IKantRead

Interestingly enough the Marxist variation on this is that, if we assume Marx's Labor Theory of Value (I know, big assumption for most, but let's just go with it and see what the conclusion would be): All surplus value is extracted from human labor, automation (including AI) is just efficiency which is used by Capitalist to gain temporary advantage over other Capitalist. This advantage is always temporary.

The second part of this logic is quite apparent: Assuming AI is game changing, early adopters will see an initial advantage, but in the long run everyone will have to use AI or no longer be competitive.

But the LTV (remember, for the sake of exploration we're assuming this is true) argues that value is always extracted as surplus (i.e. unpaid) labor from humans. So LTV would predict that if AI truly replaces humans we would ultimately see a decline in the surplus value that Capitalists extract, i.e. we would b faced with an economic crisis.

So LTV would predict that we would have either just as much or more hours of human labor happening or we would enter an economic crisis.

To add to what you're saying: It's not like the Object-relational impedance mismatch[0] is some great unknown property of ORMs. Since people have been putting ORMs into production designers of these systems have been well aware that you must always chose a trade-off between a full functional object system and a fully relational one.

And the mismatch has two sides. If one's answer is "just use SQL!" then you're going to have new problems dealing with the mismatch coming from the SQL side of things.

The pre-ORM solution to this was not simply to shove a bunch of SQL in your application logic (though this was done), but to have a much, much more complex set of data layer logic. I'm guessing most people today writing about eschewing ORMs in favor of pure SQL have never used: stored procedures, triggers, cascades etc. I personally do miss some of the features from that era of software, but there's a lot more complexity to the "just write SQL" approach than most people realize.

0. https://en.wikipedia.org/wiki/Object%E2%80%93relational_impe...

Are you confusing Wittgenstein with Russell?

Tractatus is basically the opposite of "windmill tilting", it concludes that the vast majority of philosophical "problems" end up being essentially "windmills" in the context of that expression, and that what remains is trivially solvable and not particularly interesting.

Wittgenstein, during his lifetime and after, was often confused for logical positivist, that is all that matters is logically provable statements. But it's quite clear from his further writings (and a thoughtful reading of the Tractatus) that what is being said is: if you're limiting your discussion to logical statements (i.e. philosophy has he understood it at that point) then what you can talk about is limited to logical facts and their consequences... unfortunately this doesn't cover much ground in human experience. For all of those other parts of human life, (logical) philosophy doesn't have anything to say. Hence the concluding “Whereof one cannot speak, thereof one must be silent.”

I also wished we had an academic world that valued that advancement of a field that wandering the world in thought and then showing up with a ground changing work was enough to get a Phd.

It's hard to imagine a new Wittgenstein in the world, even with LW's connections, being able to show up at 40 years old with a work in hand, defend it and walk away with a PhD.

With 10+ years in DS, I've always felt that best DS were always basically software engineers that knew math and were more interested in prototyping cool machine learning product than maintaining production infrastructure. Unfortunately this always accounted for a small fraction of DS I interacted with.

The largest group of DS was non-ML/CS/Math PhDs who started panicking once they realized their future job prospects in academia were very slim and so they signed up for bootcamps and got jobs at places hiring DS by the hundreds. Many of the people in this latter group had no idea how to write Python outside of a notebook, generally just structured problems to fit into XGBoost, and when not doing that tried to squeeze resume-boosting-complexity into any problem the could find. They also tended to have a hilariously poor understanding of creating business value.

Nearly everyone I know in the first group has switched back to just being an engineer of some sort, typically ML or AI engineer. I suspect the small set of talented people from the second group will end up in lesser paying product analytics type roles or closer to product management roles, while the majority that don't bring much to the table other than a PhD will be slowly attritioned out of the field as companies start looking for the value different skillsets bring to the table.

> "I don't see my seeing of a flower"

Then you must have a fundamentally different qualitative experience of consciousness than I do.

The flower I am holding right now, it is real to me because I can observe myself experiencing it, however the flower you are holding I cannot experience because I am not observing your experience of observing the flower.

It is quite remarkable that you can experience that flower that I am holding and say that it's real while I cannot do the same for the flower you are holding.

> "A dog which sees a flower knows there's a flower"

I certainly don't know enough about the state of my dogs mind to know that my dog is seeing a flower, as opposed to say, merely experiencing clustered sensations in the same location as I am seeing a flower. I've looked at a flower and seen my dog eat it, so I have a suspicion that he recognizes something there, but only because my beliefs about how his actions might correlate with seeing something.

