Looking for a #postdoc in modelling & simulation for neurodev at Ecole polytechnique near Paris! pls share...
Funded by a national ANR grant, collab with Jean Livet@IDV. Aim: try and explain unique conectomic data by simulating development.
More info:
https://mycore.core-cloud.net/index.php/s/jdsWRnGEXSGmA9b
Official job posting and application:
https://emploi.cnrs.fr/Offres/CDD/UMR7645-ANACHE-002/Default.aspx?lang=EN
"as [bioRxiv and medRxiv ] become essential infrastructure, it makes sense to be supported by a diversity of stakeholders..several institutions offered financial support, including Stanford, MIT, Caltech, CNRS, Fred Hutch, Imperial, UW, & VU Amsterdam"
š šš to all who have helped. The vision of bioRxiv/medRxiv has always been a non-profit, community effort and that includes funding. So if your institution would like to help too, please do get in touch!
https://investinopen.org/blog/infrastructure-spotlight-biorxiv-and-medrxiv/
Like many other technologists, I gave my time and expertise for free to #StackOverflow because the content was licensed CC-BY-SA - meaning that it was a public good. It brought me joy to help people figure out why their #ASR code wasn't working, or assist with a #CUDA bug.
Now that a deal has been struck with #OpenAI to scrape all the questions and answers in Stack Overflow, to train #GenerativeAI models, like #LLMs, without attribution to authors (as required under the CC-BY-SA license under which Stack Overflow content is licensed), to be sold back to us (the SA clause requires derivative works to be shared under the same license), I have issued a Data Deletion request to Stack Overflow to disassociate my username from my Stack Overflow username, and am closing my account, just like I did with Reddit, Inc.
https://policies.stackoverflow.co/data-request/
The data I helped create is going to be bundled in an #LLM and sold back to me.
In a single move, Stack Overflow has alienated its community - which is also its main source of competitive advantage, in exchange for token lucre.
Stack Exchange, Stack Overflow's former instantiation, used to fulfill a psychological contract - help others out when you can, for the expectation that others may in turn assist you in the future. Now it's not an exchange, it's #enshittification.
Programmers now join artists and copywriters, whose works have been snaffled up to create #GenAI solutions.
The silver lining I see is that once OpenAI creates LLMs that generate code - like Microsoft has done with Copilot on GitHub - where will they go to get help with the bugs that the generative AI models introduce, particularly, given the recent GitClear report, of the "downward pressure on code quality" caused by these tools?
While this is just one more example of #enshittification, it's also a salient lesson for #DevRel folks - if your community is your source of advantage, don't upset them.
Thought about hypothesis testing as an approach to doing science. Not sure if new, would be interested if it's already been discussed. Basically, hypothesis testing is inefficient because you can only get 1 bit of information per experiment at most.
In practice, much less on average. If the hypothesis is not rejected you get close to 0 bits, and if it is rejected it's not even 1 bit because there's a chance the experiment is wrong.
One way to think about this is error signals. In machine learning we do much better if we can have a gradient than just a correct/false signal. How do you design science to maximise the information content of the error signal?
In modelling I think you can partly do that by conducting detailed parameters sweeps and model comparisons. More generally, I think you want to maximise the gain in "understanding" the model behaviour, in some sense.
This is very different to using a model to fit existing data (0 bits per study) or make a prediction (at most 1 bit per model+experiment). I think it might be more compatible with thinking of modelling as conceptual play.
I feel like both experimentalists and modellers do this when given the freedom to do so, but when they impose a particular philosophy of hypothesis testing on each other (grant and publication review), this gets lost.
Incidentally this is also exactly the problem with our traditional publication system that only gives you 1 bit of information about a paper (that it was accepted), rather than giving a richer, open system of peer feedback.
offf, this story about how Google made google search into a pile of seagull shit hits me hard:
https://www.wheresyoured.at/the-men-who-killed-google/
Around the time of this story, I was living through a similar situation in my work life (on a much smaller scope, of course, WordPress.com first, Tumblr later).
Back in 2019, working on WordPress, I started finding myself, almost weekly, arguing against people who wanted to take the product we were working at and made it worse if that mean they could squeeze 0.1% more revenue from it
The 0.1% figure is not even a random number: I remember this speciffic A/B test on WordPress.com that was declared a success and shipped to 100% of the users because it increased the free-to-paid conversion by 0.1%. Soon after it was released, I found out that as a side effect, it increased the churn of free users by 20 something %,so I called for an urgent rollback and removal of the change. So I was promptly explained that we didn't care about free-users churn, because finance had calculated the average long-term value of the free users to be something like $2 per year, and the increase in conversion was bigger than what we could get from them.
Everything became about growth hacking. Everything became thinly-veiled dark patterns. In our private dev slack channels, we joked that since it was impossible to make it smaller or less conspicuous, the next thing the growth team was going to ask us to do was to make the 'free plan' button flee away from the mouse pointer when the user tried to click it. We kept making our product worse, we kept consciously crippling the cheaper versions so we could force people to move to the more expensive options.
Back then I was the lead of one of the two dev divisions working on WordPress.com, so my job was mainly to discuss what we were going to be doing, when and how. And I was getting drained by a constant state of fight against a constant wave of shit they wanted us to build. So much than by the end of 2020, the CEO quietly told me to follow the growth team plans and shut up or step down.
