does the brain work? Someday, we'll figure it out. 
Group Leader, MRC LMB, and Professor, University of Cambridge, UK. 
#neuroscience #Drosophila #ScientificPublishing #academia #TrakEM2 #FijiSc #CATMAID #connectomics #connectome #vEM #iNaturalist #entomology
Born at 335 ppm. 
Brains, signal processing, software and entomology: there will be bugs.

Public Key

npub1fg672rmxdzphzhahf2nqt4prsey4xqw69gqj88rheyhnggf3j0sqpr04nu

NIP-05 Address

albertcardona@mathstodon-xyz.mostr.pub

Profile Code

nprofile1qqsy5d09panx3qm3t7m54fs96s3cvj2nq8dz5qfrn3mujte5yyce8cqpz3mhxue69uhhyetvv9ujuerpd46hxtnfduqs6amnwvaz7tmwdaejumr0dsum2f58

Publishing to

relay.damus.io

nos.lol

Author Public Key

npub1fg672rmxdzphzhahf2nqt4prsey4xqw69gqj88rheyhnggf3j0sqpr04nu

Show more details

Last Notes

2026-10-26 21:24:57 UTC

Me: professor at Cambridge university.
LinkedIn: "You're a top applicant for Senior Research Associate in Translational Biophysical Chemistry and Imaging (Fixed Term) at University of Cambridge Vet School"
One wonders, did anybody ever get hired via LinkedIn. Or is it all just one giant bullshit job, a place for managers and recruiters to pretend to work?
After all, it's owned by Microsoft.
#LinkedIn #Microsoft

2026-10-24 08:31:34 UTC

From September 2nd, 2026. Such timing!
"Digital Sovereignty Is Written in PHP"
https://thephp.foundation/blog/2026/09/02/digital-sovereignty-is-written-in-php/
The article says, there is a funding gap between what governments spend in digital sovereignty and the support they provide to foundational software such as PHP itself:
«Germany is spending €108 million to move its federal websites onto a PHP application.»
«The money is the budget for the eleventh version of the Government Site Builder, the German federal administration's standard content management solution: €26.88 million for product development and support, €73.2 million for migration and relaunch, €8 million for operations, phased over four years. More than 80 federal agencies and institutions already use it, and they have built well over 250 websites with it.»
«In 2025, The PHP Foundation received $730,534 in total contributions.»
Reminded of an XKCD comic which clearly the German government hasn't read yet:
https://xkcd.com/2347/
#PHP #DigitalSovereignty #PHP2shell

2026-10-22 16:14:30 UTC

Nothing like humour in these trying times.
"AI disclosure. During the writing of this paper, we had several conversations with Claude, 55, owner of the neighborhood bar in Strasbourg. Unfortunately, none of them was of relevance to the topics of the paper."
https://arxiv.org/pdf/2610.27918

2026-10-21 08:05:48 UTC

Apple, the new unwitting ally of open access non-profit journals, by having an iOS browser that can’t let the reader click on “reject cookies” at a glamour journal website.
Congratulations to all software engineers on both sides.
#ScientificPublishing #webmaster
https://media.mathstodon.xyz/media_attachments/files/117/477/862/562/525/761/original/29f441c5763df196.mp4

2026-10-19 08:32:52 UTC

Open position: Research engineer in Bioimage Analysis. Institut Pasteur, Image Analysis Hub - 2026.
Work with the formidable Jean-Yves Tinevez at Pasteur! What an opportunity.
https://forum.image.sc/t/research-engineer-in-bioimage-analysis-institut-pasteur-image-analysis-hub-2026/122732
#PhDJobs #FediHire #BioimageInformatics

2026-10-08 08:07:09 UTC

Longest mouse brain axon ever reconstructed: 70 centimeters long.
With a soma in the Locus Coeruleus, this neuron releases norepinephrine over most of the brain but specifically skips the cerebellum, brainstem, and spinal cord.
Very readable and accurate press release: https://alleninstitute.org/news/out-of-the-blue-and-into-the-rest-of-the-brain
The paper:
"Topographic structure and function of locus coeruleus noradrenaline neurons", Su et al. 2026 (Jeremiah Cohen's lab)
https://www.nature.com/articles/s41586-026-11026-0
#neuroscience #mouse #norepinephrine

2026-10-01 11:38:58 UTC

`Wondering whether anything has changed since 2023:
"Despite an unprecedented number of women entering neuroscience, and decades-long recruitment and retention efforts, women continue to be disproportionately underrepresented in European academic tenure-track faculty and leadership positions."
"This Perspective focuses on two major career points where women exhibit diminished representation: the transition from postdoctoral fellow to junior professor and the promotion to more senior (tenured) faculty positions."
From: "A European perspective on structural barriers to women’s career progression in neuroscience", by Ashley M. Bourke, Teresa Spanò & Erin M. Schuman, 2023.
https://www.nature.com/articles/s41593-023-01467-5
#neuroscience #academia

2026-09-20 08:51:15 UTC

The words "content creators" and "feed" have, for me, an obnoxious flair to them.
The first one trivializes what goes into art in all its forms, i.e., writing, drawing, painting, sculpting, photographing, and strongly smells of a tasteless software engineer naming the components of a platform built to facilitate interactions between artists and everyone else with the intent of cashing in.
The second term is irredeemably intricate with industrial farm animals bound to a feeding bucket, choiceless, flabby, droopy, bowed.
Surely there are better terms for these, and those who create such systems but not for the purpose of putting themselves as middlemen to draw down value from every interaction could choose more humane, less general terms such as artist, maker, artisan, blogger, photographer, essayist, writer, collector, ... and streams, collections, portfolios, albums, sets, troves, repertoires, posts.
We don't have to accept the language of the oppressors, the names they assign to the rest of us. Our languages are vast and rich, why not relish in all their beauty and precision.
#ContentCreator #feed

