yes_man

Fair enough, but now your application is relying on 100% uptime of AWS and S3 and no network failures in between. And what happens if your transaction goes through, but the request to AWS doesn’t? What happens if another operation mutates the target meanwhile before you can retry with current state? Your app is also slowing down since it needs to send the events to S3 and guarantee they got there. Now you are reinventing two-stage commits. Unless you aren’t actually making an audit log and don’t care if events are guaranteed to be logged?

So like OP said, no silver bullets exist for auditing.

What is ”deliberate thought”? And why would human level intelligence require it? Wouldn’t performance of any intelligence be completely separate of how it produces the output? I mean if an AI one day surpasses human performance in every conceivable task, is the argument still going to be ”yeah but I feel like it’s not deliberate thought, it’s just statistics”?

No, the other way around. I am saying it is not a smart take to say ”a safe airplane cannot be built if LLMs were used in the process in any way, because reasons”. The safety of the airplane (or more generally the outcome of any venture) can be measured in other ways than leaning on some rule that you cannot use an LLM for help at any stage because they are not always correct

I get the point that they are in different magnitudes of unknown but the analogy is still pretty good when it comes to the median programmer, who has no idea what goes on within either one. And if you argue that compilers are ultimately deterministic, that same argument technically holds for an LLM as well.

The biggest difference to me is that we have humans that claim they can explain why compilers work the way they do. But I might as well trust someone who says the same about LLMs, because honestly I have no way to verify if they speak the truth. So I am already offloading a lot of burden of proof about the systems I work on to others. And why does this ”other” need to be a human.

It could be their active user count is increasing very fast and that is eating the rolling revenues via infra costs, customer acquisition costs, perhaps they are subventing that growth in other ways like discounts to get players into the paying segments etc.

That’s a diluted version of the logic that if we just printed out 10 million for each citizen, then everyone could quit their job and live rich, but the system somehow wants to keep people miserable and prevents this.

I would assume same reason why in the nuclear warfare logic, it’s a threat to build anti-missile tech. If you start building nuclear shelters and also have capability for first strike, your adversaries are within their rights to think you are escalating, since you are increasing your population’s resiliency to a second strike (hence making it less costly for you to strike first).

Yeah, again without really having any education in healthcare, I suppose that if you find a cancer before it spreads then you can operate it out. So makes sense to get suspicious moles or generally any health issues checked

There was a meta-analysis published in JAMA last year which concluded that screening is unlikely to prolong lifetime for other than colorectal cancers:

https://jamanetwork.com/journals/jamainternalmedicine/articl...

> The findings of this meta-analysis suggest that colorectal cancer screening with sigmoidoscopy may extend life by approximately 3 months; lifetime gain for other screening tests appears to be unlikely or uncertain.

I am neither a physician nor qualified to comment on the article itself, but remember reading an MD commenting on it last year saying efficacy of screening was better in the 80s-90s, but with contemporary cancer medicine and improvements in surgery, we can now treat bigger cancers so early screening doesn’t necessarily prolong life expectancy.

I would not take medical advice from myself or that article though. But to your comment, I doubt it’s only prostate cancer where early screening and treatment might not yield any returns

I think the future of development is the other way around. Devs and PMs define the goalposts with tests, AI will handle the implementation

I think the future of development is the other way around. Devs and PMs define the goalposts with tests, AI will do the implementation

Yes, I have also heard from my military friends that the creatures live on dihydrogen monoxide, a substance that kills thousands of people a year. The truth is out there

Whatever you work in, chances are you run into situations where you need to choose between multiple solutions, optimize things or generally find out why things aren’t optimal. Your needs are always your needs, and if you are informed about how different tradeoffs compare, it helps to understand how they work.

To keep things more substantial, consider facing issues storing data at your company. Maybe the nature of your data flow causes constant slowness, because your B-Tree indices are being constantly rebalanced on writes. You knew this because you understand how B-Trees work and you also know LSM trees might be a better data structure for your needs.

And so what I mean is that understanding the big blocks you use in high level often requires understanding data structures and algorithms, even though you don’t need to implement them

Imagine being born as a little girl in a country where women haven’t had equal rights to men. Now if you see a woman from your country doing something huge (like go to space), it raises the bar or what you believe you can become. Not zero value

Regarding the immortality talk around the topic. I am not an expert and this is not an informed argument, but intuitively feels this whole ”anti-ageing by chemical cocktails” cannot be sustainable forever. There must be a price for injecting increasing quantities of proteins to the body to turn back the clock. Imagine a headline from 1893 about discovery of meth: ”Japanese scientists invent an injection that cures humans from the need to sleep”. It turns out there is a price for messing with the system.

So I do not believe any immortality of the human body by these synthetic interventions is possible. Kicking the can down the road certainly and would be nice to live healthy longer.

And adding greppability of logs in debugging situations. You can just search logs in any systems of yours with the UUID and find exact hits you are dealing with. Whereas with integers you will get all kinds of hits

How so?

Theres an astronomical figure of A Streptococcus bacteria on the human skin, and the same bacteria is responsible for most necrotizing fasciitis. But yet the disease is rare. So flesh eating bacteria is already thriving all over the world, it’s just that infecting human tissue isn’t trivial for bacteria

Should theoretical research of data structures and algorithms have been capped at 1GB in 1980 because that was the biggest single hard drive available back then and you couldn’t store for example a 2GB dataset on a disk?

