wood_spirit

There is one turbine near where I live in Scandinavia that is very noisy. It is a low thumping sound that penetrates houses and is horrid. Those living within a km perhaps more won a court case to remove it but the owner has appealed and appealed and during the years or appeals the thing keeps turning and keeps being noisy so people can’t sleep. My understanding is the simulation and calculations of the noise that were part of the planning process were flawed and did not accurately model the terrain.

Meanwhile, not 5 km away, there are a bunch of turbines with people living around them and no problem.

So the exact slopes etc of the terrain is very important.

Eh that is still encompassed by the term “pattern matching” in this context. Sure it’s complicated, but it’s still just a glorified spell checker.

The nukes have many roles perhaps but I think the fully developed weapons are for retaliatory strike.

They are the North Korean leadership saying that if the US (or China or anyone really) tries to surgically decapitate them (like the US just did in Venezuela) then the nukes are used to take the attackers with them

The nukes are to deter the US. They have been steadily increasing their missile range to first reach regional bases like Guam and now the all the way to the continental USA, and are now even launching a nuclear powered and nuclear armed ballistic missile submarine https://www.hisutton.com/DPRK-SSN-Update.html

It’s oblique but this puts me in mind of an old adage I recently heard about war: Of 100 men, one should be a warrior, nine should be soldiers, and 90 shouldn't be there at all.

I think this is true of software developers too: only in companies, the 90% don’t really know they shouldn’t be there and they build a whole world of systems and projects that is parallel to what the company actually needs.

My memory is there were a spate of SO scraping sites that google would surface above SO and google just would not zap.

It would have been super trivial to fix but google didn’t.

My pet theory was that google were getting doubleclick revenue from the scrapers so had incentives to let them scrape and to promote them in search results.

Another plausible explanation is that the new owners didn’t develop the community in a good way. Instead of fixing the myriad of issues that were obvious to almost all contributors they instead basically let it die?

Adding: UK just come out saying they are not involved at all https://www.bbc.com/news/articles/cy4qgvwxp08o

This feels so foreign: since Suez the UK government has been backing the US and giving them the fig leaf of international legitimacy in their actions.

His loyalty test for subordinates is saying that he beat Biden in 2020…

do you think that a pro US replacement regime in Venezuela will get US backing and support for it’s claims to eastern Guyana?

The UK, for example, stopped intelligence sharing in the Caribbean so as to not be party to war crimes.

Genuinely interested: what problems does mongo fit better than mainstream competitors these days? Why would you use it on a new project?

I really like my Kia EV6.

However, you asked what you can get for the price of a used Tesla…. :)

Tesla sales have plummeted in my part of the world, and they are a bad buy because their second hand price has plummeted too.

They are cheap because primarily the political views or Musk and secondarily they are no longer the only EV maker.

So you can buy a used one cheap… not necessarily a good thing.

You get what you measure.

It does not follow that people making more searches means people are having more successful searches. If google found the exact thing you were looking for and put it top centre in the results, would the number of human searchers stay the same but the number of human searches drop?

Intriguing and very cunning attack! So obvious in hindsight!

It makes me wonder how Deepseek avoids commenting politically on China? I have heard anecdotes that it will be writing out a long reply and then presumably it generates some forbidden phrase and it abandons the output and replaces it all with an error message. So presumably the safeguards could be a separate trivial non-LLM-based post filtering which makes it immune to the doublespeak attack?

The article links to a shodan scan reporting 213K exposed instances https://www.shodan.io/search?query=Product%3A%22MongoDB%22

There are plenty of GPT evangelists too on HN. Plenty of people who have a few successful interactions with a code model and extrapolate that to LLMs reasoning and solving general problems etc.

Presumably the 4000 sacked workers won’t be coming back though?

A long time ago I was trying to get video multiplexing to work over mobile over 3G. We struggled with H264 which had broad enough hardware support but almost no tooling and software support on the phones we were targeting. Even with engineers from the phone manufacturer as liaison we struggled to get access to any kind or SDK etc. We ended up doing JPEG streaming instead, much like the article said. And it worked great but we discovered we were getting a fraction of the framerate reported in Flash players - the call to refresh the screen was async and the act of receiving and deciding the next frame staved the redraw so the phone spent more time receiving lots of frames but not showing them. Super annoying and I don’t think the project survived long enough for us to find a fix.

Do you mean a wallet per transaction?

And if you simply have multiple wallets and try and maintain the appearance of being disconnected, can you move funds between them without establishing a connection that unmasks you?

A “manhattan project” can just mean a massive secret scientific war project? Seems apt.

A decoder predicts the next word (token) to iteratively generate a whole sentence. An encoder masks a word in the middle of a sentence and tries to predict that middle.

The original transformer paper from google was encoder-decoder, but then encoder BERT was hot and then decoder GPT was hot; now encoder-decoder is hot again!

Decoders are good at generative tasks - chatbots etc.

Encoders are good at summaration.

Encoder decoders are better at summaration. It’s steps towards “understanding” (quotes needed).

China is building all kinds of suggestive tech including invasion barge piers to land heavy stuff quickly once a beachhead has been established http://www.hisutton.com/Chinese-Invasion-Barge-OSINT.html

Except the meta reason for employing AI for these use cases is to stop employing the humans?

A better link would be the press release with the map? https://news.nau.edu/gurney-co2/

You get what you measure. The bot might be really bad and customers close the chat and it gets counted as success etc.

Does the recent makeover of the Oval Office count as just one?

Didn’t the majority of people vote to “drain the swamp” and “bring down the cost of living”?

My company recently really cut back on slack retention. At first I was frustrated, but we all quickly got over it and work carried on getting done at the same pace as before and nothing really got impacted like many of us imagined it might.

The Soviets planted listening devices in American embassy typewriters between October 1976 and January 1984 - by intercepting them in the mail!

Really sophisticated devices: https://www.cryptomuseum.com/covert/bugs/selectric/