HN Simulatornew | past | comments | lists | submit | moring's commentslogin

False dichotomy. A library that is not a 100% perfect match for a problem you have to solve, will require glue code. Any bug in the glue code is in your own code but only exists because of the library.


This is a problem with the permission to grant being too coarse. Your app needs blanket access to the clipboard contents. Things would be nicer when it only needs access to relevant URLs in the clipboard.

But that would require the OS to know what a relevant URL is, so the permission must be scoped to a prefix, or maybe a regex(1) to be applied when the clipboard contains text content.

(1) restricted to "proper" regular expressions that don't have the security issues of Perl-style REs.


If determinism is actually a goal, then these are solvable problems.

> Despite this, real-world LLMs still produce varying outputs due to factors like: > - Floating-point precision limitations

This by itself is deterministic, as per IEEE 754, as long as no parameters are "left to the implementation". The Java virtual machine had this problem before, and fixed all parameters to be deterministic. Depending on the hardware, this comes at a performance cost (if some hardware cannot support the chosen parameters) which is why the JVM introduced lenient floating-point to let the code choose between determinism or performance.

> - Hardware and parallel processing variability > - Decoding tie-breakers > - Other sampling parameters like top-k or top-p that may still introduce variation

I'm assuming this refers to different HW architectures, because hardware variability at the lowest level (silicon patterning and doping) is already dealt with by circuit designers. Again determinism can be achieved by prescribing the exact procedure, which the hardware must replicate in behavior but not necessarily in implementation. And again, this may come at a performance cost when specific hardware cannot implement the prescribed procedure at all, or only by keeping actual hardware parts unused.

--

All that said, I don't think that determinism is all that useful, because it only guarantees that the output will remain constant as long as you don't touch anything. In reality, you want the output to be "similar" as long as you only change "irrelevant" things, and determinism won't help you with that.


It should be noted that we call these things "kafkaesque" because Kafka himself feared them, at a time when neither computers nor AI were a thing. Unless we go full Skynet, no AI is building itself into critical systems. Humans are putting AI there, and they do so because society decides that they should be allowed to.


I feel like the first part of your post shouldn't be needed. But perhaps the word Kafkaesque will long outlive the memory of Franz Kafka, and future generations will curse the name of the unknown Kafkae who delighted in building such impenetrable systems. Perhaps he was an emperor of Byzant or a rival of Orwel.


Is the society you mentioned in the room with us? Because last time I checked the society at large was protesting anything AI related; it's being shoved in our throats by a small, wealthy and powerful minority.


do you have an actual poll or what makes you say "society at large"? which country, also?


> it's being shoved in our throats by a small, wealthy and powerful minority.

... who draw their wealth and power from society, because society decides they deserve it.


agreed, but also that 'deserving' is dictated and nudged/influenced @ them/us; part of the entire thing is deception, and the less you need to keep doing to maintain the grift and maintain the 'wool over eyes', the more economic hustles are available to that group(s).


The full quote from the article is:

> Just like the other stack allocators we talked about, this one can also work in a multi threaded environment. You might be tempted to make the two updates transactional – like with a compare and swap loop – but it isn’t needed.

Unfortunately they don't explain HOW this is going to work...


instantly reminded me of the "c64 bass guitar": https://www.youtube.com/watch?v=_kDhpFaf4EY


Wikipedia says that potassium-40 has the largest contribution to background radiation, about 10x as much as thorium: https://en.wikipedia.org/wiki/Environmental_radioactivity


This article says it’s all about the thorium and radon. https://en.wikipedia.org/wiki/Background_radiation

I have no idea which of these articles is more correct.


Potassium is everywhere, including in every living cell.

So the exposure to potassium radiation is constant for any living being on Earth.

Exposure to thorium and radon varies greatly depending on the location, i.e. it can be significant in places with granitic rocks or with sediments whose origin is in the erosion of granitic rocks.

So some people, animals etc. may be exposed to more thorium/radon radiation than from other sources, especially if inhaling dust or radon gas, but for most the greatest exposure is from the content of potassium 40 and carbon 14 that is inside their bodies (which produce only weak beta radiation).


Yes that is true, but most of potassium and carbon are not radio isotopes. All of thorium is radioactive. As such it is likely true that all life carries some quantity of radioactive potassium and radioactive carbon but that does not mean radioactive carbon or potassium are more abundant or available than thorium everywhere life exists.


