stefs
stefs on Hacker News
i'll have to recommend "practical doomsday" here -
https://lcamtuf.coredump.cx/prep/ (edit: because it spends a lot of time on the topic on how to secure your wealth). or ask lcamtuf himself.
there are areas where almost all options for pumped hydro are already developed and there's little room for expansion. additionally, the heat waves from the climate catastrophe affects pumped hydro negatively (evaporation and higher water temperature).
that's a one time cost over many years that's quickly amortized. lets assume an upfront cost of ~100kg CO2 per kWh of battery storage and 250g of CO2 per kWh generated by a gas turbine. this means the carbon ROI is there after 400 kWh.
i'm also using a "type safe" templating engine (for kotlin, as a gradle plugin) of my own design that goes the other way - instead of embedding a model class into the template, a template class is generated that takes all the values as parameters.
i.e. MyTemplate.tpl.html
{if $foo}hello world, {$myname}!{else}goodbye, world.{/if}
which creates a kotlin class that can be used like this:
MyTemplate(
foo = true,
myname = "bar"
).toString()
the biggest upside for me is that this way there's full, free IDE support, i don't need to write extra view models if the mapping isn't 1:1 and performance is pretty good because in the end there's no runtime parsing/compiling/interpreting except for the baseline JDK stuff (though there are even faster engines).
https://codeberg.org/h34tnet/tempolin
warning: so far i'm the only user though, so don't use it in production. also, i'm working on some improvements right now which haven't been pushed yet.
i looked up what alternatives there'd be and openrouter also makes its search API available. it's exa too, but $4 per 1000 results.
iirc i'd say it's the latter, as it's not scientifically possible to prove the absence of god. as far as i remember, the books don't introduce anything that might suggest new proof for the existence and the "aggressive absence of god" is just the continuation of the current trajectory where no further proof can be found for another 55 years.
Open weight and -source models aren't without biases. Giving a model away for free would be a good way to distribute an agenda. With a sovereign model you can at least control the agenda and align it with your values.
the assassination of archduke franz ferdinand was just the trigger but not the reason. the war would have started anyway.
i agree with this. the kremlin sees this as a win-win strategy. there are two options:
1. either their actions are ignored by the victim states, which makes this very cheap and effective sabotage to weaken EU members.
2. or alternatively, if there's military retaliation, putin can finally do the mobilization he promised not to do for his war on ukraine, because he can now assert that the country is under attack by foreign nations.
the EU / the member states are now in a pickle: they know it's russia but can't say it out loud because then they'd have to take action. they also can't do nothing because then russian sabotage could wreak havoc with the population and economy. so what can they do?
1. military support of ukraine
2. further sanctions of the russian economy, which includes getting rid of dependencies on russian goods
3. re-arming and modernizing their own military (i.e. introducing modern drone warfare capability, launching their own satelite intelligence, ...)
4. start opposing/controlling russian influence by local malicious actors, i.e. russia friendly right wing political parties (which is somewhat complicated but not impossible in a democratic setting)
part of the answer is the peace time dividend. the decrease in european spending has mostly been used for lowering or abolishing taxes on the corporations and the wealthy. if this was to be reversed (because we're not in a peace time anymore), the added tax income could be used to finance the re-armament.
> Congress to close the loophole
But i think one of the prediction markets more important uses is to enable insider trading for the current administration.
i don't think that's correct. SEO has been popular long before google. many websites were overloaded with long lists of unrelated keywords at the bottom.
google became the most popular search engine exactly because those SEO keyword spam lists didn't work anymore and, after a while, they vanished. blogs, on the other hand, worked exceptionally well, because they used to link to other blogs' entries a lot, and blogs also used to be (comparably) high quality content.
so the main SEO guideline for getting a good page rank used to be: have a blog/news and publish lots of high quality content that bloggers link to.
i feel like that's still a kind of security by obscurity. the LLM isn't guaranteed to trigger a booby trap.
in my opinion there's no _fundamental_ reason between an AI agent and a human adversary attacking your code, just a practical difference. in the long run the only way is to fix the bugs.
but i agree with you that in practice, it's also a question of cost effectiveness. booby-trapping your code is probably making attacks cost prohibitive for almost everyone (which is the same situation as before, when attackers had to employ security researchers/hackers or buy zero-days from 3rd party providers).
running SOTA models is expensive, so the attackers aren't going to do drive-by attacks. next generation cheap models might do the same, but they can also be used by the defenders.
that's not a solution, that's just an impediment - at least if the agent can keep persistent memory about its state between attempts.
this is addressed in "Solution #3: Allow marketplaces to represent and manage micro-enterprises collectively as if it were a single producer."
so apparently, what you are describing is not what's actually happening.
i'm a fan of fettes brot too. i didn't know that about schiffmeister!
ps: austrian here, i'm about as landlubber as possible.
same here, on my amiga 500 and later on the PC. i was a kid and barely able to understand what was going on (but that didn't matter that much anyway), my english was absolutely inadequate for the text. i remember pestering my father about the vocabulary: "what does 'squashbuckler' mean? what's 'swashbuckler'? what's 'merchantman' and 'the flying dutchman'?" (the last was actually a misreading of "she's flying dutch colors" when you looked at a dutch ship).
the worst was when i got too far north-west and had to sail back into settled territory against the wind, which took ages.
something else that stuck in my mind was the soundtrack - iirc mostly händel and bach.
ps: years later i found the "mount & blade" series (which i too love), which is roughly the same but medieval and on land.
the ibex install worked flawlessly on the first try while installing windows (on the same machine) failed due to driver issues.
years ago my father got a decomissioned, wiped clean notebook from work and asked me to set it up for him. i didn't want to shell out a hundred bucks for windows, so i suggested trying linux (ubuntu) and he agreed. "intrepid ibex" was installed without issues. a few weeks later i asked him if everything was ok and how he liked it, and he had no complaints, except one: he didn't like the edgy skull on the desktop wallpaper:
https://en.wikipedia.org/wiki/Ubuntu_version_history#/media/...
> My hope is that in a couple of model generations, we'll trust AI to review MRIs the way we trust it to proofread our emails.
the doctors may have financial motives to suggest unnecessary treatments. but i fear, by the time we'll have models powerful enough we'd be _theoretically_ able to trust their expertise, the financial motives will have shifted to the model creators.
somewhat funny to the immature mind: "fut" - the name of the compiler binary - is austrian slang for vagina, but in the most obscene, vulgar way possible. a bit like "cunt", but worse.
i think what they're talking about is an attacker poisoning the data the agent is trained upon to include functionality/a backdoor that can later, after training and when the agent is deployed, be used to induce unwanted behaviour.
that study is widely regarded as unreliable and outdated.
increased testosterone from working out is probably around 10-30% long time, which is a far lower variance than natural level variance in healthy adults. i think i heard from several (claimed natural) strength and bodybuilding athletes that their total testosterone is at the lower end of the scale.
that said, natural free test levels are at a fraction of what enhanced pro bodybuilders tend to supplement, and there are mass monsters with hair. cutler, yates, ferrigno and golden era bodybuilders like schwarzenegger, zane, columbu all had full heads of hair.
yeah, but if you really really wanted to and/or your livelyhood depended on it, you probably could afford it.
