Moritz Negwer

@moritz_negwer@mstdn.science
988 Followers
2.6K Following
13.8K Posts

Neuroscientist by training, tinkerer by nature. Now scanning transparent mouse brains with light-sheet microscopes. Microscopy, clearing, data crunching, tinkering.

Working as postdoc at @radboudumc with Nael Nadif Kasri and Corette Wierenga, looking at neuron-distribution differences in mouse models of ASD.

Married, father of two. Toots in English, German, Dutch. Boosts a lot.

opted into tootfinder for full-text indexing. Check it out: https://www.tootfinder.ch/index.php?join=1

anyone need some maps or cad done remotely? i need to make some money lmao

If you drive in #France and suddenly see something milky on the roads, it's not a new farmers' protest. It's called "lait de chaux".
This #whitewash consisting of lime and water is a try to protect the bitumen in the roads from melting due to more and more heatwaves: https://www.franceinfo.fr/environnement/evenements-meteorologiques-extremes/vagues-de-chaleur-canicules/canicule-du-lait-de-chaux-repandu-sur-les-routes-pour-eviter-la-chaleur-du-bitume_7326369.html The streets are about 10 degrees cooler due to the white colour.
This week-end, we will have up to 37°C.

#climateDiary #ClimateAdaption #ClimateCrisis #white #colours #heatwave #road #roadSafety

Canicule : du lait de chaux répandu sur les routes pour éviter la chaleur du bitume

Dans l'Allier, un mélange d'eau et de chaux est répandu sur les routes pour faire baisser la température du bitume. Les opérations ont commencé avant la vague de chaleur du vendredi 20 juin.

Franceinfo

"The writing is getting better. The ideas are getting worse."

Great post on the impact of GenAI on students' ability to learn how to write and therefore how to think

https://www.forkingpaths.co/p/the-death-of-the-student-essayand

The Death of the Student Essay—and the Future of Cognition

One professor's reflections on the end of an era, as AI tools such as ChatGPT have murdered the student essay (RIP). Here's why that threatens the future of human cognition—and how to save ourselves.

The Garden of Forking Paths

A beautiful pair of Solitary Rugose Coral fossils.
Eroded into perfect cross-sections, you can see their internal structures.
County Donegal, Ireland.

Cormacscoast.com walking tours

#wildatlanticway #walkingtours #discoverireland #keepdiscovering #fossil #irishfossils #coral #rugosecoral #Ireland

It’s live: The Anti-Autocracy Handbook- The Scholars’ Guide to Navigating Democratic Backsliding. The need for this is self-evident given current events around the world and in particular in the U.S. The team of authors includes experts from relevant fields and several authors with first-hand experience of living under autocracy.
 
a pdf of the handbook is freely available for download at the short link https://sks.to/autocracy and there is an associated Wiki that will continue to be expanded and updated.
 
Please share!
The Anti-Autocracy Handbook: A Scholars' Guide to Navigating Democratic Backsliding

The Anti-Autocracy Handbook is a call to action, resilience, and collective defence of democracy, truth, and academic freedom in the face of mounting authoritarianism. It tries to provide guidance to scholars navigating the growing global trend of democratic backsliding and autocratization, in particular in the U.S. To this end, it sets out how autocracies often follow a common playbook, built around the “3 Ps”: populism, polarization, and post-truth. Leaders present themselves as voices of “the people” against “corrupt elites”, inflame societal divisions, and undermine facts to avoid accountability. This leads to a cascade of dangers for scholarship, including censorship, restrictions on funding and research collaboration, and even violence. The Trump administration serves as a contemporary example, with policies that curtail international scientific cooperation, revoke research grants, and suppress studies related to public health, climate change and minority issues. Because open inquiry and dissent are central to science and academia—qualities antithetical to authoritarian control—academia is often among the first targets of autocrats. To help scholars resist authoritarian developments, the handbook highlights both historic and contemporary measures aimed at attacking scholars, their institutional environments, and their scholarship. The handbook also sets out a framework for action based on personal risk level—low, medium, high, or extreme. This is designed to help scholars think about their own risk and purposefully choose actions in line with it. The handbook considers tools for enhancing digital safety and highlights the importance of ongoing documentation, preserving imperilled data, and creating distributed archives as a defence against erasure. It also calls on scholars to tell their stories—publicly or anonymously—to inspire others, maintain accountability and preserve a historical record. Accompanying the handbook is a living wiki that will continue to incorporate new developments and provide updates on global efforts by scholars to push back against authoritarianism and safeguard the democratic foundations that enable free inquiry.

Zenodo

Heute war ein Schulfest. Da gab es einen Quizstand. Eins underer Kinder (9) wollte schummeln, kam zu mir und bat, ob ich "schnell bei ChatGPT" nachfragen könne nach der Frage, ob die Legislative in den USA im Weißen Haus oder im Capitol sitze. Danach folgten weitere solcher Fragen. (Edit: Das Quiz warn nicht für 9jährige, sondern einfach ein Stand auf dem Hof, wo vor allem größere Kinder waren, aber kleinere Kinder turnten da rum und wollten auch mitmachen).