I also have friends who have seen a purple flower in the exact space I am looking where I cannot see a purple flower. What is the explanation for this?

> Scepticism does not falsify reality

I don't understand why you're even making this statement, no philosophical skeptic would disagree with this. Skepticism has always attacked our understanding of reality, not reality itself.

If anything it is the idealists, like Leibniz that argue that "reality is only in our minds", verging on solipsism. You seem to be misunderstanding Hume who more so points out that the only thing we can really "know" is our inner experience of reality, which is a bit closer to what you're arguing.

I'm not even sure what "really is a flower" even means.

Surely you'll at least agree that the concept of "flower" is a mental object that we use to classify a set of related sense perceptions.

If you're going to call skepticism "naive" then you have to at least outline what you mean by "a flower" in relation to the experience of observing one. I think you'll find the number of propositions required is not small.

> That era of philosophy had yet to learn that philosophical propositions are themselves subject to scepticism

Hume's fundamental point was that even causality cannot be meaningfully verified as real so skepticism about philosophy itself was already alive and well at that point.

It feels a bit like contemporary physics has caught up to the concerns of 18/19th century philosophers.

Descartes, Leibniz, Hume, Kant and others, while disagreeing with each other in major ways, all agree on one big concern: We're fundamentally limited in our knowledge about "things in themselves" outside of our consciousness and the perceptions we receive.

Naively we think "oh, looks there's a flower!" but for these philosophers a big question was: how much can we really know about the actual thing that is the flower, beyond the (potentially suspect) information that our mind receives through the senses and our mental models (include the concept of 'flower').

In a very practical way I think we might be bumping into these limits. Which isn't surprising since what was driving these questions for those philosophers was concerns about the foundations of the then rapidly emerging scientific fields.

To add to this sentiment: I increasingly find the idea of "reading a book" (in the sense of "completing a book") when talking about serious works of literature, philosophy (or even technical books) somewhat nonsensical.

In technical books people often criticize recommendations of TAOCP or SCIP because "nobody really reads those", however this misses the point that books of this nature are perpetual projects for the reader. I have moved my eyes over every word in SCIP, done ~20% of the exercises, but would still say I haven't "read it" in the sense of completing it.

This is even more true for works of philosophy. I would be far more impressed with someone saying that had spent the last 3 years working through the first section of Spinoza's Ethics (a fairly lean volume) than someone who claims to have read Heidegger's Being and Time over a break.

I was talking to someone recently who remarked that it's embarrassing to see a bookshelf covered with bookmarks visibly only part way through the text, implying the reader has never "completed" the work. I found this view somewhat shocking as the older I get the more I realize you never finish a good book.

I suspect many people on HN have never had a proper office with a door.

While I do agree with your overall premise, having had offices at multiple roles over my career, it is a major improvement over cubicles and open office.

Having a space to work in complete, uninterrupted quiet is wonderful. The simple social signal of the door being opened or close implying availability to chat works great to balancing heads down time and chat with colleagues time. Have a white board of your own that you can stare at throughout the course of a project, that evolves with discussions has not be replicated yet either remote or in a shared space.

You can also get pretty wild with customizing an office for comfort. I used to know many people that brought their own lighting in, completely solving the problem of constant overhead fluorescence. And brewing your coffee the way you like it in your own space in the afternoon is very refreshing.

Even given all that I'm sure many people would still prefer remote work (I would at this point), but I can image a fairly large number of people that would be very interested in a set up like this. It also mean your home is completely work free again.

I should have specified gasoline consumption since this more closely maps to what we assume EVs will replace.

That has been increasing overtime, barring of course the pandemic, we're on track to pick up from where 2019 left off this year: https://www.statista.com/statistics/188448/total-us-domestic...

> We are making progress and as long as the pessimists don't get their way.

I've been hearing a similar statement for decades now, but the global emissions only continue to rise.

In fact, I've been hearing that logic since the late 90s and in that time we have doubled the cumulative emissions up to that point.

We have seen that emissions drop with decrease in economic activity (multiple times). Yes, batteries, renewables, and nuclear would all part a big part in a truly sustainable future... but we have to survive until that future first.

> Reducing carbon emissions isn't a solution

We're well past the "solution" stage of this problem. Where in the "harm reduction" phase. Reduced emissions is absolutely superior in this regard to increased emissions... and yet you're arguing that we should be increasing them because a magic solution is just around the corner, like it always has been.