2026-08-30 12:05:09 UTC

Induction cooking has so many advantages, surprised that restaurants switching to induction is news. Much cleaner air, faster cleaning, less waste heat, less need for air conditioning, lower running costs. To which I'd add: geopolitical independence, since all you need is solar panels or windmills, plus batteries, to power a kitchen locally. No need to haul fossil fuels from petrostates half the world away just to cook.
https://www.theguardian.com/environment/2026/aug/28/the-chefs-turning-down-the-heat-by-switching-from-gas-to-electric

2026-08-28 09:15:50 UTC

A must-read for anyone who uses LibreOffice and cares about the possibility of continuing to use it. Note:
“Advertising-supported software is funded by attention; subscription-based software is funded by the ability to revoke access; and data-driven software is funded by surveillance. TDF has ruled out all three of these options, leaving only the possibility of donations.”
And if you use LibreOffice at work, get your company to buy it. It’s the right thing to do, in the best interest of the company.
#nevent1q…r4d7

2026-08-25 20:38:53 UTC

Superb Dalmatian wall lizard, Podarcis melisellensis. Its tail is about twice as long as its snout-to-vent length!
http://www.inaturalist.org/observations/394458272
#iNaturalist #herpetology #lizards

2026-08-12 08:24:59 UTC

- reply

@nprofile…ynd9
For Spain beware of past (and present) corruption that led to building construction on top of seasonal streams, that often run dry for years until they don’t, slamming everything on their path with a 5-meter tsunami wave. We call them “riera” (plural “rieres”), also “barranc” (or “barranco” in Spanish).
If you are looking for a place with good food, proximity to both mountains and sea, and many sunny days, the city of Girona (and nearby towns) is hard to beat. Likewise but with less sun, and directly on the coast, Donosti.

2026-08-03 11:54:55 UTC

- reply

@nprofile…rxg6
As a native coastal Mediterranean person living in Cambridge, UK: often Cambridge is way too moist. Mold is a widespread problem.
The graph shows percent rainfall less or more than average, whereas what I'd like to see is an indicator of whether plants are doing well or not, and how close the rainfall is to their critical threshold.
Further, latest I saw the city water distribution network had losses of about 50%. Fix that as a matter of immediate and urgent need?
In other words, "drought" is a relative term, and I posit it's relative to its impact on nature and agriculture, not on the ability to water golf courses or the probability to bring about a garden hose pipe ban.

2026-08-02 10:41:00 UTC

I'm reviewing a grant proposal. Someone I know said "just use an LLM, then read what it says, use what you like, send it in." That is many things but above all irresponsible, and here is why.
I read the grant and wrote my own evaluation. The grant is not exactly in my field of research, so there was quite some looking up of papers, techniques, and jargon in the process. What I wrote is my perspective (what I have been asked to provide) and makes sense to me only because of the leg work and time I put in.
Now suppose the LLM had written an evaluation, with the "correct prompting": "You are a senior neuroscientist and have been asked to evaluate this grant proposal ... write an evaluation of maximum so many words and in a neutral tone without emphatic or bombastic terms, using rigorous language".
No matter what the LLM would have generated, I could not have assessed whether it was correct or not, and hence, I wouldn't be able to send it in. Plus it wouldn't be my perspective.
Had I used an LLM, I would have learned nothing, and sent in what amounts to, possibly, gibberish, or brilliant commentary: I would not have been able to tell either way. That it'd be written in beautiful language and grammatically correct amounts to nothing, or worse than nothing: the appearance of my thought process without the substance.
Took me some hours, that weren't paid. In the face of plummeting salaries in academia, that is no longer OK. My response should have been to decline to review it, to contribute to pushing grant agencies (and scientific journals) to pay for our work, but alas, I am, somehow, an idealist fool. But I don't do that for for-profit journals; there I draw the line nowadays.
#academia #LLMs

2026-08-01 20:59:16 UTC

Proper water management. Shocks me that it sounds too good; everything written here is rather sensible and down to earth:
"Labour MP Helena Dollimore said Burnham should “end the era of extraction” in the water industry"
Turning failed companies into cooperatives:
"Companies that come out of a SAR [special administration regime] would be required to do so as not-for-profit cooperatives, rather than being sold to the highest bidder"
"Coppard said: “Water is an essential utility that simply should not be farmed out to the private sector – that’s a lesson we’ve learned the hard way. A cooperative model would put control back in the hands of people, help keep bills as low as possible, and allow us to finally intervene in the inexorable dumping of sewage in our rivers.”"
A politician talking sense? That's news in the UK!
https://www.theguardian.com/business/2026/aug/01/turn-failing-water-firms-into-not-for-profit-cooperatives-mps-and-mayors-tell-pm
#UK

2026-07-31 20:33:38 UTC

- reply

@nprofile…ryua
Dean's and Fender's song does not sound like material for posterity.