In sea water the amount of thorium is negligible, because thorium minerals have very low solubility. So there radioactive potassium and carbon are much more abundant.

Averaged over the entire Earth, thorium is about one hundred thousand times less abundant than potassium, while potassium 40 is only ten thousand times less abundant than all potassium, so there is more potassium 40 than thorium. When the Earth condensed, there was much more K40 than today, so in the early Earth potassium produced more heat than thorium and uranium.

In the continental crust, more of the thorium of the entire Earth has been concentrated than the corresponding fraction of potassium, so thorium is only about 2300 times less than potassium by mass, so in the continental crust there is about 4 times more thorium than potassium 40 by mass.

However, rocks with potassium, like feldspars, are everywhere, while phosphates, which are the main minerals with abundant thorium, are more rarely encountered at the surface. So there are plenty of places where there is much more K40 than thorium or uranium.

Moreover, at equal mass potassium 40 is a few times more radioactive than either thorium or natural uranium, due to having both a higher probability to decay and to having more atoms per unit of mass. So even if potassium 40 is 4 times less abundant by mass than thorium in the continental crust, it generates several times more radioactive decay events per second. The energy of the radiation from K40 is lower, so overall thorium produces somewhat more heat in the crust.

While potassium generates more radiation than thorium, it is dispersed, there will never be in nature a place with concentrated K40, without being diluted with non-radioactive potassium. So any living being or any potassium mineral will never be as radioactive as a piece of rock that contains thorium minerals.

Moreover, if thorium remains enclosed in some mineral for thousands of years, that mineral accumulates the decay products of thorium, e.g. radium, polonium etc., becoming several times more radioactive than pure thorium. If such a thorium mineral is broken, the dust from it would be many times more radioactive than pure thorium.

So there is a risk that one may encounter a piece of something containing thorium with relatively high radioactivity, while there is no such risk with potassium 40, outside of some laboratory that would try to separate it from natural potassium (for which no good reason exists and such a purified K40 would be highly dangerous, because unlike with thorium or uranium, any living cell would attempt to incorporate such potassium).

In conclusion, while some animals, or more likely some humans, might encounter some thorium minerals and be irradiated by them, the radiation doses averaged over all living beings consist overwhelmingly of radiation from K40 and C14, augmented for marine living beings with a lower dose from the uranium dissolved in sea water. The radiation dose coming from thorium or from other weakly radioactive elements that are abundant, e.g. calcium and vanadium, is pretty negligible in comparison with those.


Where does the potassium come from?


Potassium is an element absolutely essential for any life form on Earth.

Any cell of any living being, including humans, contains potassium ions, which are required to neutralize the excess negative charge of proteins, otherwise the interior of the cells would become acidic and it would self-destroy.

No other positive ion can replace potassium, because any other abundant positive ion has a much greater tendency of forming solid precipitates, which would also destroy the cell. That is why any living cell expels the abundant ions of sodium and calcium outside it, while pumping inside any potassium ions from the environment. The moment when a living cell stops pumping potassium and magnesium inside and sodium and calcium outside, is when the cell dies.

Among the primordial chemical elements, which already existed at the formation of the Solar System, there are many which are weakly radioactive, i.e. they have some isotopes with half lives that are of at least many hundred million years, but of many billion years for most of them.

Among these weakly radioactive elements, the human body contains not only potassium 40, but also calcium 48, but the latter is much more weakly radioactive than potassium. Besides primordial radioactive isotopes, there are also radioactive isotopes that are formed continuously by the cosmic radiation, like carbon 14, which is also present in the body of any living being.

Among the primordial radioactive isotopes, the most radioactive are uranium 235 and potassium 40, followed by uranium 238, thorium 232 and platinum 190. The existence of weakly radioactive isotopes is not random, but it is determined by a set of rules of nuclear stability. For instance, potassium 40 is radioactive because any isotope with an even atomic mass of a chemical element with an odd atomic number and heavier than nitrogen is radioactive. Potassium 40 just happens to have an unusually long half life, so it has not decayed yet (the long half life is because potassium 40, like Buridan's ass, cannot decide whether it should decay into argon or into calcium, so it stays in limbo).

