Mikael Huss

I am starting a computational lab at LIOS, Riga, Latvia, and going to recruit a few people. Please DM if you are interested in a postdoc or PhD position in my Metabolic Research Group within the TARGETWISE project. I can highly recommend Riga, it is a beautiful city with very good living conditions.

… naive, as I think a lot of people are not really feeling the love from our government so to speak

That’s true! Many things are good here - otherwise I’d move - but I thought the original post sounded overly

Wealth inequality stats from wid.world/world/#shwea...

Well, I live here…

For many, the poor healthcare access and delivery (not talking about quality, but the difficulty of getting care in the first place). For some, the rampant bombings and other gang-related violence. Large and growing economic inequality and segregation.

I think many people in Sweden would disagree that the leadership protects us from childhood to old age.

Probably Unknowing-Interested though I aspire at Aroused-Opinionated status

Lao Ganma 😍

I have that feeling too!

Works for me too (also in Sweden so does not add much information)

He could literally shoot a child to death, and his followers would find a way to treat it as "fake news", "setup by the Democrats", or something similar. He can do no wrong in their eyes.

As I wrote in the original post, because looking weak is something he has seemed very sensitive to in the past. It is a psychological question. But thank you for your input.

My question was psychological, why Trump does not care about looking weak. Not sure what you are getting at with this question.

Of course, that might be a bit exaggerated, but at least most people in EU, Russia, China would view it in that way, no? I suppose he only cares about US voters, but even so...

Serious Q. If there is anything Trump (governor Krasnov of the Amerika Oblast) seems sensitive to, it's status and power dynamics. Why doesn't he seem more concerned that most of the world now sees him as Putin's lapdog? Surely Putin looks the "alpha" here in the repulsive framing these guys have

That’s a nifty idea!

Therapists these days, am I right?

Yeah 100%. I am just writing from some sort of Realpolitik (Putin probably fears Xi and knows China is vastly more powerful than Russia, while Europe is still fragmented).

Sure, I agree in principle. But what if the US and Russia become true allies? We will need some sort of strategic relationship with China then. Will be interesting to see whether Macron starts reaching out to Xi.

Yup, he is perhaps driving Europe into the arms of China. I wonder what his idea about the endgame there is.

Taiwan, Beyond the People’s Republic of China and the Republic of China This text was written by two cousins who identify as Taiwanese with mixed Han settler ancestry. Szu-Han Ho lives in Tiwa lands (Albuquerque, New Mexico) in the United States. Meng-Yao Chuang lives in…

I think it's a bit more complicated than that. See e g thefunambulist.net/magazine/dec...

2023 letter | Dan Wang A walk-and-talk; Crumb; running and tangping; Knausgaard; Elon as Palmer Eldritch; memelords; Chicago as a megacity; Norton; and a pause.

Perhaps that's when Elon started his transformation into Palmer Eldritch ;) (pretty far down in this very interesting blog post danwang.co/2023-letter/)

"envenoming" is a fantastic word

Håller med! Har på senare tid gått och funderat på om man kan säga att 1900-talet definitivt tog slut 2024

Me after working together with an LLM for coding too long:

Yesterday I solved a React front-end problem with Gemini 2.0 Flash (free tier) by pasting a series of screenshots and code for the component. It solved it after a couple of tries; free ChatGPT and Claude did not even try (due to attachment and length restrictions). I don't know anything about React

I’ve been following the DL for tabular data space for years and always been disappointed when trying the DL models in the wild. The approach described here makes a lot of sense and might move the needle. Eager to try!

Flyers advertising witch doctor services

The harder part (rather than training models on all that sequence data) might be to connect all the relevant experimental data and metadata so that the resulting model does something useful. But in principle, I'm all for training on all of SRA for example