Wir haben später nochmal über diese Situation gesprochen. Ich habe gefragt, warum es nach ChatGPT gefragt hat. Die Erklärung war, dass der Lehrer bei einer Frage vorher mit den Kindern die Lösung mit Google und ChatGPT "gesucht" habe und nur ChatGPT hat sie richtig gewusst.

Kleine Anekdote, aber da steckt viel drin. Ich habe nochmal erklärt, was eine Suche mit Google eigentlich ist und das Google keine Antworten liefert bzw. eigentlich liefern sollte und dass sie eventuell ein Versuch einer KI gesehen haben, die dafür bekannt ist, besonders schlecht zu sein. Aber auch, dass ChatGPT gar keine Wissensdatenbank ist und halluzinieren kann. Das Kind war stark irritiert davon, weil im Sprachalltag von Kinder, Schule, Medien ChatGPT quasi als Superpower imaginiert wird und überhaupt nicht klar ist, dass ChatGPT nicht dafür da ist, um Wissensfragen zu beantworten.

Wir konnten das jetzt klären, aber sonst wird das wohl nicht geklärt. Das macht mich nicht besonders glücklich. Ich habe in den letzten Jahren nun schon mehrfach kurze Medienunterrichtseinheiten an den Schulen der Kinder zu Computer-nahen Themen gemacht. Aber das ist eben nur ein Tropfen auf den heißen Stein. Das läuft wirklich extrem schief gerade. Und hier ist auch nochmal eine Alterslinie erkennbar: Je jünger die Kinder, desto selbstverständlicher sind für sie LLMs in der Lebensumgebung, während die älteren Kinder sie als "neu" auch eher skeptisch betrachten.

"A cyclist can go 3 or 4 times faster than a pedestrian, but uses 5 times less energy. The bicycle is the perfect transducer to match metabolic energy to impedance of locomotion. Equipped with this tool, man outstrips efficiency of not only all machines, but all other animals as well."
—Ivan Illich, 1974.
Grafieken over verkoop en gebruik antibiotica in veesector NL in case toegevoegd aan site.
https://datagraver.com/antibiotica-voor-veesector/
Antibiotica voor veesector

Jaarlijks rapporteert de Autoriteit Diergeneesmiddelen over de verkoop en het gebruik vn antibiotica voor de veesector. In 2009 heeft de Tweede Kamer, na jarenlange stijging van gebruik, bepaald dat voor 2015 een significante verlaging noodzakelijk was. Dat doel is in dat jaar net gehaald. Hier de ontwikkeling eerst in verkochte hoeveelheid en daarna in gebruik...

Datagraver

reminder that I have 20+ years of data experience and I'm looking for some kind of Data Scientist/Analyst/Analytics Engineering gig, remote in the US (I'm in Portland, OR)

I would like to be doing something that makes the world a bit better, and every one has data, so it seems like that should be possible.

https://www.linkedin.com/in/amsantos/

Reboosts and links to job posts are most welcome!

#GetFediHired

fMRI unifies categorical/dimensional visual cortex views. Face/body/scene areas encode multi-dimensions, tuning to sub/cross-category features, with local clusters & sparse distribution, suggesting multidimensional tuning key. #Neuroscience www.biorxiv.org/content/10.1...

biorxiv.org/content/10.110...
×
"A cyclist can go 3 or 4 times faster than a pedestrian, but uses 5 times less energy. The bicycle is the perfect transducer to match metabolic energy to impedance of locomotion. Equipped with this tool, man outstrips efficiency of not only all machines, but all other animals as well."
—Ivan Illich, 1974.
@straphanger Would love to see ebikes on this chart. Must be somewhere between normal bike and motorbike, but I can't begin to guess where exactly.
How much does an electric bike consume per kilometer?

Learn about electric bike energy consumption, optimization tips, and Fiido recommendations to reduce costs, extend range, and choose the best commuting option

Fiido EU
@nix @straphanger there's actually research that finds pedal assist ebikes to be more efficient than acoustic bicycles! A laptop battery and a 500w motor combined with muscle power on two wheels make for a very efficient vehicle
@nix @straphanger
The chart backs up other studies that show humans walking or running are as inefficient as automobiles at turning calories into work. If batteries and motors improve on that then replacing all the humans with robots should offer the huge energy savings the future needs.
@nix @straphanger I have read they’re more energy efficient than pure pedal bikes because battery-motor power is much more efficient than digestion-muscle power.

@straphanger

Why are personal cars more efficient than taxis?

I would think taking people to a spot, picking someone up close by, and so on would save on fuel.