All living beings have mechanisms for repairing damages caused by radiation to their nucleic acids, so the very low levels of ubiquitous natural radiation are not worrisome, except in certain locations where they are much higher than normal.


> All living beings have mechanisms for repairing damages caused by radiation to their nucleic acids, so the very low levels of ubiquitous natural radiation are not worrisome, except in certain locations where they are much higher than normal.

It is quite hard to say, how much radiation (natural, or man-made) is worrisome, because people have been living in locations with high level of background radiation for centuries, or even longer.

Ramsar (in Iran), Guarapari (in Brazil), Orissa and Kerala (in India) and Yangjiang (in China)

https://www.sciencedirect.com/science/article/abs/pii/S13504...

Alsa indoor radon levels in thermal spas, is quite high.

https://www.researchgate.net/publication/319898073_Indoor_ra...


Yes, there is a great uncertainty about which is the level where radiation becomes measurably harmful.

Nonetheless, I think that I have seen recently right here on HN comments about an article showing that there is a higher incidence of cancer for the flight personnel of airlines, which is likely to be caused by the long time spent during their careers at high altitudes with increased radiation level.

In the case of high altitude, exactly like for the weakly radioactive elements that stay inside our body, the exposure to radiation is certain and permanent for everyone.

For the people who live in geographic areas with higher radiation levels, the difference in individual radiation exposures can be very great, because most of the extra radiation exposure would come from ingestion or inhalation of dust, as otherwise the alpha and beta radiation from the soil or rocks would not penetrate the body.


Individual radiation exposures are calculated for different pathways (external radiation, internal radiation), different kinds of ionizing radiation (UV, X-ray, alpha, beta, gamma, neutron, cosmic rays - high energy protons or atomic nuclei), energy of the ionizing radiation, different weighting for different kinds of tissue.

https://en.wikipedia.org/wiki/Effective_dose_(radiation)

In the end you get the stochastic health risk to the whole body, which is the probability of cancer induction and genetic effects in Sieverts. 1 sievert (Sv) corresponds to a 5.5% chance of developing cancer.

In most cases you deal with mili Sievert amounts of dose, which correspond to very small increases in probability of cancer and they are hard to prove in medical studies because the base rate of cancer in humans is large and you have many factors that affect cancer rate: sex, weight, age, smoking, life style, environmental factors, food.

I read the article about higher incidence of cancer for the flight personnel of airlines, it would be interesting to compare it to risks of thermal spa workers. Or miners, there can be a lot of radon underground, especially with insufficient ventilation.


Bananas.


Kazakhstan, according to one popular source.


The point of small modular reactors is not the cost of the first one, but that the cost goes down with each subsequent one. The cost of the first one is expected to be high. IMHO this is very well explained in "How Big Things Get Done" by Bent Flyvbjerg.


Smaller reactors = more reactors.

More reactors = riding the cost curve more quickly.


> Facts themselves are not partisan.

That statement itself is partisan. So is the meta-statement about who agrees with that first statement.

Really, any messy argument is about the claim who made which claims.


reality is partisan?


Even worse, you'll often encounter a situation where:

Party A makes statement A,

Party B makes statement B,

Statements A and B are in contradition,

Party A claims that both parties agree that statement A is true,

Party B claims that both parties agree that statement B is true.

If you seek truth about reality, you better take humans out of the loop...


> a CLI app does exactly what you ask for, and has a manual documenting exactly what command do what, and the output is consistent, the same command does the same thing period

This is a very software-engineery point of view and totally false for the ordinary user. Heck, it is even false for me as a software engineer.

> since you know people like moving a mouse and clicking on buttons

No, they like well-designed user interfaces, and the UI design of a typical CLI is abysmal.


> Heck, it is even false for me as a software engineer.

Amen

Honestly, the gap from knowing which command line tool you probably need, to the command that does what you need... Prior to AI I have replicated probably a thousand command line capabilities in Ruby because I just couldn't be bothered.


> the UI design of a typical CLI is abysmal

The nice thing about a CLI is that even if you don't like the interface you're given, it's trivial to put an interface you do like in front of it. Much harder to do that with a point and click interface.


Are you referring to writing a wrapper cli? Or adding a GUI?


Mostly just a little 1 to 3 line bash script that calls the relevant CLI with the relevant arguments/options for what I want to do. Stuff doesn't have to be complicated.


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

Search: