Those supermen... when I first wake up I crawl to the coffee pot.
Forget the grinds and it’s kind of weak.
Forget the water and nothing comes out.
Forget to turn it on and nothing happens at all.
Switching to a single serve coffee maker adds new failure methods. Forget the cup and it brews onto the floor.
You fill it in the evening while you're still coherent. It starts on its own 10 minutes before the alarm in the morning.
Hype happens when people don't really need a product and the marketing machine has to compensate. It helps if the product seems like magic for the first month and then becomes obnoxious and boring without marketing.
When these came in, I remember my dentist being clearly very excited. There was absolutely hype... we just weren't the target for it.
There was a real rage in the press and in the street after it was released. Now it can still be bought, but most people would consider it uninteresting.
What we see now with "AI" is that the hype is mostly by online shills with some connection to the funding. The traditional press as well as the alternative YouTube press is increasingly negative.
But the granddaddy for your question would be the Polio vax.
I love to "folks these days" so I shall. Folks these days don't know what a revolution Firefox was. Before that there were no "web standards" or anything. You had IE6 compatible and Netscape compatible and this and that. Linux web browsers would struggle to render pages. They came up with these ACID tests and these standards.
And Firefox was crazy back then. People now would lose their shit over the "lack of consent". People were talking about how they'd install Firefox and give it the IE logo so their parents would use it and everyone would talk up its leanness and this and that.
You couldn't get Netflix on Linux. It needed Silverlight to enforce DRM. These days the web is the platform. The complexity in the runtime and every application runs on Linux. Man, I never could have guessed 20 years ago that this would happen.
For me the fastest I've seen mind-blowing technology go from sci-fi to banal is LLMs. When TalkToTransformer first allowed you to do it there was this completion model that seemed insane. Like magic. Now, everyone I know uses an LLM for all sorts of things. It's so integrated into life. Wild.
I always interpreted those sort of stories as kind of… somebody doing something they consider at least a little duplicitous.
At the time though, there was the concept of a “family computer,” which isn’t really as much a thing anymore. Also it was assumed (whether or not it was true) that kids knew tech better than their parents, so there could at least be a kind of justification. “I’m modifying the shared resource that I know best in a slightly underhanded way so that my parents will use it better.” I dunno. I didn’t do that sort of thing but I was an annoyingly conscientious kid.
> You couldn't get Netflix on Linux. It needed Silverlight to enforce DRM. These days the web is the platform. The complexity in the runtime and every application runs on Linux. Man, I never could have guessed 20 years ago that this would happen.
Was 4K Netflix ever finally fixed on Linux? IIRC it was limited to lower resolutions for the longest time.
Often it's a situation where someone doesn't really know what programs are, they just know that the blue E is the button for Internet. In that case it's not trickery to keep using the blue E on the Internet button, just practical labeling.
Could have fun by looking at any social spread. Not just of a technology through the society. Fashion similarly spreads. Behavioral patterns spread. Language usage. All of it.
The point of "hype", on the other hand, and the reason to be "hype-optimistic" as the OP is, is that it can result in the spread of technologies that are not necessarily better. In other words, tech hype or marketing will not necessarily spread technologies in a meritocratic way. But actually what many cultural anthropologists know, such as by studying the spread of early human tools, is that there is a wisdom of the crowd effect where large social adoption of suboptimal technology tends to improve that technology unintentionally.
Those who decry "hype" are operating from an engineering background, where good tools are designed from the top down. But "hype" is more like a social force, which can emergently produce better tools from the bottom-up. The latter dynamic is robust yet uncontrollable, but it's definitely a real thing.
I'm seeing similar with so many people hyping solar power. People are fawning at how useful today's panels are. And they really are. But, that is largely only true if you have all of the other technology that we have to use the power it can generate. And this cuts in multiple amusing ways. Without electric lights, a lot of the power you'd have at home would be wasted. And, without modern electric lights, far more of it is lost in non-light generation than what we see today. (Lights are particularly interesting to look at. The amount of energy used by lights had such an extreme drop off that it is almost hard to really grapple with.)
Directly to your post, fashion and language may seem more subjective, but I question that. Obviously parts of it are driven less by utility and more from some other factors. But that is almost certainly true for technology, as well. Is why many places hold out from technology for longer than others, as an easy example.
* https://en.wikipedia.org/wiki/Technological_Revolutions_and_...
And The Rise and Fall of American Growth by Gordon:
* https://en.wikipedia.org/wiki/The_Rise_and_Fall_of_American_...
Most actually innovation (in the US?) occurred from about 1920-1950,† with mainstream acceptance from 1950-1980, with very little newness (leading to economic/productivity growth) occurring post 1980s: the main exception of course being computers and the Internet, but that seems to have maxed out in the '00s.
† The (first?) Industrial Revolution is discussed, as well as early scientific discoveries that led to later engineering development that turned into usable products.
It’s hard for me to imagine a KPIs that hasn’t been affected by mainstream internet and mobile smartphone adoption, which happened the last 15-20 years.
The responses are very cultural and nothing like what would be required to accomplish anything large.
To give an example. We could make tubes with hot and/or cold water running from Siberia to the Sahara. There are plenty of design challenges in such project. The correct mindset is to take a broad outline calculating ROI so that you have an idea how much the state of the art needs to improve for it to be viable. Then you take the largest expenses or technical hurdles and ponder what could be done to ease the pain point. You can ponder the diameter vs the temperature vs the flow rate or pressure, the distance between the pumps etc etc, not to build anything but to train the brain noodle.
