HN user

freefaler

968 karma
Posts5
Comments380
View on HN

There is also another benefit to books, on average they are much better than a random 3 hour podcast. If you care about what you read, you'd be getting something that the author has spend a lot of time, skill and energy to write, the editor would have spend a lot of time and skill to improve with the author.

I have a measure for all content I consume, quality/hr of reading/listening. If it's just a long video that has 2-3 questions that has caught my attention I'd be listening only those. If it's a long text that I might find something interesting I'll ask the LLM to summarize the main ideas as a filter before I decide to dive in.

Books, and their audiobooks version have on average much more bang per hour than random podcasts, because they're structured, authors had spend more time on them and you can cherry pick from a structure.

I also have caught myself using sloppy content as excuse not working on planned tasks with excuses like "this might be useful", or watching "productivity porn" videos. I think LLMs are good as a pre-filter for that.

I got this problem with my own employees, LLM are fine, but lazy slop is not permitted. Current idea is to have a clear "best practice" template for most of the research/specs/problem definition they submit and it reduced the slop to a manageable level. But this might work in a smaller company where the management is reading and is strict about these things.

I have one IKEA Apstuga on my desk, sitting right next to a good CO2 monitor. Since Apstuga uses worse approach (heat) rather than light as the good sensors, it diviates around +/- 100 ppm. For example the correct CO2 is 610 ppm and IKEA's sensor shows 552 ppm with is reasonably close. So the trend will be correct and the values will not be.

But when it goes over the safe limit it should be enough to decide to ventilate.

It's strange how people search for cartels where the basic laws of supply and demand are at work. It seems to me, that such a simple concept should be better understood, especially in any market economy, where most of the people has sold something online and priced it somehow. But alas, that's not what's happening, is this because of the discourse of "if I don't like something it's because of some evil corporation tries to squeeze me"? (many examples of those, but not all markets are Apple Iphones)

Cory Doctorow had a very profound talk about it very long time ago (10+years).

https://www.youtube.com/watch?v=HUEvRyemKSg

As the internet become the place where people do a lot of things, no government (and especially no security services) will be able to keep themselves from trying to control it or at least monitor it. And with the new LLM features they can automatically do much more than before.

Human nature is a constant and when the government sees an easy way to enforce something, many more bureaucrats will try to do it.

It works with relatively good scans, when there are bad/skewed scans and especially something with many label/value pairs, that aren't nicely tucked inside sentences, the more context you have, the more you can find the correct words and fix the errors.

There is a whole class of tricky documents. A decent (if you ignore the marketing bias) post about this problem can be found here:

https://getomni.ai/blog/ocr-benchmark

I did, and also it matches what I know from history on how the flag of AU has changed so it matched several facts I knew already. Also mentioned 2 of the producers of freezes and etc in the city back during that time and I've found a some images from their catalog online.

I live in a house that is at least 110 years old, built around the time Australia became a federation. The house has been repainted and extended several times, but it still has 3.5-meter-high ceilings, ventilation holes with intricate metal bars, and a lot of original details.

Even though the house is really old, it has been taken care of. There is no mold, the doors are still the same original ones, the fireplaces with decorative tiles are still there, and the wooden fireplace parts are still in good condition. I don't know how they did it, but it was built rather well.

Interesting tidbit: on the ceiling there was something like a Star of David. After asking LLMs what it was, one of them said that when Australia became a country, the Federation Star had only six points, denoting the newly incorporated states, but later a seventh point was added. Gemini told me that the frieze details were typical for the 1901-1910 period, and this helped me date the house.

This took about 10 minutes. Before, it would probably have taken me at least several hours of Googling.

Also they're mostly a VPS provider with some "dedicated servers" available after you talk to sales.

Hetzner is great because you can build from dedicated servers your own infra and not suffer the multi tenant issues on virtualized servers. (IOPS exhaustion for example)

There is a good paper about the effects:

https://www.sciencedirect.com/science/article/pii/S105113772...

Basically if you're in a "rent controlled" unit, you benefit, but live in a dumpster, because the owner don't have the reason to invest in the place.

But the supply of new units is down & all other renters are paying higher rates, because there is no incentive to build new units. NIMBY-ism has the same effect.

