Just read @pluralistic's latest and now I'm angry all over again about how I can no longer call myself a Realist or a Rationalist because assholes have appropriated those terms to justify their bigotry and greed.
FWIW? A true Realist believes in full equality and rights for all without favor – because, realistically, any alternative system might get applied to YOU. It's irrational to think you are special.
The world Philipp Mainländer describes in “The Philosophy of Redemption” is the world I know.
I see myself as a adherent of Mainländer.
Mainländer described the world that I experienced for most of my life. There are several ideas and images of his which were part of my inner world. Before, I had no way of seeing them clearly.
Branded a traitor during the Reign of Terror, French chemist Antoine Lavoisier, who was also a tax collector with the Ferme générale, is tried, convicted and guillotined in one day in Paris.
The day after Lavoisier's execution, the great mathematician Louis de Lagrange commented: "It only took them a moment to knock that head off, and perhaps a hundred years won't be enough to reproduce a similar one".
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.
Just found out that the entirety of the 1997 election coverage is on the BBC iPlayer, and given that we are very likely to see something very similar this year, it seems like a good time to reboost my piece about how culturally, we've yet to escape 1997...
Radboud University Nijmegen (Netherlands) seeks to fill two PhD positions (4 years) in for the ERC project “Avicenna Live: The Immediate Context of Avicenna’s Intellectual Formation.”
The currency of life isn't money, it's options. A monk can choose to leave the meager life, a prisoner can't. Money also gives options, but so do contacts, skills, knowledge, ... #life#currency#philosophy#showerthoughts
It sounds obvious in hindsight but some #philosophy researchers did a study analyzing the hundreds of thousands of posts and millions of comments in the popular subreddit r/AmItheAsshole.
On May 7 at 4:30 PM EDT, Jon Leydens will deliver the Bovay Lecture in the History & Ethics of Engineering at Cornell University. His research concerns how engineering education can contribute to social justice, sociotechnical thinking, and humanitarian engineering.