tmalsburg2

Which is just another way of saying that it doesn’t really work, except perhaps for party tricks.

Yanis Varoufakis recently called crypto currencies “oligarchic” which is very fitting. Quote:

“I do not believe that the… ‘oligarchic,’ by definition, cryptocurrency like Bitcoin is ever going to replace [conventional currency]. It shouldn’t, it can’t, and it would be a nightmare if it did.”

Full interview in which he also discusses his proposal for cryptocurrencies run by central banks and why that’s (according to him) the way to go: https://www.kitco.com/news/2022-05-27/Gold-and-Bitcoin-won-t...

How can it help you to speed up development but not be worth 10$/month. Your hourly rate can’t be that low.

In lack of any evidence, you can and actually must rely on your priors. There are things that were a-priori unlikely but are still considered true. However, we believe that they are true because solid evidence has been provided to override our priors. In the present case, there is no such evidence.

> "LaMDA is not sentient because come on"

I think this is a perfectly fine argument because the claim is so outrageous. Extraordinary claims require extraordinary evidence and none has been presented. I'd also say that defining sentience is the job of the person claiming it.

This means that you pay ~4 USD per kWH. That's almost 12 times more expensive than in Germany where electricity is supposedly expensive. Where do you live?

Doesn’t support audio output to dlna devices on the network. Seems I will be stuck with pulsesaudio for now.

Question: Why is a tracker necessary? Is there no solution that works with just a qr code that contains some cryptographic signature, similar to train tickets for instance?

Some retweets are useful, true, but I’m more interested in people’s original thoughts. Retweets also used to be 90% of my feed which means that the vast majority of tweets in my feed were by people who I wasn’t following. Original content by people I was following got drowned out. Some accounts also retweet a lot and Twitter didn’t deemphasize these, which allowed some people to basically take over my feed. Having said that, the search function hack that I mentioned does include quote-tweets which tend to be higher quality.

It’s true that some of best accounts are anonymous (that’s why I have rule 7), but also 90%+ of the most toxic accounts are anonymous. Hacker news is a different population, and heavily moderated. Can’t usefully compare IMO.

I settled on a similar approach. I used to liberally follow all sorts of accounts that seemed even remotely relevant, but signal to noise was abysmal and there was way too much drama. Several times I was on the verge of quitting, even though Twitter was quite useful for promoting my own work and that of my students. But recently I decided to try a new approach and systematically unfollowed a lot of accounts. Now I stick to the following diet and it produces a much more healthy and interesting feed:

  1. No politicians  
  2. No journalists
  3. No institutions
  4. No companies
  5. No entertainment
  6. No anonymous accounts
  7. Some notable exceptions in all the categories above
  8. Mute words to taste

What's left is mostly genuinely interesting people who post about things they understand and truly care about.

I primarily browse the chronological timeline without retweets courtesy of a nice search function hack (you must be logged in for this to work):

https://twitter.com/search?q=filter%3Afollows%20exclude%3Are...

The algorithmic timeline has a bad reputation but it's actually useful for stuff you missed. But I only check it after looking at the chronological feed.

I also like and mostly follow these suggestions for using Twitter effectively: https://twitter.com/AlanLevinovitz/status/151946437478365184...

Yes, it's so very obviously useless for the stated purpose that you've got wonder why they want it so badly.

Agree about Ursula von der Leyen, the President of the European Commission. She’s completely serious about online surveillance and has been for many years. Now she has the power to push this really far.

Can't replicate this bug in current Firefox on Ubuntu 21.10. They've probably fixed it.

From the article:

> Data were collected from 229 participants working in a management position in the United States through Amazon Mechanical Turk. Participants were compensated at a rate equivalent to $7 per hour.

I have a hard time imagining people in "management positions" working on Mechanical Turk for less than (US) minimum wage.

If I understand correctly, participants also didn't rate applications with photos, but just photos. Seems possible and perhaps likely that the pragmatics of this specific task would activate biases that may not be at play (or to a lesser degree) in an actual application process where the photo would be just one piece of information. The task really seems designed to maximize effects of beautification while sacrificing generalizability to real-world settings.

How does it work?

I was also worried about filenames, the database, custom ids and all that. It all felt over-engineered for my modest purposes. In true Emacs spirit, the logical conclusion was that I had to implement my own Zettelkasten system:

https://github.com/tmalsburg/mwk.el

Basic word proximity would be enough to deal with the basic stuff like "red cube" but it's struggling even with that.

But see these examples where it utterly fails at even the most basic form of compositionality:

https://mobile.twitter.com/david_madras/status/1512573390896...

Where “energy independence” means that Germany becomes dependent on much more expensive LGN from the US. This may still be better than what Germany has now but it’s not independence.

False. Emacs has a wonderful vi(m) implementation in the form of evil-mode.

Which language Google uses is different from product to product. In Google forms, they use the language set in the Google account (neither location, nor browser language), which is frustrating, because as a designer of a survey you have no way to control the language the UI elements in your survey.

Perhaps you detected them because you knew there would be 50% fakes? If you'd see one of these fakes in a CNN article, would you still reliably detect it? I think the role of the prior expectation should not be underestimated.

I was pretty good at detecting the fakes (video and sound), but made some mistakes with the real ones. But I've also relied on the 50% fakes prior which I couldn't use in a real-world setting. Idea for a follow-up experiment: Manipulate the prior and see how it affects people's performance. If you tell people it's just 10% fakes (proxy for the reputation of the source), performance might be much worse.

I think OP understands this and their comment was a joke.

Except, perhaps, as Putin‘s successor?

You write that without NS and NS2 gas cannot reach Germany. This is incorrect. Only about third of German gas imports are from Russia. 31% are from Norway, 22% from the Netherlands, and the rest from other sources. Gas from Norway and the Netherlands does not enter through either NS or NS2.

That’s interesting but it doesn’t answer the question why it’s okay for the US to do business with Putin but not for the EU.

Even more shocking then that they are buying massive amounts of oil from Russia, thereby filling Putin’s pockets, no?

The US relies heavily on Russian oil [1]. Why the sole focus on European imports from Russia?

[1] https://www.nationalreview.com/corner/why-is-the-u-s-still-i...