So I requested to move to tumblr, because I thought the pastures were greener over there. But it was all the same: Adding login walls to what we were pretending to be "the last bastion of the free internet", cramping in embarrasingly obvious money-making schemes disguised as features, and making them silently opt-out instead of opt-in so the less people the possible would deactivate them, having to fend off the pressure from the CEO to make everything algorithmic timelines because, you know, tiktok makes a lot of money and why aren't we, etc etc.
I found myself in a place where building something good that people enjoy using was no longer a priority, but tricking people into generating more money for the company was. And when I looked around me, I could see that happening everywhere else, not only in my company. Experiencing the start of the enshittification years from inside wasn't easy.
And, as in the article, the people who decided to turn the shit-metter up to 200%, have a name, in every case. And these people, no matter if they are called Sundar and Prabhakar or Matt and Mark, are destroying the internet. These people are milllionaires, or billionaries, and are destroying our shared, common spaces to squeeze some extra cash from us.
That's why the fediverse and its principles are important. Because that's how we take back internet from their dirty hands. That's how we make internet resilient against them. That's how we build the commons.
I accidentally found a security issue while benchmarking postgres changes.
If you run debian testing, unstable or some other more "bleeding edge" distribution, I strongly recommend upgrading ASAP.
My discipline, anthropology, is not seen as a āgrowth" discipline, and departments are being closed down. But the world needs Anthropology and Anthropologists now more than ever!
Here are my 8 reasons for this:
1. POSSIBILITIES
At a time of polycrisis, when the destructive fallouts of capitalist modernity are ever more apparent, anthropology highlights that there are myriad alternative ways of thinking and living; that there is so much to learn from other peoples in the world. 1/n
Processing neuroanatomy data in napari? You may find brainrender-napari (https://github.com/brainglobe/brainrender-napari) useful.
We're slowly porting the original brainrender to napari to provide a consistent neuroanatomy data analysis and visualisation environment.
If you think this could be useful for you, please give it a go and reach out if you have any questions/suggestions.
- Tutorial: https://brainglobe.info/tutorials/visualise-atlas-napari.html
You need to stop using Chrome NOW. Itās not hyperbole: Google just rolled out a change to Chrome that tracks the sites you visit, builds a profile, and shares that with any page you visit that asks.
This is real. Itās not tech bro conspiracy shit.
Vivid description of emergence!
Mathematical biologist Jack Cowan loves to describe the difference between biophysicists and theoretical biologists ā¦. ātake an organism and homogenize it in a Waring blender. The biophysicist is interested in those properties that are invariant under that transformation.ā One couldnāt get a more graphic image of the difference between aggregativity and emergence.
From Wimsatt:
https://www.hup.harvard.edu/books/9780674015456
A common question is: āhow do we improve thingsā. One way is to de-emphasise and de-incentivize academic publications. The current incentives drive us to bad scientific practices (e.g. salami slicing, high amounts of low-quality papers, paper mills, data manipulation, metrics gamingā¦). Once the career and financial incentives are gone from publishing, we can get back to publishing only what matters, when it matters, as a small part of the scientific process. Or so I dream. š
Tomorrow are again the APAC/AU/EU- friendly NGFF & Friends Office Hours.
If you have questions, feel free to stop by!
Scientific #data is becoming increasingly complex, making data #curation increasingly difficult.
In this #PLOSBiology Perspective, a group of biocurators present some simple guidelines for data #accessibility that will help to increase the reach of published studies.
The villagers watched a mighty paladin battle the demon under the full moon. At long last, the demon was vanquished.
"Did you see an old lady?" the villagers called. "It took her!"
The paladin looked at the moon setting, then shuddered and seemed to shrink. The armour turned to mist, the sword became a cane.
"I'm here."
"You're a... werepaladin? Is that a- How?"
"In my youth, I was bitten by a paladin." The old woman smiled fondly. "Many times."
I love this inversion of Gibson's famous line from Neuromancer, found in Tim Maughan's "Infinite Detail":
āThe dead screens are the color of the night sky.ā
A great line and so much added depth knowing the original version too.
My neighbor is a good friend and engineer and one of those "linux is fun to play with" people.
I am a librarian and open source fan and birdwatcher. Together we got BirdNet up and running from my back porch, linked it up on BirdWeather and now you can see what tiny dinosaurs are flocking to my backyard feeders.
https://app.birdweather.com/stations/2588
Build your own: https://github.com/mcguirepr89/BirdNET-Pi
A gamedev realized Linux users were just 5.8% of their sales, but represented 38% of bug reports.
Then they looked at those numbers closer, and realized. Linux users were not experiencing more bugs. Almost none of the Linux-user bugs were Linux-related. Linux users were simply *more likely to file bugs*.
Their conclusion: A linux port pays for itself bc it nerdsnipes ppl into giving u free QA, see:
https://techhub.social/@ozone89/111337250473454154
Via @Rainer_Rehak
Naturalizing Relevance Realization: Why Agency and Cognition are Fundamentally not Computational, with Anna Riedl, Alex Djedovic, John Vervaeke & Denis Walsh.
https://osf.io/preprints/osf/pr42k
It's massive indeed! I'll do a TL:DR thread on its argument one of these coming days.
Announcing the beta of Zotero for Android: https://forums.zotero.org/discussion/110371/available-for-beta-testing-zotero-for-android
A Christmas Message from the President of Ireland
"...may I express my gratitude to the migrants who now call Ireland their home. Their presence enriches our culture, contributes to our society, bringing as they do experiences, traditions, and perspectives that make us stronger as a nation.'
http://telescoper.blog/2023/12/24/a-christmas-message-from-the-president-of-ireland/
