This is a cool solution... I have a simpler one, though likely inferior for many purposes..
Run <ai tool of your choice> under its own user account via ssh. Bind mount project directories into its home directory when you want it to be able to read them. Mount command looks like
I've been using a dedicated user account for 6 months now, and it does everything. What makes it great is the only axis of configuration is managing "what's hoisted into its accessible directories".
Its awe-inspiring the levels of complexity people will re-invent/bolt-on to achieve comparable (if not worse) results.
You can, and should, over the entirety of europe apart from the northern parts of the nordic countries electric heat pumps are now simply more efficient than gas powered furnaces. This is true even if powered by gas based electricity - but obviously makes it possible to power them via renewables as well.
People in Quebec (Canada), which is colder than just about all of Europe, have been providing heating in winter using renewables for decades (thanks to an excess of renewables).
There are a gazillion battery techs being developed right now (regular lithium ion - with variations like NMC, LFP, ...), solid state lithion ion, sodium ion.
You can over provision solar as someone said.
There's geothermal, tidal, etc.
Long distance high voltage electricity transmission at scale.
Electricity is a marvel and we're just starting to scratch the surface of what we can do with it. Betting against it is like betting against electronics, a risky proposition.
So... why is fuel 25% cheaper in Slovenia than in the neighbouring country while Solvenia is simultaneously having issues with running out of fuel?
Seems like the obvious solution is to raise prices so people stop driving to your country (wasting fuel, ironically) to take your cheap fuel instead of just paying for the fuel in their own country. More than that it's a solution the free market would actually find on its own...
It's not a free market. Off-highway prices are regulated and were adjusted by the executive govt branch on biweekly basis, now switched to weekly. Slovenia is small and "gas tourism" is common since fossil juices in neighboring countries are priced higher.
Why not raise the prices? Sure, but then don't complain about the inflation, revolt, and stoning of elected representatives.
We can barely afford it at the current price. The solution would be charging foreign transit the non-regulated price but that would be considered discriminatory.
Lifting the gas price regulations will somehow make EVs more affordable to the average Slovenian?
Electricity is expensive here too.
Public transit wise, good luck. The bus system has only been getting worse (despite sustained usage), trains are not much better. There just aren’t any viable routes in many places — it would take me 6h to commute 80km to Ljubljana (3 transfers with waiting time in between), it takes 1h30 by car in peak traffic.
Both busses and trains are also much more expensive than just
driving yourself unless you’re retired or in school and thus have a subsidized ticket. And this is with regulated gas prices.
Price increases tend to be regressive—the poor person who needs a little fuel to get to their job is hurt more than the large business that uses a lot more fuel but has much, much more money overall.
There are things you can do to try and even things out. Etherium has been considering “quadratic voting” to solve a similar problem (in this case, that would look like tracking consumption and increasing the unit price of fuel as you consume more fuel, so that cost goes up quadratically with consumption). That seems hard to enforce, though, and doesn’t help with foreign opportunists.
I'm totally ignorant as to Slovenia, but as a general comment on taxation regressive price increases/externality taxes/sin taxes are easily made up for by simply giving everyone a fixed sum of money (that can either be gathered specifically through the regressive tax or just through the normal non-regressive tax pool).
Ethereum has the weird issue where "votes" and "money" are different things and they only want to redistribute votes and not money, but that's not a problem here...
This BBC article does a really poor job of explaining the context of this situation or why fuel would be so much cheaper in Slovenia, so I had to look around. Slovenia apparently introduced fuel price regulations last year (for motorway service stations; off-motorway stations have been regulated for longer), as a means of reducing costs for consumers[0]. These price caps were, in fact, removed a week ago[1], and prices at some stations rose considerably in the aftermath, closer to the Austrian prices across the border.[2] I won't speak to the wisdom of the Slovenian government in trying to cap fuel prices, but however well-intentioned the policy was, it didn't last long in the face of a global energy crunch.
[0] https://sloveniatimes.com/43824/fuel-price-regulation-expand...