2026-07-25 16:51:18 UTC

A student was telling me, they haven’t read some papers I asked them to read because it takes them a very long time.
Well, let me tell you: it took me 20 days (with 2-4 hours per day) to read a single large paper in my field, to understand the multiple years of work that went into it with its many implications for my own research and for the broader field. 20 days. And it was worth it.
There are no shortcuts.
#academia

2026-07-24 16:17:33 UTC

How to destroy US scientific research: a roadmap. Is there really no one at the helm.
https://science.house.gov/2026/7/full-committee-hearing-unleashing-golden-age-science-ostp-research-enterprise-priorities-fy27
#academia

2026-07-22 07:15:12 UTC

English language as a poor choice for training language models [*you don't say!*]. David Eichler proposes Basque, a language with very explicit consistency rules in its sentence structure that double as constraints for correctness.
"The hypothesis predicts the Basque-first model will show superior performance on tasks that require robust predicate-argument reasoning, independent of surface language fluency."
"We propose treating English — and any other natural language — as a last-mile rendering layer. [...] Lossy, human-readable projections of representations built on the substrate beneath them."
From: "Curriculum Order and the Geometry of Learned Representations: The Basque First Hypothesis", Eichler 2026.
https://zenodo.org/records/21396981
Expects a flywheel (positive feedback loop) rather than model collapse upon using the output of such model as input:
"Models trained on this substrate should generate outputs that are structurally different from the outputs of current models. When asked to produce arguments, we predict they will produce arguments with explicit logical structure. When asked to prove theorems, proofs with traceable inference chains. [...] These outputs — the proofs, the structured arguments, the logical derivations — are *training data.* [...] Reasoning quality compounds."
Quite the claims in this paper. Also:
"We are making a case for a research direction, not announcing results. The results do not exist yet."
#LLMs #LanguageModels #Basque #euskera

2026-07-18 09:35:17 UTC

- reply

@nprofile…63pg @nprofile…qurt
Using ChatGPT for references is among the worst possible uses of ChatGPT; the are plenty of examples of when that went wrong. To on top of it need a copy editor to check whether the references are legit is really something ... first, the editor in chief of Science claims not to know before writing the piece whether the few references cited in the piece exist, as in, he looked for references to back up what he was writing rather than knowing of those references and use them to inform his statements, plus, once queried, didn't know whether they are real or not. Second, everybody who cares uses Google Scholar, OpenAlex, ORCID, and others, specifically built for this purpose, not a random sentence generation machine.

2026-07-17 20:13:21 UTC

My spam filter has such good sense. Filters out of my inbox all my employer's requests to do "training courses" on pearls such as "IT safety", "slavery", and others. All are subcontracted to a shady industry of course providers that feast themselves on these employer requirements, running without any accountability whatsoever as to whether any behavioural change is achieved. A fantastic waste of time and money all around.
Such emails are indistinguishable from phishing attempts. To the point that employers follow up such third-party requests with further emails about how they aren't phishing attempts. 👏
#academia

2026-07-17 20:06:09 UTC

- reply

@nprofile…yx65
Who is this character? Because, didn't Science answer "Can ChatGPT help science writers?" with a "no"?
https://www.science.org/content/blog-post/can-chatgpt-help-science-writers
There was a @nprofile…kkfs video about this:
https://www.youtube.com/watch?v=akHL0DMnHYU&t=34s

2026-07-16 20:39:49 UTC

- reply

@nprofile…px8z
«For a student who used AI to write a paper
Now I let it fall back
in the grasses.
I hear you. I know
this life is hard now.
I know your days are precious
on this Earth.
But what are you trying
to be free of?
The living? the miraculous
task of it?
Love is for the ones who love the work.»
Joseph Fasano

2026-07-16 10:29:07 UTC

"Small circuit principles for learning in large worlds", Raman et al. 2026
https://www.biorxiv.org/content/10.1101/2025.11.13.688216v2
#neuroscience #Drosophila #LearningAndMemory

2026-07-05 22:37:30 UTC

- reply

@nprofile…thcn @nprofile…pfuj
Bees are wasps that forgot how to eat meat [1]; ants are wasps that forgot to fly; sawflies don't know that they are wasps. It's wasps all the way down ...
[1] Except for vulture bees, also known as carrion bees https://en.wikipedia.org/wiki/Vulture_bee

2026-07-05 16:26:05 UTC

Happy anniversary to the NHS, founded on July 5th, 1948. Employs "one in every 25 adults in England", which is rather very little for a healthcare system.
https://en.wikipedia.org/wiki/National_Health_Service#Employment_and_waiting_lists
#NHS #UK

2026-07-04 20:52:31 UTC

Given my success rate (under 50%) in identifying whether a review was written by a bot or by a human, if I had any doubt that reviews in Google Maps, Amazon, or similar are largely fake, now I am somewhat even less trustful.
Try: bot or not
https://lancasteruni.eu.qualtrics.com/jfe/form/SV_77BzuG0OQBlwX7E

2026-07-04 10:31:45 UTC

- reply

@nprofile…gydw
What a photo. And love it when animals incidentally carry uniquely identifying markers, like this masked bee dragging about a dead ant, mandibles clapping its leg:
https://www.inaturalist.org/observations/30098847
#iNaturalist #Hymenoptera #nativebees #entomology #insects

2026-07-04 08:14:00 UTC

- reply

@nprofile…rxg6
In Cambridge (UK), in May, some of our tomato plants froze, and a week later some dried out from excess heat. In the span of ten days. Shockingly all survived, sprouting again (I attribute this to deep roots; deep because of how I plant them between trenches). A roller-coaster that will only become worse. Will see what happens through the summer.

2026-06-23 10:54:26 UTC

- reply

@nprofile…63pg
Chris Russo @nprofile…t7wq I think got a Leverhulme a while back.
And there must be a public list of MRC Career Development awardees, some in neuroscience, that you could email?
#neuroscience #UK #Leverhulme #MRC

2026-06-22 14:28:38 UTC

"Efficient robot navigation inspired by honeybee learning flights", Ou et al. 2026 https://www.nature.com/articles/s41586-026-10461-3
" ‘Bee-Nav’, a highly efficient navigation strategy inspired by the visual learning flights of honeybees. In equivalent robotic learning flights, a tiny neural network is trained to map omnidirectional images to a home vector based on path integration. After learning, the robot can fly far away from home, come straight back using path integration and cancel integration drift using the visual homing network"
"In real-world indoor and outdoor experiments, a small drone successfully returned to within 0.5 m of home for 100% of 30–110-m flights and 70% of 200–600-m flights in windy conditions, using 3.4-kB and 42-kB neural networks, respectively."
Honeybees still do better, but hey, it's a good start. Now imagine we had the whole honeybee #connectome mapped.
#neuroscience #robotics #honeybees #drones