> There is no motive for truth

This also ignores the larger question that has been a known issue for at least 2,000 years: "Quid est veritas?"

The hype has also been great at fueling economic activity which contributes even more across a wide range of sectors.

We know how to reduce carbon emissions immediately, we did a great job of it 3 years ago, we just don't like the answer: reduce economic activity.

Consume less, produce less, work less, spend less. These things will do more to reduce CO2 emissions than if everyone rushed out and bought electric cars (in the US ~oil~ gasoline consumption has been positively correlated with electric car ownership).

But we don't want that, and more importantly, the stock market doesn't want that. I think most people would be fine with reduce work and reduced consumption (to a point), but the people invested deeply in the status quo tend to have asymmetric influence.

The idea that "experience" is an outdated concept would only make sense coming from an industry that really doesn't even really have "senior" people.

Having worked in multiple fields throughout my life, I deeply miss the presence of truly experienced coworkers.

For example one of the lessons of experience is that skill in the immediate task at hand is only a part of the overall engineering skillset. Experienced professionals (in any field) can often solve problems better with a tool they are unfamiliar with than less experienced professionals can in their area of expertise.

One trait of experienced devs I've seen, that I've not seen in even the sharpest but less experienced ones, is the ability to see problems that will arise down the road from miles away.

Likewise I also rarely see less experienced devs, no matter how smart and talented, see solutions outside of the main path massively reduce the complexity of the original problem. I fondly recall working in research orgs when I was much more junior, and asking a 30+ year industry vet how to solve a tricky problem only for them to, time and time again, that the best solution was to avoid the problem.

The trouble with climate change has always been that by the time you notice the "big nasty effect" it's already too late to prevent catastrophe.

To be clear, it's never too late to improve our fate, but by the time something like the AMOC shutting down destroys Europes climate, you've already signed up for an incredible amount of climate pain. Maybe such an event will wake us up enough to prevent out right extinction, but if this happened the collapse of industrial civilization would be virtually guaranteed.

> Where are we in the hype cycle on this?

Can we stop acting like the Gartner "hype cycle" is anything more than a marketing gimmick created Gartner to validate their own consulting/research services?

While you can absolutely find cases that map to the "hype cycle", there is nothing whatsoever to validate this model as remotely accurate or valid for describing technology trends.

Where is crypto in the "hype cycle"? It went through at least 3 rounds of "peak of inflated expectation" and I'm not confident it will ever reach a meaningful "plateau of productivity".

Did mobile ever have "inflated expectation"? Yes there was a lot of hype in the early days but those people hyped about it, rushing to build mobile versions of their websites... were correct.

The "hype cycle" is a neat idea but doesn't really map to reality in a way that makes it useful. It's only useful for Gartner to create an illusion of credibility and sell their services.

oh thanks for this! I had no idea, and it would have been better than nothing at all!

I was recently stuck on the tarmac for nearly 3 hours over night. It was an awful experience (especially since they were unable to give us a real estimate, we kept hearing 15-30 mins for the entire duration). I would have much rather they simply cancel the flight than have to go through that.

The issue is that all these airlines are running things with much lower tolerance for variance in the system than they used to. They just don't have the resources to support that any more. Rather than lose money by being honest about what they can handle, they'll let passengers suffer. The reason is obvious: no matter how awful my experience so far, I'll continue to keep flying.

I suspect air travel will get worse until the limit case that it's so awful people start refusing to fly altogether.

We currently have no reason to believe this, and information we do have seems to suggest that is very unlikely to be the case. I'm also guessing from my username you can infer that I don't think we even know enough to concretely say what is this "world" you are referencing.

LLMs and essentially all neural networks can be viewed as learning compression algorithms where the behavior of the compression algorithm is learned and subject to potential constraints beyond mere file reconstruction.

Highly recommend reading Ted Chiang's "ChatGPT Is a Blurry JPEG of the Web"[0] to get a better sense of this.

Keeping this fact in your mental model neural networks can also go a long way to demystify them.

0. https://www.newyorker.com/tech/annals-of-technology/chatgpt-...

The problem is funding.

The best psychology work tends to be much closer to philosophy than science. But philosophy generally doesn't get much funding.

Anyone interested in psychology will ultimately be forced to optimize for funding, which inevitably leads to fraudulent work since psychology based on scientific principles simply isn't possible yet as we don't know nearly enough about the function of the mind to explore the areas psychology is interested in.