All businesses run on a profit margin. It's a resource game, thinly veiled by the money layer as means of exchange. If you make the price of buying labor by decree, the owners of the business will either buy less of it (supply/demand ratio), automate it, raise the price to offset the increased cost or close the business down, because the costs are too high.

Businesses are shaky constructions and when you regulate them out of existence who will be generating the taxes for the government to redistribute?

About the "single payer health system", it's not a panacea. Many countries struggle with single payer systems, because there are also problems there too. The best working example for a medical system is Singapore. Compulsory medical saving account with government subsidies in that account for the poor. All prices are transparent and patients pay from that account and have premiums over the basic treatments, thus creating the market dynamics needed for the market to work and still requiring personal responsibility, by making sure the patients sees the costs and chooses himself.

The wizardry is not in knowing the law or the case law, especially in commercial law it's mostly selling their experience because the have seen it all before and if they're good they'll show you the ways you can get abused by the other party. History of negotiations and deals is the leverage they provide.

The price of their work will go down and it might not be economical to do it at all. Theirs skills (as also many IT skills) will not be needed at that scale. In the same way as typing on a typewriter was a skill that gave economic opportunity not so long time ago. Now everything is an email and part of it is speech to text. When something becomes a commodity, the skilled providers need to find something new to sell on the market.

About the law firms, part of the job of a law firm is to give the corporate employee a "guarantee" that he won't be held accountable for doing something legally stupid. So a new lawyer is at great disadvantage if he don't have the contact he has build trust with. From a freind's law company with 50+ lawyers I know that junior lawyers fresh from uni need at least 4-5 years to build their client base. Then, they can leave and start their own taking part of that client base with them. This limits the number of people who can start their own company and most of them won't risk it in the age of AI, because it will be sales and marketing that will feed them, not their legal wizardry, especially when tasks like "check this agreement" won't be billed at the current insane rates.

Ferrari Luce 2 months ago

The price is the reason. Veblen explains that.

Buying an ultra-premium EV Ferrari over a faster, cheaper is a evolutionary broadcast (Costly Signaling Theory), proving the buyer possesses such immense excess wealth that they have no practical need to optimize their dollar-to-spec ratio. Everybody drives Teslas, the highly exclusive Ferrari satisfies a deep human drive for elite group differentiation (Social Identity Theory) while perfectly mirroring the buyer's aspirational ego and public identity (Self-Congruity Theory). Ultimately, this choice optimizes for intense internal sensory and emotional pleasure rather than objective efficiency (Hedonic Consumption Theory) by making (at least at the beginning) the owner feel that he is a super special dude.

Good design follows the function. Not distinctiveness per se.

If it's an interface and not an art object, then the design is secondary to the function of said interface.

Good hammer is a good hammer, not a "distinctive" hammer.

The legacy compatibility of the protocol has brought all the hacks on top of it for identity verification like SPF, DMARC, DKIM ...

Even with those, the amount of farmed accounts from a reputable platforms is still high, and it will go higher with the cheap AI targeting that will make the texts much more well crafted and spam filters much more aggressive.

My other conjecture is that the big mail providers would have enough data to catch the spammers based on a number of signals.

A classic "the tragedy of the commons" with the SMTP protocol.

When the cost of spamming is near 0.00, all open platforms will be abused to the tilt. We have seen the email channel get less and less reliable with our own clients (password recovery, notifications and etc).

This might evolve into a couple of oligopolies (Microsoft 365 Outlook, Google Gmail, may be some legacy email providers like Yahoo) and if you want delivery you'd need to pay them, because they'd be the verifiers that you're not a spammer.

And these platforms will have a hell of time to fight the spammers that will create millions of email addresses and spam trough them.

They have a working operational system and battle tested tactics, not only procurement. It's not the rifle that distinguishes the special forces, but how it's used.

They built a network centric warefare with starlink and cheap android tablets down to the drone teams in the field.

They built a network of cheap acousting sensors (old phones) as passive sensors and using ML models to find the drones cheaply and increase the coverage. (Radars are expensive and easy to hit because they emit).

What they achieved is a "sensor fusion like" distributed system buid on cheap components and updated realtime. And all this is battle tested in the new environment of transparent battlefield (there is always a drone looking).

Also a lot of real-life electronic warfare stuff and drone applications.

This is what's missing in the US army. They are optimized for a symetrical 20th century warfare.