2026-06-22 08:05:06 UTC

The lab of Xiaotang Lu is hiring, at University of Illinois. From her lab's webpage:
«Three Postdoc positions are immediately open
a) Biochemistry: experience in protein engineering, molecular probe design, preparation, and the application of advanced imaging platforms to complex tissue samples.
b) Neurobiology: experience in systems neuroscience, circuit neuroscience, and related diseases.
c) Computational science: experience in large imaging data segmentation and analysis, and/or computational protein design.»
https://neurolab.chemistry.illinois.edu/recruit
See also her research page, on the chemistry of #connectomics
https://neurolab.chemistry.illinois.edu/research
#neuroscience #PhDJobs #FediHire

2026-06-19 11:43:34 UTC

- reply

@nprofile…ytjz
For anyone wanting to use Chrome, there is still the ability to block ads system-wide, for example with hBlock in #Ubuntu #linux :
https://ubuntuhandbook.org/index.php/2025/03/block-ads-trackers-system-wide-ubuntu/
There's also #AdGuardHome but I haven't tried or read much about it:
https://github.com/AdguardTeam/AdGuardHome#comparison-adguard-dns

2026-06-17 09:00:50 UTC

Dendrites are what machine learning is missing. (And inhibitory neurons, but hey, one at a time.)
"What can a neuron compute", by Ido Aizenbud, David Beniaguev, Noam Pnueli, Idan Segev, Michael London, 2026
https://www.biorxiv.org/content/10.64898/2026.06.08.730984v1
All about dendrites. Extending prior work by Beniaguev et al. 2021 and Jones and Kording 2021.
#neuroscience #CompNeurosci

2026-06-15 13:17:56 UTC

- reply

@nprofile…az86
How does Deepkseek-4 compare to Gemma4 and qwen3.6?
Related: what's the best way to load a graph into an LLM?

2026-05-31 16:14:34 UTC

#SilentSunday #Scotland
https://media.mathstodon.xyz/media_attachments/files/116/670/072/691/541/323/original/badf9d01d517baba.png

2026-05-25 20:52:30 UTC

Today in random plaques inside St. Giles’ cathedral, Edinburgh:
1/n
#chloroform #Edinburgh
https://media.mathstodon.xyz/media_attachments/files/116/637/188/759/070/591/original/8e3af34b63d327dc.png

2026-05-24 20:48:52 UTC

- reply

@nprofile…y789
I wish I could buy this sort of ticket 6 months in advance.

2026-05-22 09:30:54 UTC

Can someone recommend anti-midges clothing for Scotland? Ideally there is a shop in Edinburgh or nearby that sells them?
Also repellents, which ones are effective for midges?
(Search engines are showing me clothes for little people, and that just about summarises the state of internet search today.)
#Scotland #midges

2026-05-21 13:57:15 UTC

- reply

@nprofile…pfuj
Diptera – two wings – would have gotten you there faster.

2026-05-18 14:23:09 UTC

A lot more interesting than I expect it to be, to the credit of the excellent @nprofile…sl5y
#nevent1q…xaew

2026-05-18 12:40:15 UTC

- reply

@nprofile…az86 HN interest in PSOS all of a sudden ... sorry for your loss.
http://www.csl.sri.com/users/neumann/psos.pdf

2026-05-15 07:50:39 UTC

So many chipping in with ideas for ID'ing this artifact. Outstanding community here in the #fedivers !
#nevent1q…6kxn

2026-05-08 11:38:30 UTC

- reply

@nprofile…63pg
Wishful thinking, comfort voting for neighbors, a desire for a change of some sort. Because everyday life became untenable and even the so-called left-wing parties are in cahoots with the powerful and the rich.
What such voters don't know yet is that their everyday life will become worse still by voting the far right.

2026-05-05 21:04:15 UTC

So much talk about data, but no funder wants to support databases.
FlyBase is in trouble, yet a key reference in biology research.
Raising funds to support curating connectome datasets, including their large, nanometer-resolution volumes, is hard. Yet then evaluators point out that we aren't doing enough to make it easy to access – yet with what funds?
And likewise for every scientific dataset ... either there is an internally funded institute behind it (and even then) or it is hell.
Everyone wants new data, but nobody seems to pay attention to the need for old data to support the generation and interpretation of new data!
#academia

2026-05-05 20:44:13 UTC

5 days left to apply to open position in my lab in Cambridge, UK, for a software engineer / data scientist in connectomics research.
#FediHire #PhDJobs
#nevent1q…cvd8

2026-05-05 20:42:08 UTC

"This is an ongoing list of quality, Open Access (OA) journals in the social sciences, who won’t sell your stuff to AI companies, and don’t charge you fees."
https://ubisurv.net/2024/09/04/radical-open-access-in-the-social-sciences/
(updated as of May 2026)
Curated by David Murakami Wood https://www.uottawa.ca/research-innovation/centre-law-technology-society/people/murakami-wood-david
#ScientificPublishing #SocialSciences

2026-05-05 20:36:53 UTC

The International Brain Laboratory is hiring:
Open positions:
* Neuroscience Community Engineer
* Neuroscience Research Software Engineer / Data Scientist
https://www.internationalbrainlab.com/opportunities#jobs
Read about the IBL:
https://www.internationalbrainlab.com/faq-1
#FediHire #PhDJobs #neuroscience #IBL

2026-04-27 12:18:19 UTC

- reply

@nprofile…v2v2
If only man pages had examples of typical usage in them. I could never understand why not.