@Madagascar_Sky @straphanger I have the same question. The only reason I can think of why taxis may be less efficient is that they make a lot of stops as they drop off / pick up passengers, which is less fuel efficient than when making long, uninterrupted trips.

@daihard @straphanger

Thanks. Yeah, that seems to make a lot of sense.

@daihard
Hmm I'm not convinced. Their engine is always warm, the drivers have incentive and expertise to drive economically. Also I don't understand why trains score so bad. Transporting hundreds of people in one go using electricity on low friction rail, that's really efficient in my book
@Madagascar_Sky @straphanger
@Madagascar_Sky @straphanger not sure about other countries but in UK the long and relatively efficient motorway/autoroute/freeway journeys tend to be made by private cars, the short city trips at a walking pace tend to be taxicabs

@MatthewNewell @straphanger

Thank you. Yeah, that would do it.

@Madagascar_Sky @straphanger I don't think this chart is taking into account the total amount of distance traveled; i.e., the number of megajoules/passenger-km is the same regardless of whether you drive 5 km or 10 km.

A taxi is likely to travel fewer km total due to fewer round trips, but probably spends more fuel per km on average because taxis spend a lot of time looking for passengers or idling while waiting for somebody.

@minneyar @straphanger

Yeah, that is a lot more logical. Along with the explanation others gave about them doing a lot of short trips whereas personal cars just go to the destination and stop.

Thanks!

@Madagascar_Sky @straphanger taxis need to travel from wherever they are to the starting point of the journey, before the journey even begins. Then, they're one person heavier than an equivalent private vehicle.
@straphanger I've seen that chart many times before and it's pretty out-of-date/inaccurate AFAICT. The Boeing 727 is a 40 to 60+ year-old aircraft that hasn't been in passenger service for years, so not a good comparison metric. And I don't really understand how taxis are way less efficient than passenger cars on their own? And how can electric and diesel rail have the same efficiency rating? Isn't electric rail a lot more efficient than diesel rail in practice?
@straphanger (Obviously the point still stands, but that chart really needs some clarification and updates)
@Quinn9282 @straphanger
Well the taxi is hauling around a driver. Taxis also spend a lot of time travelling about with no passangers. The car is with the owner at home, travels to its destination without the additional weight of a taxi driver, then stays there not moving when passangerless. At times when the owner as ferrying around other people it becomes about as inefficient as a taxi. 99% of the time I drive my car I'm staying at the destination with it.

@Quinn9282 @straphanger For your last question, I suspect by “Electric & Diesel” what they mean is a diesel generator powering electric motors on the wheels. That’s how almost all locomotives for long-haul train routes (think Amtrak) have worked for decades.

I’m not sure how that differs from “heavy rail”, though. The example for heavy rail is a subway, but subways get utility power like light rail, so they don’t have to spend much mass on batteries. Seems unlikely a subway would be less efficient than a train which has to haul its own fuel.

@bob_zim @straphanger Thanks for the clarification. But yeah, I don't know why they split up rail travel efficiency in that way on the chart. Technically the "heavy rail" terminology as a whole can encompass both subways and commuter/intercity rail, so mentioning it in both contexts probably wasn't necessary since the efficiency is going to be essentially the same either way (assuming we're talking about a fully electrified system; not sure how that pairs up with diesel in practice though).
@Quinn9282 @straphanger If the chart is old then there's probably not a lot of renewable energy in the mix. So the efficiency is the same whether you burn the fuel in the train or a power plant.

@straphanger as a mad-keen cyclist I don't wanna argue. But water-borne transport is missing (planes there so not all land). Canal boat shifting tonnes powered by a horse. The old tea-clippers are still close to record holders for lots of searoutes (only beaten by speedboats).

But the point stands. Bikes are cheap, simple, available (hopefully) to all, and eco-friendly. Doesn't get better than that.

@MatthewNewell Well there's another dimension to the Pareto frontier, canal boats are slow. If I want to get somewhere quick an e-bike would be high on the list
@babble_endanger oh definitely. I was just a bit surprised that water-borne transport was omitted. I should probably check out the original article
@straphanger I will say that the Elizabeth line has significantly lower energy usage than the London underground train figure, probably about 0.8MJ/km estimated pessimistically with public data but the overall message still stands
@straphanger ... What are lorries doing in a chart about transporting humans? *Confused*

@straphanger I wonder where an albatross or shearwater would fit on that graph. They travel extraordinary distances expending little energy.

I guess wind power is cheating though. A land yacht would also be more energy efficient than a bicycle, as long as the wind was blowing in the direction you wanted to go.

@straphanger The car value of 2.1 here seems based on combustion-based cars, which is fair because the vast majority of cars still burn fuel. But it would have been nice if they had also added (battery-)electric cars as its own category, since they clock in at a much lower 0.7 MJ/km (approximately: I converted 0.2 kWh/km which is about 3.1 miles/kWh).
@straphanger doesn't consider the need of a infraestructure (plain surface), i guess, but good to know!