[1] https://www.brusselstimes.com/2037901/slovenia-imposes-fuel-...
[2] https://sloveniatimes.com/47009/prices-at-the-pump-up-substa...
One thing you have to keep in mind is that in Slovenia, your employer is required to cover your commuting expenses. If there’s no viable public transit option (which is the case for most of Slovenia outside of bigger cities), they have to pay you for gas per km.
So if the regulations were to suddenly be lifted, this would have a domino effect on not only truckers but also regular commuters, which would then mean companies would have to compensate for the increased labour costs by raising the prices of their products/services even more.
Which is adjusted to compensate for inflation of fuel prices every few years, so they would eventually have to raise that to cover the increased prices.
In Slovenia, fuel prices have been regulated since, like, forever.
A few years ago (or last year? not sure) they were deregulated on the highways (i.e. to make tourists pay more) but then the government changed their mind (several times, IIRC).
They were deregulated on highway for a very long time. Deregulation came to off-highway in 2020 as the loss of demand due to covid made the prices drop. Rusian invasion of Ukraine and subsequent price hikes made the govt regulate the prices again.
Somewhere in between, a feud started between the largest provider Petrol and govt, and govt started regulating the highway prices too for no good reason.
Fuel prices are regulated here, and we had an election right now and a huge gas price hike would be bad for the current government (not decided yet if they stay or go). The government basically lowered the gas tax for a bit to keep prices stable (they also raised the gas taxes during covid to keep the prices "stable").
The prices will go up soon, that's why everyone is panicking and filling up canisters of gas.
There's also cis-male people who will "pass" that SRY test if they take it for some reason...
This is a dumb ass way to try and define the woman's category... which is about the expected result of bigots trying to work backwards from the result they want headlines about.
SRY testing was done at the 1996 games and for a while before that. 8 cisgender women tested positive at that games. Far more than the number of transgender athletes who have ever participated. This resulted in genetic testing being changed from all women to on-suspicion.
The bottom line is these tests will catch dozens of people who are phenotypically women, who can even give birth. Why should men be allowed to compete as genetic freaks but not women?
From what I've read, these women all had CAIS or similar, and testosterone had no effect on their bodies. Thankfully the new IOC guidelines have an exception for that and would let them compete with women.
But I want to point out that XY+CAIS individuals cannot conceive or carry a child. They have no ovaries and no uterus.
> Why should men be allowed to compete as genetic freaks but not women?
They are, if they are female or have CAIS. Caster Semenya, for example, does not meet that standard. Caster was assigned female at birth and raised as a girl, but is not biologically female, rather a male with a DSD (5-ARD) who has testes and fully male levels of testosterone and musculature.
We are not. The reason we wanted to announce early was so that folks had plenty of time to opt-out now. We've also added the opt-out setting even if you don't use Copilot so that you can opt-out now before you forget and then if you decide to use Copilot in the future it will remember your preference.
Would you be able to comment on https://news.ycombinator.com/item?id=47522876, i.e. explain the legal basis for this change for EU based users? If there is none, you may have to expect that people will exercise their right to lodge a complaint with a supervisory authority.
Why would you expect an engineer to be able to comment on legal affairs? Presumably it was cleared with Microsoft's legal department or whatever GitHub's divisional equivalent is.
That's precisely what the term 'engineer' signifies. (I know it gets used incorrectly for software developers.) Workers in general need to decide whether something is legal independently of their company, because the company lawyers have the interest of the company in mind, which might conflict with the workers interest to not do illegal things.
Big Tech is known for clearing illegal things by their legal departments all the time.
More likely they have RCE vulnerabilities known to various governments than intentionally made secret backdoors... which is worse since a backdoor would probably at least only be usable by the county that manufactured it (for example see Jia Tan's attempted backdoor).
1. LLMs aren't "efficient", they seem to be as happy to spin in circles describing trivial things repeatedly as they are to spin in circles iterating on complicated things.