2026-04-24 18:10:08 UTC

- reply

@nprofile…49as
What a find. Crab spiders always take a moment to see them ... often after realizing the bee couldn't be so still for so long.

2026-04-24 15:04:47 UTC

2+2 years position for a software engineer / data scientist with serious programming chops in my lab at the #MRCLMB to develop tech for #connectomics and #neuroscience :
https://fa-evzn-saasfaukgovprod1.fa.ocs.oraclecloud.com/hcmUI/CandidateExperience/en/sites/UKRI-Careers/job/2250
Please do share with suitable candidates, and feel free to reach out via email to discuss details.
Deadline for application: May 10th.
Starting date: any within 2026.
Requires Masters plus industry experience, or a PhD in a suitable field – and demonstrable experience in software engineering.
Please help disseminate – either here or by copy-pasting onto emails or messages. Thank you!
#PhDJobs #neuroscience #SoftwareEngineering #jobs #ScienceJobs

2026-04-22 21:26:02 UTC

1. Invent an illness;
2. Put up a couple of preprints about it;
3. See AI companies' web crawlers swallow it hook, line, and sinker whole;
4. LLMs start telling users that the condition is real. WTF.
5. See published papers cite them, likely from authors using LLMs to write them. WTAF.
https://www.nature.com/articles/d41586-026-01100-y
#LLMFail #LLMs #chatbots #Bixonimania

2026-04-08 11:15:41 UTC

- reply

@nprofile…ryua @nprofile…v3r4
Esthetics are important. We have to live with these things.

2026-04-07 21:28:43 UTC

"In the UK, a quarter of total university income is generated by international students, who pay much higher fees than home students. To attract them, British universities rely on vast, largely unregulated networks of agents operating primarily in Asia and Africa."
"The reason British universities spend so much to attract international students is simple: if they were to rely solely on home students, they would go bust."
What could possibly go wrong ... with university finances. This system is not well-grounded, not reliable or stable in any way or form.
https://www.theguardian.com/education/ng-interactive/2026/apr/07/brutal-reality-of-life-as-a-foreign-student-in-the-uk
#UK #academia

2026-04-05 20:02:04 UTC

- reply

@nprofile…px8z
In the UK alone, every person, adult or child, subsidizes car drivers to the tune of £800/year, whether they themselves drive or not. That's the current cost of infrastructure and public services used by cars and their drivers.
Reads cheap until one thinks, suppose that was allocated to cycling infrastructure and public transport instead ...

2026-03-30 08:28:27 UTC

- reply

@nprofile…v2v2
Did you use the credit card trick.

2026-03-29 19:17:26 UTC

#SilentSunday #Bloomscrolling #CambridgeUK #blossom
https://media.mathstodon.xyz/media_attachments/files/116/314/063/594/095/635/original/5549a606a5d5cec1.png

2026-03-29 09:40:44 UTC

- reply

@nprofile…az86
In that case, why not make your own CSS class, or arbitrary HTML tag ID.

2026-03-29 09:39:40 UTC

- reply

@nprofile…x9tq
The government is the representative of a sovereign state. Doesn't need to justify severing ties with a company, much less one known for its exploitative practices. If anything, the government ought to emit a public apology for having contracted out work to Palantir in the first place.

2026-03-29 09:38:22 UTC

- reply

@nprofile…az86
I don't know, but explicitly labeling them as such would put you ahead of 99.9% of the web in honesty.

2026-03-28 08:42:55 UTC

- reply

@nprofile…x9tq
While surely the war is stressing the supply chain, the UK by design is always weeks away from shortages of one kind or another. The UK's is not a self-sufficient society in any shape or form. Likewise for most if not all Western countries.
The shock to me was when the UK voted for Brexit when 50% of its food – I mean, its FOOD – came from the EU. How far can manipulation go to make people not know where and how their food comes from, and to then vote against their very own interests.
There are good reasons why, e.g., France heavily subsidises its farmers.
#UK

2026-03-26 23:08:40 UTC

- reply

@nprofile…px8z
But back to humans ...
"Women used to dominate the beer industry – until the witch accusations started pouring in" (2021)
https://theconversation.com/women-used-to-dominate-the-beer-industry-until-the-witch-accusations-started-pouring-in-155940
«Humans have been drinking beer for almost 7,000 years, and the original brewers [1] were women. From the Vikings to the Egyptians, women brewed beer both for religious ceremonies and to make a practical, calorie-rich beverage for the home.»
«From the Stone Age to the 1700s, ale – and, later, beer – was a household staple for most families in England and other parts of Europe. The drink was an inexpensive way to consume and preserve grains. For the working class, beer provided an important source of nutrients, full of carbohydrates and proteins. Because the beverage was such a common part of the average person’s diet, fermenting was, for many women, one of their normal household tasks. [2]»
[1] https://www.womenshistory.org/articles/women-and-beer-forgotten-pairing#:%7E:text=Women's%20involvement%20in%20brewing%20beer,in%20Mesopotamia%2C%20and%20possible%20earlier.&text=Historically%20women%20were%20involved%20in,also%20brewed%20their%20own%20beer.
[2] https://www.atlasobscura.com/articles/women-making-beer
#beer #fermentation

2026-03-26 23:04:16 UTC

- reply

@nprofile…px8z
Not only human's:
«Yes, there are very real dangers to drunkenness among animals, yet there seems to be no shame attached to it. Except perhaps for honeybees. They are nature's hardest drinkers capable of knocking back 100% proof, or as one researcher remarked: "I know of no other organism that drinks pure ethanol, not even a college student."»
«In Australia, the summer heat ferments nectar and, according to the entomologist Errol Hassan, bees all too easily slip into alcoholism. "The bees can't find their way back to the hive, bang into trees or simply fall to the ground," he says. A drunken bee often faces disgrace back at the hive, ostracised by its teetotal fellows. "They are not allowed in by the guard bees at the entrance to the hive," Hassan says. "They are pushed away from the landing platform or attacked." And with good reason. If their fermented nectar was turned into alcoholic honey, it could end up paralysing the entire colony.»
https://www.theguardian.com/science/2001/dec/13/research.highereducation1
#honeybees #alcoholism #alcohol

