Hacker Newsnew | past | comments | ask | show | jobs | submit | jemmyw's commentslogin

I think Germany has tax rules that make exits harder, whereas it's very easy in the UK to sell. If you have a more free market next to protective ones it makes sense that your IP is going to flow in that direction.

People who want to have kids have kids. There have been worse economic conditions, and far worse living conditions for folks in the not very distant past and they still had large families.

There has been a slow burn change to social pressure and autonomy. It seems like women don't want to have a large family, or some a family at all, if the choice is there. The rationale about why they put it off are unlikely to be worth much.

I think every economic remedy will fail. But it'll probably pick up again because I imagine social pressure will turn. All this noise people are making about it right now is the start. Personally I see that as a negative, we should be celebrating a downward population trend. We had so many years of warning about the effects of an ever larger population and now get hand wringing the moment that looks wrong.


My take for a while is that if the low birth rates persist and do become a critical problem, we might just have to do a boring thing that we do for other areas where the unregulated market or society doesn’t produce enough of something or doesn’t solve a problem. We might just have to subsidize it and basically pay people to have kids.

I've tried rtx and lean-ctx and these tools seem to end up confusing the agent more than helping. Any saving is irrelevant if the agent decides to work around the tool and makes even more calls than it would otherwise.

I don't know about cost saving, but if it's keeping the context size down I've had a lot better results using subagents to keep a higher order conversation clean for longer.


I looked into lean-ctx and decided not to use it. It has a very specific use case, and it's good when your interaction with the repository is read-only. When you want to edit, then the model has to read the whole file anyway. It's a cool tool, but it has a very narrow use case where it delivers the performance it claims.

Subagents help with costs too, as they can run on much cheaper models.

Also, most of the time when I'm having an agent look through logs or output, it's grepping for the bits of data relevant to its actions.

I find it quite interesting the people on the thread mentioning face blindness or being bad at names and how it's embarrassing. I'm bad at names, really bad, but I've given up embarrassment about it. For one thing I can remember the names of some people and I've come to realize that being great at learning names is meaningless, I remember the names of people I really like... sorry others, we can't all click.

Secondly, put yourself in the opposite situation, do you really care if someone forgot your name? Does it even reflect how well you know someone? I had a friend at scouts when I was a kid and we were inseparable for a year. Never remembered his name. Didn't matter.


It was a bit more nuanced than that. Each politician was trying to leave in their own image and failing to get anything through parliament, until Boris Johnson got a majority party.

What would have been better is if the politicians who started the whole thing to settle party politics didn't bugger off the moment things didn't go their way and instead did a follow up referendum with the options for leaving. I don't necessarily think the eventual deal would have been different, but it would have stopped a lot of the bickering. And maybe have exposed what leaving meant to a greater number of people.

I don't live in the UK any more, but I do enjoy reading the stories along the lines of "I voted for brexit and now my business has gone to shit" of which there are a surprisingly large number. The whole "we didn't want them telling us what to do but didn't think they'd apply rules to us like we weren't in the EU" crowd.


I ran pre AI articles through pangram last week and it gave similar kind of numbers so I think it's pretty bad at distinguishing.

Someone ran the Declaration of Independence and U.S. Constitution through an AI detector and apparently our country was founded by an AI.

I just checked on Pangram and it turns out that the U.S. Constitution is 100% human-written. Who would have thought?

Saying "an AI detector" is a bit like saying "an AI" - there's a pretty big difference between Llama 3B and Opus 4.8.


I KNEW IT!

To the degree that AIs are good at re-creating the style of the training data, you would expect there to be non-AI-generated text that would "look like" AI output.

> The wallet was supposed to be the constraint. It turns out, as we will see, that the wallet is the constraint after all.

I can't tell if you made a mistake and meant the wallet isn't the constraint. These short burst sentences are really hard to read. Write "As we'll see, the constraint is x.". There's no need to split that, a single sentence conveys the whole point.

The article is full of similar wording, and that's why it feels choppy to read.

> The rest of this essay is about why that is harder than the press understands. And about a second problem hiding underneath it

I'd describe this as chain of thought writing. It's fine in casual conversation, with the words just tumbling out of our mouths, but it doesn't work in writing or speeches. There are so many ways the two concepts expressed there could be worded, combined or separated. "The press has an unfortunate tendency to use hyperbole and simple descriptions, but even with those stripped away there are deeper misconceptions..."

It's interesting that folks have honed in on AI as the problem. I'm my view the issue is that you haven't decided on your writing style, and as a non native speaker, you're unable to write a simple phrase and get AI to embellish it. Writing simple phrases is surprisingly difficult. Try making everything concise, with no repitition, and then adding style and flowery language afterwards.

Edit: sorry I may have read another person's comment about being a non native speaker. Writing concisely is something we can all work on.


I call bullshit on these social interactions having any meaningful impact on work. I've been in very social offices of a large company where we all lunched together, spent a lot of time at the coffee machine, went out together during and after work. Lots of fun. I didn't once see, hear or participate in cross team discoveries as a result that improved work. And in smaller orgs that were also social, the social part is extremely inefficient at moving work information.

My current remote employer does as good a job at building trust between employees with 6 monthly on-sites. But they also do things that expose cross team productivity issue: rotate people in leadership roles between all the different company meetings, so the CEO might be in the planning meeting this week. Get different people in different roles to join customer calls. Not just anecdote at the coffee machine, actually see what's happening across the company.


Sergei Beseda chief of the FSB "fifth service" was arrested after the start of the war. https://meduza.io/en/feature/2022/06/20/no-infighting-here is a story I found around it. I guess we can't be definite, but that journalist in Russia was citing sources saying that the service had essentially embezzled money and made up the data about their resources on the ground in Ukraine, and misrepresented polling data they had that Russia would be welcomed by the general population.

"Soldatov argued that decision makers in Moscow had approached the invasion as a police operation, not a military campaign, because they believed the Ukrainians would actually welcome Russian forces."

" a Moscow FSB agent once tried to convince him not to investigate the October 2002 Moscow theater hostage siege"

Ok, so the source is one russian journalist who was known at least since 2002 to the FSB as someone to keep away from inner sanctum information. It is something, but I doubt he got really access.


Andrei Soldatov is not merely "one russian journalist". Since 2000, he has been operating agentura.ru, which publishes information about Russian security services. He is well-connected and widely considered one of the most serious independent experts on the Russian security apparatus.

But he is not in russia anymore since a while. I doubt his ability to get insights into the highest ranks of security let alone Putin's head himself.

The people who owned the house previous to us put plastic weed mat all over the place. Looked fine until it wasn't, and then it was a huge job to remove it. Still haven't got it all out 10 years on.

I'm not sure about their vinegar claim - I buy a horticultural vinegar and I haven't noticed any bad effects on the soil where I've used it. The main issue is just that you have to spray more often.


Guidelines | FAQ | Lists | API | Security | Legal | Apply to YC | Contact

Search: