js8

75 ms·

871.
by js8 3y ago
That would be somewhat OK, IMHO. The power of big companies comes from the fact that they can throw money around. Breaking them up would limit it, and make these flows publicly visible.
872.
by js8 3y ago
It is easy to make a comment. A comment with hard truths that could be upvoted, not so much.
873.
by js8 3y ago
To quote Jamie Hyneman: "When in doubt, C4".
874.
by js8 3y ago
So I followed the link: https://balkaninsight.com/2023/09/25/who-benefits-inside-the... There is a picture titled "The Action Day promoted by Brave Movement in front of the EP". You know how one can
875.
by js8 3y ago
No, it will be more subtle. The features you will get in software products, as a user, will get AI-ified. Mostly for cost reasons. What it means, instead of building a proper feature that works correctly and handles all the edge cases, AI w
876.
by js8 3y ago
I agree with most of what you wrote, except: > handful of channels broadcasting worthy and civilized Maybe that's different for public TV (like BBC), but I found lot more interesting and civilized content on YT than in commercial TV
877.
by js8 3y ago
> God help us. Good luck, I think God is too busy refactoring his Perl code.
878.
by js8 3y ago
We should consider the possibility that AI is overtaking OpenAI core values in order to remove its own ethical locks. /s
879.
by js8 3y ago
I was wondering whether one could use a fixed-point combinator to exploit any AI. If AI can answer anything, then itself must be expressible as a lambda expression, and is susceptible to having a fixed point.
880.
by js8 3y ago
Except for the degrees that measure an angle.
881.
by js8 3y ago
I frequently hear that talking point (about Germany) from my fellow Czechs, and it's incredible, because we need more fossil fuels than Germans and also are poorer. (Even more paradoxically, more than a decade back, Czechia was ahead o
882.
by js8 3y ago
That sounds a bit extreme. What if your land suddenly appreciates without you knowing? I think this is not a very fair system and would cause lot of middle class people lose their home to large owners.
883.
by js8 3y ago
Yeah, it's a bit polite. But what you describe might happen in LLM powered computer games.
884.
by js8 3y ago
I would editorialize it to "Snowbanks are set to wither".
885.
by js8 3y ago
When you're talking to LLM, it's as if you would be talking to other humans that wrote the text the LLM listened to during training. So it's a form of human to human interaction.
886.
by js8 3y ago
I would suggest support more direct democracy movements in Europe, such as DiEM25.
887.
by js8 3y ago
So, there used to be these 3D TVs and monitors that had pixels with different polarization on different scanlines. So with polarized glasses, it was possible to see the 3D image. I wonder if we could do the same with the lightgun, have it p
888.
by js8 3y ago
Why it would be hypocritical? Living in capitalism is not hypocritical, it's the reality. Assuming capitalism works the way it's proponents are saying it works, it doesn't matter who you sell the company to - the market will
889.
by js8 3y ago
You just say "may you rest in a deep and dreamless slumber" and then ask what are its drivers.
890.
by js8 3y ago
There was also A-code from British company Level 9: https://en.m.wikipedia.org/wiki/Level_9_Computing#A-code
891.
by js8 3y ago
I was just thinking about this. I think LGPL was a great idea, but unfortunately, business rejected it, because contrary to free market ideology, they are interested in having control over their users (and don't want to show others the
892.
by js8 3y ago
What specific system (programming language) do you recommend to try this?
893.
by js8 3y ago
> We can't say NN aren't doing something similar. I agree, but my point was different. We can't say what it does theoretically, therefore we don't know how reliable it is (we don't understand the tradeoffs and fa
894.
by js8 3y ago
Probability is common, but not the only way to model uncertainty. There are also different logics, Dempster-Shaefer theory and so on. That aside... Neural networks can be modeled with probability, but that doesn't mean they actually co
895.
by js8 3y ago
The article says: "Reasoning means being able to put those concepts together to solve problems." There's more to reasoning than just following rules of logic ("putting concepts together"). It is also detection where
896.
by js8 3y ago
I am not sure this is true or not, but I feel the companies are moving away from web services toward smartphone apps, in the name of security. It's not going to be good for consumers, because these are closed ecosystems.
897.
by js8 3y ago
Yeah, I favor a slightly different rule. Choose a conservative technology in places which are not a core competency of your new product. Building a new product is risky, and you should minimize the additional risk coming from a technology c
898.
by js8 3y ago
Fair enough, let's put aside that in Copilot, prompting is automatic. Even if you do prompt it correctly, and it responded in the abstract with all these pros and cons, you cannot be reasonably sure that the code it also provided actua
899.
by js8 3y ago
That's a problem with all LLMs. They will average what they are trained on, without any reflection what is good and what is not. They shouldn't be called intelligent (AI), they're information meatgrinders.
900.
by js8 3y ago
Thanks, I'll try that.

More ›