2026-03-26 13:58:49 UTC

Snow flies now less mysterious but even more fascinating:
"Coordinated molecular and physiological adaptations enable activity at sub-freezing temperatures in the snow fly Chionea alexandriana", Capek et al. (Stensmyr and Gallio labs).
https://www.cell.com/current-biology/fulltext/S0960-9822(26)00247-2
How they stay alive, balancing resource expenditure and the need to remain not frozen:
"direct measurements reveal that snow flies produce brief bursts of endogenous heat in response to cooling at sub-freezing temperatures, indicating active thermogenic capacity."
And as always, lots to learn, in this case for medicine:
"our results demonstrate that Chionea has evolved mechanisms to cope with high levels of reactive oxygen species (ROS), a byproduct of mitochondrial activity and a hallmark of cold exposure."
#Snowflies #genetics #ColdAdaptations #thermogenesis #Chionea

2026-03-25 08:35:15 UTC

- reply

@nprofile…px8z
Except, US citizens can and do travel through Europe with Meta's glasses on ...

2026-03-24 11:54:57 UTC

Postdoc position: Career Returner Fellowship
"The LMB is proud to launch a three-year fully funded fellowship for scientists who have had a career break of 12 months or more."
"At the LMB, we understand that stepping away from research – even briefly – can affect career progression. That is why we have created a dedicated postdoctoral fellowship to support researchers who have taken a career break of 12 months or more, for any reason, for example parental leave, health reasons, caregiving, a career change or other personal circumstances."
https://mrclmb.ac.uk/careers-and-people/postdocs/career-returner-fellowship/
Happy to discuss applications to our lab in #Drosophila #connectomics, #ElectronMicroscopy or #ComputerVision applied to #BioimageInformatics.
#MRCLMB #PhDJobs #neuroscience

2026-03-19 22:41:52 UTC

Reflecting on eLife's new publication model, 3 years in: "the most important thing we have learnt is that our new approach to publishing works. Authors, reviewers and editors routinely tell us that they have had a more constructive experience with the new approach."
https://elifesciences.org/articles/110392?_hsmi=98220609
Proud to be an eLife editor. eLife's publication model gets the best from everyone:
* from authors, who remain in control and can reply to reviewers without fear and without being overly apologetic or sycophantic;
* from reviewers, who engage constructively in a semi-anonymous way (they aren't anonymous neither to each other nor to the editors, all practicing scientists in their field), knowing that their comments are suggestions, not mandates for authors;
* and from editors, who don't have to deal with any nastiness from any party, everybody being far more relaxed that I've seen in any other journal, concentrating their efforts in the scientific content.
1/3
#eLife #ScientificPublishing

2026-03-19 21:06:37 UTC

For fruit fly and vision enthusiasts and specialists:
"Electrical synapses mediate visual approach behavior", Frighetto et al. 2025 (from Mark A. Frye and Larry Zipursky's labs)
https://www.biorxiv.org/content/10.1101/2025.10.14.682373v1
"Our findings reveal mechanisms underlying visual approach, and highlight the interplay between electrical and chemical neurotransmission for rapid object detection and action selection."
#neuroscience #Drosophila #vision #GapJunctions

2026-03-09 23:49:38 UTC

- reply

@nprofile…yx65
Snark aside, I find peer review helpful, to a point. Human peer review, which is about ideas, vision, concepts, and suffiency of methods and data, not about pretty turns of sentence.

2026-03-09 21:23:28 UTC

- reply

@nprofile…yx65
Loss of what the point of the paper was.

2026-03-09 20:17:56 UTC

- reply

@nprofile…63pg
I laughed so hard when I saw that headline. Grifters gonna grift when the marks dole out the dough. To hear a politician talk about "AI" ... it's either pathetic or embarassing, and often with deadly consequences.

2026-03-03 15:38:38 UTC

I keep hearing about the "slopocalipse" coming to scientific publishing, and I can't but think: it was already happening before LLMs became "good enough" at writing academic prose. What hasn't been addressed is the root cause: the incentives. The publish-or-perish approach to science evaluation, for grants and positions.
Want a solution? Limit publications to one per author per year. You can publish more when in collaboration, i.e., make it fractional. The end goal: make people think what brick they want to supply to the edifice of science. As a bonus, *reward* scientists when they publish *less than* one paper per year.
And the business models based on journal APCs can all die a sudden death now.
#ScientificPublishing #academia #slopocalipse

2026-02-24 09:50:12 UTC

- reply

@nprofile…v2v2
The Signal desktop app works very well – except when on a call the chat isn't available to paste links or drop comments for later. Is there a workaround? @nprofile…kyv9

2026-02-10 21:29:40 UTC

- reply

@nprofile…px8z
There's always a small group who doesn't. As always, the novella "Profession" by Asimov comes to mind ...
https://en.wikipedia.org/wiki/Profession_(novella)

2026-02-10 20:09:57 UTC

- reply

@nprofile…px8z
Some of us still understand the machine, still read the code of the dependencies, still know how to fix it all when it breaks, and can write code to save the day – in my everyday, to handle hundreds of terabytes of data with enough efficiency to derive sensible results with an average annual compute budget smaller than the purchase price of a single H200 GPU.
I understand the sentiment, but, there was always such noise. Frameworks du jour, abstractions upon abstractions. Some of us still learned to write low-level code and to care about the inner workings and difference between registers, CPU caches, round-trips to RAM, and disk caches. The analogy of the food in the fork, in front of you in a bowl, in a box in the fridge, in the market, and sowing the seeds for harvesting 6 months later still rings true and makes sense, to many of us.
What has perhaps happened is that those who never cared much now can code, or rather, are vibe coding, which are entirely different things but lets them get stuff done. Good for them, if only it wasn't boiling the oceans over for mundane stuff like writing a school essay or crafting an email reply.