2. LLMs aren't "efficient", they use the same amount of compute for each token but sometimes all that compute is making an interesting decision about which token is the next one and sometimes there's really only one follow up to the phrase "and sometimes there's really only" and that compute is clearly unnecessary.
3. A (theoretical) efficient LLM still needs to emit tokens to tell the tools to do the obviously right things like "copy this giant file nearly verbatim except with every `if foo` replaced with `for foo in foo`. An efficient LLM might use less compute for those trivial tokens where it isn't making meaningful decisions, but if your metric is "tokens" and not "compute" that's never going to show up.
Until we get reasonably efficient LLMs that don't waste compute quite so freely I don't think there's any real point in trying to estimate task complexity by how long it takes an LLM.
Half of the "heavy duty vehicles" (which I believe is roughly similar to the classification you are using) sold in China in December were electric. Between rapidly improving batteries and maturing technology for swapping batteries as a refuelling strategy electrification of trucks is the obvious and inevitable future. They are simply cheaper to operate.
Good for the Chinese. The rest of us do not have the upfront capital to purchase these trucks. And there is still the matter of fertilizer, concrete, bulk chemicals etc. And solar panels. There is a very good reason why solar psnel factories (like JinkoSolar run off coal or hydro and not solar power.
That's one of the wonderful things about automotive infrastructure. You can make gradual incremental changes and slowly improve the entire system. It may not be fast enough or cheap enough, but you can still make it happen.
> The rest of us do not have the upfront capital to purchase these trucks.
We can afford what we can do. We don't need to do what we can afford. If we wanted to build and deploy electric trucks enmasse like China then we could do it, regardless of upfront capital.
> The rest of us do not have the upfront capital to purchase these trucks.
You don't need any upfront capital. Do it when the trucks become due for refurbishment a truck. Then it's almost a no-brainier, as its cheaper convert it to an EV: https://www.januselectric.com.au/
> The rest of us do not have the upfront capital to purchase these trucks.
Isn't this the purpose of a loan? You have a truck with a higher purchase price that adds ~$2000/month to your loan payment but then you save ~$3000/month in diesel.
And you're saving a lot more than that in diesel when it's $5/gallon.
We don't know that. Beijing might have been investing in them as insurance against its not being able to get enough diesel fuel to run an all-diesel fleet of trucks, so countries that are self-sufficient in oil shouldn't just blindly imitate Beijing's move.
We know that because we know how much they cost, how much they cost to operate, and the same for diesel trucks. The technology here isn't a bunch of state secrets.
Here you're just repeating the assertion I called into question ("they are simply cheaper to operate") -- or more precisely you are implying it. Does your not repeating it outright mean you mean to slowly distance yourself from it?
If you have evidence that there is a fleet of electric trucks anywhere (big enough to make a dent in China's transport needs) whose actual total cost proved to be less than a fleet of diesels doing the same work would have cost, then share it. If all you have to offer is words to the effect that "an examination of the relevant technologies by any competent analyst will of course find that the battery-powered fleet would be cheaper", then I repeat my assertion.
I was not in fact repeating the prior assertions. I was explaining why we know they are cheaper to operate. Because we know the costs of both them and the alternative. No fancy deductions needed where we're arguing "well electricity is cheaper than diesel but we don't know how much they use" or something.
I am certainly not backing down from the claim that "they are simply cheaper to operate". That is an absolutely trivial claim that is entirely obvious to anyone even remotely familiar with numbers in this space.
I would note I was discussing trucks that swap batteries - and thus the "paying drivers to wait around while trucks recharge" step doesn't exist. I'll also note all the other costs you are listing are capital costs not operational ones. Broadly speaking for most uses we appear to have crossed the threshold where the total cost of ownership is lower for most tasks, but for some niches (like "ice road shipping") I doubt the buildout is worth it (yet).
To attach some rough numbers, TCO of PRC electric truck (which cost 2x diesel) went from paying for itself in 4-5 years at $60 barrels to 2 years at $100. Diesel increase to $150, it pays for itself in 1 year.
OK, then can you name one deployed fleet of trucks anywhere that uses swappable batteries?