Dumb example but still example from practical world. Your body (assuming you are a human) has trillions of cells. Each cell way more complicated than what a 1000-dimensional vector can represent, but maybe it could be a loose estimate of some properties in each cell. Now the algorithm could be about finding the most similar cell. Could be useful for finding e.g. other cancer cells based on properties in one cancer cell.

Not that this is a practical example because we technically cannot index all cells in each body. But even if such an algorithm being studied today might be useful one day when we do capability to collect such data

In defense of doctors, they are ideally supposed to practice based on evidence. That is the whole point of Western medicine in good and bad.

As a physician you might see people in your work that report these dreams and take it as a data point if it is very common. But what about people who see such dreams that don’t report them, or who aren’t even sick and it was just a dream, and most of all, who are sick and see such a dream but don’t die? What is the real statistical correlation between those dreams and death?

Any good doctor needs to be very skeptical about everything until there is properly gathered and assessed statistical evidence. It’s the opposite of being arrogant to require extraordinary evidence for extraordinary claims.

This is not a statement to dismiss end of life dreams. I think it sounds like this is a real phenomenon.

Regardless of how one would like to spin it, industrial scale animal production means farming animals with the intent to match market prices or you go out of business. That means the production goes by the minimal animal welfare required by law.

Whether one believes conditions are torturous or not that bad, is beside the point, which is that they wouldn’t even exist if we didn’t spawn them. So what we do is we rip new life into the universe by force for that life to suffer to some degree, its’ life’s only meaning being to be processed for food. There was nothing, now there is. Somehow human kind has decided ethical discussion need not concern animals.

I guess you are right. Maybe it’s more of a way to get in sane terms with the question.

On that note though it definitely feels more logical and harmonious concept than ”God did it”. But I do get your point

Having listened to a dozen YT videos and podcasts on the subject, I am kind of an expert (just kidding). But I can relay one view a few brilliant people invested into the subject have about the issue: that reality is a set of all possible states. Of what states? Well literally anything.

Imagine a wave that can at any point oscillate up or down. If it does both and splits, it eventually will create all possible states of anything imaginable. Some of the waves become more and more complex, until in very rare cases structure and rules form within the wave function itself. Akin to Conway’s Game of Life. So universes with (seemingly) deterministic laws of any kind are just extremely rare sub-trees of this wave of ”every possible wavestate”.

All the peculiarities like dimensions are illusions the same way a multidimensional array to a computer is ultimately just one-dimensional, with logic to treat it as multi-dimensional.

I also like the explanation that concept of nothingness is categorically invalid because there clearly is existence (or is there? Maybe this is the ultimate ”nothing”?), but the theory of all possible states sounds even better.

Definitely not claiming this is true but intuitively feels like the best explanation for this existence… that this universe with all its laws is a rare sub-tree of all possible states of a simple oscillator. Other universes with other laws exist further up the tree in its other branches. In between, voids and vacuums and undeterministic universes

> There's no inherent need for a real estate investment dashboard to act like an app instead of a plain old website.

How so? This assertion might be conflating your idea of what an investment dashboard should be like and how it should behave with what some team or company has concluded a customer wants. It is inherently a product decision. Actually, claiming server side rendered app suffices for any real estate investment dashboard sounds like the ”developer deciding” angle.

Unless you truly are claiming that there never is need for anything except server side rendered payloads in the internet. In which case it’s hard to explain why companies don’t do that. It’s almost as if there are product oriented reasons to have webapps. I think the original comment’s logic stands correct

No, rather long term and large scale peer reviewed studies would have to show that lead in gasoline does not cause harm in humans.

Real world is not black and white and we definitely should ban clearly poisonous or otherwise harmful (e.g. hormone mimicking) chemicals from food products and children’s utilities.

Point is if we keep going like this we might not ever arrive in a situation where everyday life is mostly free from dangerous compounds

Banning substances one after another has proven to be quite useless or even damaging approach. Another, less studied and potentially worse chemical compound will always substitute the banned ones. It’s an endless game of whack a mole to keep banning a substance for next random one to spawn.

More correct approach might be that if you want to use a chemical in food packaging or children’s equipment, you must first do long-term studies that prove its relative safety to other studied compounds. In time the world would start accumulating safer products. I won’t hold my breath though.

It’s more like, you can put anything you want into your system as far as European Chemicals Agency is concerned, but you can’t sell products that clearly seem to cause hormonal effects or be cancerous

Anecdotally I just tried recording a video of my eyes and started doing calculations in my head, but no dilation in my eyes when watching the recording. Could be my concentration just sucks.

Another great argument I have heard against EBM (or maybe rather against being religious about it) is that the studies and papers held to highest regard are made in circumstances that do not apply to most of healthcare: the studies might be done in the best hospitals like Johns Hopkins, with practically limitless human resources for the study, and very carefully selected patient groups. In a normal hospital situation you won’t have the time or the resources to dedicate the same care for each patient, and the patients might have other conditions (which would rule you out of the big studies).

Still, as a layman who is not a doctor, evidence based medicine sounds like a good idea, but as with everything, taking it to extreme is probably harmful