2026-02-08 23:18:35 UTC

One of those days when the trade winds [1] ("vents alisis" in Catalan) are clearly delineated:
https://earth.nullschool.net/#2026/02/08/2200Z/wind/surface/level/orthographic=-39.55,21.19,1451
[1] https://en.wikipedia.org/wiki/Trade_winds
#geography #winds

2026-02-06 14:17:34 UTC

- reply

@nprofile…xuzd
And the whole of UKRI, and the Wellcome Trust, and so many other British institutions. They are hopelessly addicted to Microsoft Teams.

2026-02-03 11:41:32 UTC

- reply

@nprofile…px8z @nprofile…vh6n
Thanks, indeed, a nice summary of key takeaways from Anil Seth's essay https://www.noemamag.com/the-mythology-of-conscious-ai/ There's a lot to agree with there, including all the assumptions and bad metaphors that those who equate brains with computers repeatedly fall into.
A few points that come to mind:
* The software/hardware separation isn't as clean in computers either, from the CPU microcode to the OS that it runs. In humans, I don't have an analogy for microcode, but languages change the way we perceive the world, and those of us who speak multiple languages do feel sometimes that we express a different character when speaking in a different language.
* On time, computers operating under real-time constraints are a lot closer to what brains do. The optimal outcome is not the best possible one or a designated one, but rather, that which can do the job in the time and resources allotted.
* Operating systems, and broadly most software, operates on an event dispatch loop, whereby the system awaits an event and then reacts, followed by a restart of the loop body. Having the loop not stop is a key characteristic of our cerebral cortices, where most inputs to any one neuron arrive from other cortical neurons, with as little as about 6% of inputs to some select neurons (in layer 4) arriving from external sensory stimuli via the thalamus (https://onlinelibrary.wiley.com/doi/full/10.1002/cne.22133 ) . A recurrent neural network is not that different, conceptually, bridging the huge distance of self-driven input seeking.
* Biological systems are entropically favored https://arxiv.org/abs/1908.08374
* The latest technology is never a good metaphor for the brain, as Matthew Cobb's book details over and over: https://profilebooks.com/work/the-idea-of-the-brain/
#neuroscience

2026-02-02 13:53:56 UTC

- reply

@nprofile…xuzd She does make the distinction between e-bikes and e-motorcycles.

2026-02-01 22:20:16 UTC

Pavlovian conditioning in the (urban) wild:
https://idiallo.com/blog/teaching-my-neighbor-to-keep-the-volume-down

2026-02-01 13:36:58 UTC

- reply

@nprofile…pfuj
At what point will moderation be necessary?

2026-02-01 10:08:25 UTC

- reply

@nprofile…x9tq @nprofile…v2v2 @nprofile…mrcf
If that's the desirable action, what will create the conditions for it? Because anyone in power will only move if they see it as a win for themselves.

2026-02-01 10:02:32 UTC

- reply

@nprofile…x9tq @nprofile…v2v2 @nprofile…mrcf
So how do we create the necessary circumstances of available information and public opinion to shift government policy towards reinstating professional schools? How can that be made acceptable and desirable to those in charge of now called universities that used to be polytechniques? And to the government to supply the funding?

2026-02-01 09:52:38 UTC

- reply

@nprofile…x9tq @nprofile…v2v2 @nprofile…mrcf
And more than that, a change in perceptions. For when I ask undergraduates here at Cambridge University why are they here, to learn or to get a credential, almost all say the latter. None of them should be here, they should be in a professional school where the credentials will match the jobs they seek.

2026-02-01 09:46:13 UTC

- reply

@nprofile…v2v2 @nprofile…x9tq @nprofile…mrcf
What's more, without the earlier UK policies that enabled anyone from a EU country to immigrate to the UK regardless of Schengen status the UK's trades like plumbers, electricians, and much more would be in tatters. Now that innane politics have closed that avenue, there's a time bomb awaiting.

2026-02-01 09:41:43 UTC

- reply

@nprofile…v2v2 @nprofile…x9tq @nprofile…mrcf
Indeed, the comparison in professionalism and abilities of the workforce between the UK and countries that have kept their work schools vs universities separate is stark.

2026-01-30 19:57:11 UTC

- reply

@nprofile…y789 There surely are quite a few about the promises made about Spanish trains and the Catalan railway network ...

2026-01-30 16:04:42 UTC

Postdoc position open in my lab (UK) to study a disease model in #Drosophila with comparative connectomics:
https://mrc.tal.net/vx/mobile-0/appcentre-int/brand-3/spa-1/candidate/so/pm/1/pl/3/opp/2765-Postdoctoral-Scientist-Neurobiology-Dr-Albert-Cardona-LMB-2765/en-GB
My lab:
* our page: https://syn.mrc-lmb.cam.ac.uk/
* institutional page: https://mrclmb.ac.uk/research-leaders/albert-cardona/
#neuroscience #PhDJobs #Parkinson

2026-01-29 13:47:34 UTC

"The official European joke", by @nprofile…m0rx
https://ploum.net/the-european-joke/index.html
Hilarious. Because it's tragic. Because it's true.
#jokes #EU #Europe

2026-01-29 09:47:41 UTC

- reply

@nprofile…xuzd Buses free during rush hour – why shouldn't they be, they save the NHS millions by drastically reducing air pollution and increasing frequency of light exercise. These buses are an essential complement to the shuttle buses from the Park and Ride.