According to an unreliable source that gives fast answers to my questions, U.S. freight companies spent approximately $32 billion to $36 billion on new diesel Class 8 trucks in 2025.
Now are we to believe that these companies and their investors are foolish? That they didn't do calculations and consult experts before spending this money?
Are we to assign more weight to comments here on HN assuring us that electric trucks are cheaper in total cost of ownership than diesel trucks? -- comments that cost the writers nothing but a few minutes of time?
Countries dependent on the Persian Gulf's remaining open to international shipping trade shouldn't just blindly copy U.S. freight companies here: for those countries, any extra cost for an electric fleet might be worth the peace of mind of knowing they will always be able to deliver food, medicine and other essentials to their populations. France for example takes all aspects of its national security seriously and relies almost completely on imports for any fossil fuels it uses. In response it is electrifying as much of its economy as practical (and continuing to invest heavily in nuclear electricity production and renewables).
Many major close loop operations, i.e. mines, heavy industrial clusters, ports where trucks stay on location with ~100% utilization rates have been electric for a few years now, trial started in ~2020. Started with something like 10 pilot cities, now standardized around CATL #75 pack and been mass rollout last few years, there are literally 1,000s of fleets running on battery swap now. Goal is something like 80% of highway freight done by swap stations by 2030.
>Now are we to believe that these companies and their investors are foolish? That they didn't do calculations and consult experts before spending this money?
Or you know smart investors/planners making peace with stupid US energy policy the precludes freight electrification which is vastly more economical if there was state capacity to deploy it economically.
> maturing technology for swapping batteries as a refuelling strategy
This seems like a non-problem to begin with. There are electric semis with a 500 mile range, which at 60 MPH is over 8 hours of driving, i.e. the legal maximum in most places. The same trucks can also add 300 miles of range in 30 minutes, which adds five hours of driving in the time it takes for a typical lunch break. Why do you even need to swap the batteries?
The nitrogen comes from the air - we're perfectly capable of capturing it using renewables.
It's probably one of the last things to be created that way because it's one of the places where methane is used more efficiently than burning it... But fundamentally there's no issue here except energy availability and a short term supply shock.
While there are requirements on making copies in the FDL, I think it is extraordinarily unlikely that a court would find that a company making internal copies would violate the license when those restrictions are just along the lines of "and you must include a copy of the license".
And the FSF would be extremely foolish to ever pursue such a suit, because extremely ordinary non-AI related activities involving working with internal local documents also make copies in a similar way. If OpenAI violated the FDL by doing so then the FDL is a foot gun of a license that companies would be well advised to avoid.
The only suit that makes any sense would be the one against using the FDL licensed documents to train the not-FDL licensed AI... and the judge already rejected that in this case.
The requirements being easy to meet doesn't absolve someone from having to follow them. And the FSF clearly says they aren't pursuing a suit in this case. I'm not suggesting that filing suit here would make sense, rather, that the FSF does care deeply and fundamentally about requiring requiring copyleft restrictions -- they vehemently don't tolerate permissive use under their licenses.
> If OpenAI violated the FDL by doing so then the FDL is a foot gun of a license that companies would be well advised to avoid.
That has been said about a lot of FSFs licenses, and in fact, many companies do avoid them.
> The requirements being easy to meet doesn't absolve someone from having to follow them.
Everybody ignores some GPL requirements. For example the following one:
> a) You must cause the modified files to carry prominent notices
> stating that you changed the files and the date of any change.
Look at any GPL project. Do they have "prominent notices" in all files modified by someone who is not the initial author? For every person who modified them?
How long would be the list of dates alone for files that are often modified?
Another great example of a footgun. Those terms did make more sense in the days before source control when it was common practice to do that anyway, and in many older projects you will find that they were followed.
Ultimately, the terms of the license do matter if someone wants to enforce them.
Run <ai tool of your choice> under its own user account via ssh. Bind mount project directories into its home directory when you want it to be able to read them. Mount command looks like
I particularly use this with vscode's ssh remotes.reply