2026-01-29 09:39:25 UTC

- reply

@nprofile…xuzd
I am planning to setup a webpage for this proposal: close the city to any entering cars in Cambridge from 8: AM to 9 AM, while simultaneously making the Park and Ride free during that hour. Want to drive into town before 8 AM or after 9 AM? Sure, please do*.
The move would free up all the streets for buses to zip by, driven by professional drivers in well-maintained vehicles full of sensors – buses are not a menace. Plus one bus replaces upwards of 60 stopped, non-moving cars stuck in their own traffic.
A ~5 minute wait at the Park and Ride would be just fine. In practice would be even less than that, given multiple buses running the route simultaneously.
The removal of cars from just this one hour would make the afternoon rush hour likely disappear altogether.
Professional services like plumbers, electricians, food delivery – they could all drive before 8 AM or after 9 AM, largely without any traffic. And for food delivery over multiple houses, once inside the town by 7:55 AM, they would service multiple houses in their 1-hour slot. Nothing lost.
Everybody would win. There are no downsides.
* To enable clearing of prior traffic, the ban will likely have to start by 7:45 – but that's on car drivers normally entering between 8 and 8:15 not choosing to now all enter by 7:45, leaving streets clogged until 8:30.
#Cambridge #UK

2026-01-29 09:31:27 UTC

- reply

@nprofile…xuzd Exactly. We are talking here about the other 99%.

2026-01-29 09:17:26 UTC

- reply

@nprofile…xuzd
And to further add: nobody needs to drive into town *at the speed of a walking toddler*. Even a young child walking would traverse the distance from the Park & Ride facilities to the centre of town faster than the queuing, idling, helpless cars.

2026-01-29 09:03:24 UTC

- reply

@nprofile…xuzd
The Cambridge Shared Equity Scheme is ... suboptimal, to not write any major bad words. Plus, the housing problem in Cambridge was entirely created by the University and the colleges with their Science Park and various other money-making schemes that resulted in attracting many highly paid jobs to a city without the housing base to absorb them. Compounded by the massive drop in purchasing power of the academic salary over the last 30 years, a problem also created by the university – the same number of undergrads could be serviced by half the existing faculty at double the salary, at least in the School of Biological Sciences of the University of Cambridge.
Leaving is an option. That is also why retirement isn't contributing to new job openings as much, because what contributes more is professors leaving for better paid posts elsewhere. Who stays behind are those who signed up early enough that they could solve their housing before it became a severe problem.
In between, lots got caught into crushing situation. I'd like to see statistics of faculty by age; I sense a gap between 40 and 55 – those who left early, with those younger continuing the trend of leaving early. At least some departments are a bimodal distribution of young, short-lived posts (even when tenured) and near-retirement professors.

2026-01-29 08:49:28 UTC

- reply

@nprofile…xuzd We could also close the roads during rush hour. Nobody needs to drive into a town; that's what Park & Ride services are for. And unimpeded by car traffic, shuttle buses from the car park to the city centre would fly, be on time and at high frequency.
#cycling

2026-01-29 08:42:55 UTC

"Should professors be forced to retire?"
https://www.nature.com/articles/d41586-026-00055-4
The data points to retirement age limits not resulting in new faculty positions at the same rate as faculty posts become available. A pungent point, since retirement age limits were justified on this basis. Further, non-academic posts (i.e., administrative and support posts) are increasing at universities. Plus:
“In many disciplines, your abilities don’t fall off a cliff at the age of 69”, says Oliver Linton.
That is true, but: our abilities often fall off a cliff *far before* that. Sometimes, even decades before that. In particular, our ability to perform research work with our own hands, and to pivot into new research lines, or entirely different fields when the present one is exhausted.
While retirement per se doesn't matter, what matters is the capability of the system to engage in new research approaches, and to explore so far unexamined questions or newly opened fields of research.
How to square the circle? Grant emeritus status to any academic who wants it, at any age. Salary stops being the university's problem; laboratory space frees up for others; and if any grant body wants to provide support funds to the emeritus, by all means; or if the research is engaging enough that other faculty members want to support it in collaboration, absolutely. The relief in voting power while retaining talent, experience, a no-fucks given attitude to questions in seminars, and often the willingness to lecture, at essentially no cost, is a no-brainer.
#academia

2026-01-26 20:41:26 UTC

- reply

@nprofile…px8z
On having to be tech savvy: I'd like to say no, but I am not sure. I am too far into the rabbit hole of tech to gauge the outsider perspective. Likely you will have to at least be comfortable running commands from a terminal.
Yes, a base MacOS install is necessary to install Asahi Linux. Just shrink its partition (as per instructions) a lot, so that not much is lost. But once you are booted in Asahi Linux you never have to go back to MacOS for anything. You can, but you don't have to.
User-friendly: as much as any Fedora Linux. Behaves the same way. The default Gnome desktop is very functional, with gestures and all. Feels very natural to me.
Hope this helps.

2026-01-26 19:26:42 UTC

- reply

@nprofile…px8z
I use Asahi Linux, works great.

2026-01-26 08:59:41 UTC

- reply

@nprofile…nwd6
The original paper:
“Three decades, three climates: environmental and material impacts on the long-term reliability of photovoltaic modules”, by Ebrar Özkalay et al. 2025.
https://pubs.rsc.org/en/content/articlehtml/2025/el/d4el00040d
A study done in Switzerland:
“This study analyses the long-term performance of six PV systems in Switzerland over three decades, with more than 20 years of high-quality monitoring data.”
#SolarPanels #RenewableEnergy

2026-01-21 23:40:53 UTC

- reply

@nprofile…49as The start of the run to mid-term elections.