If it wasn't a stupid idea people would be doing it already.
Many countries, if not all, had a period where great things were done. After that comes defeatism nihilism and a preference for talking about tarifs, Epstein, 911, etc
With tiny improvements everything shifts around. The steam engine had many challenges that have been solved long ago. Imagine that the 100 mph carburator didn't happen because you lose much of the throttle control. How silly this excuse is in the age of hybrids? It's pretty much a non-argument.
* reduce infant or child mortality?
* reduce maternal mortality?
* increase average life spans (by, say, double-digit percentages)?
* increase health or quality of life such that people at the end of their lives are more mobile, and perhaps have less onerous medical conditions?
* produce more food in the same amount of land?
* produce more food with fewer inputs (energy, labour, fertilizer, water)?
* produce more manufactured things with fewer inputs (energy, labour, metals)?
* allow for faster transportation with more efficiency?
It's all very well to invent something, but in what ways does it improve people's lives:
> Fortunately Gordon indicates that it is not an end to innovation, but a decline in the usefulness of future inventions that is taking place. Documenting the impressive rise in standards of living between 1920 and 1970, with rising TFP, he claimed that it will be more difficult than before to replicate such improvements in advanced countries like the United States. In the earlier period, the American standard of living doubled every 35 years; in the future this doubling (for most people) may take a century or more. Moreover, the newer innovations do not seem to be benefitting all segments of society, which in turn reflects rising inequality in the advanced countries.
> Gordon’s thesis raises many questions: Is he correct in saying that the era of rising labor productivity associated with innovation and technological change over? Will the digital economy imply a new rise in the standard of living? From a development perspective, what implications does his thesis have for middle- and low-income countries?
* https://blogs.worldbank.org/en/developmenttalk/declining-pac...
Well yes, they're aiming for those goals.
On the grandparent comment's link, you'll notice that they mention striving for the UN's sustainabile development goals, which includes all of that (+ others like gender equality).
Or on this link about theme weeks[1] which includes more of the same.
And everyone wants to create a longer-lasting light bulb, or a beer that tastes great and is less filling, but what are the odds of that actually happening?
LED bulbs are great, but how much of an improvement are they over CFL bulbs? Both are better than incandescent, but electrical light was a quantum leap over candles and gas/oil lighting. See absolute plumeting since 1800s, and electric light from 1900 to 1950:
* https://ourworldindata.org/light-at-night
In the last ~50 years things have improved, but the differences are tiny compared to past gains: and that's Gordon's thesis.
The Internet got going in the 1990s, and everyone in the US was basically connected by (say) 2010. Can you be "more connected" that already having your business or home being online 24/7 (i.e., DSL or cable, not dial-up modem)? Certain transfer speeds have gone up, and going from a V.92 56kbps to DSL is improves things, but how much productive is going from 10 Mbps to 50Mbps: do you get a 5x rise in economic productivity?
Can you point to a FRED chart or cite a DOI that shows that GDP is better in societies pre- and post-smartphone?
If all KPIs have been affected, then it should be easy to post a graph and see where there is an inflection point or change in the slope.
If you find it so hard to believe, shouldn't it be easy for you to cite the data showing so?
Diminishing returns has kicked in for me. But hard to rule out any future bursts until the baseline raises. Perhaps lower latency could make things like VR and AR more appealing and useful.
In 1900 people were mostly in horse and buggies (and trains for longer distance), and by 1960 we had the Boeing 707, with the 747 entering service in 1970: how much have been moved forward from that?
The business cycle would likely exist in the absence of emotional choices. Part of it is that businesses that are near-failure tend to fail when the economy declines, which causes additional decline. This means that near-failure tends to build up during good times and then fail en masse during a decline.
Most human behavior, especially commercial and economic, is a function of one's own received beliefs—and one's observations about and participation in the beliefs of others.
Whether a given assertion is factual is secondary to whether it alters belief.
Though I suppose even "belief" is too strong a word. Let's say "behavior."
One need to not believe the hype, to seek to extract value from a market.
The question I'd look it is, how are beliefs and behavior shaped by various assertions, and how well do those assertions correspond to various observations and predictions about the nature of reality—including collective behaviors.
The funny thing about ponzy schemes and hysterical markets is that the usual account about how they form and how they collapse, hand waves around the core of the whole question: the source of value.
A good socialist materialist might attempt to link value to utility amid other prosaic and pragmatic qualities,
but our nature as a social species—one motivated in increasing part as we shuffle down the hierarchy of needs by intangibles such as status and its signaling—means that even the concept of utility is just another contentious ambiguity.
Nearby is the concept of novelty: fashion in the clothes sense obviously chases a convoluted but ultimately cyclical path through relatively limited spaces.
Many of us who have been on HN or in the industry prior to its inception have seen the pendulum swing in technology fashions as well—a regular example being the architectural decisions made in front-end frameworks.
The fiction of classical philosophy of science was that it was additive, and that information was well preserved over time, and hence the entire process of scientific progress one of, well, progress.
The equivalent fiction here might be that there is some obvious way in which the over-hyped and the real, usually, are determined not by intrinsic merit but rather through the tidal swinging of fashion.
Usually. Every once in a while actual progress is made, thank god.
ML and AI are obviously one of those cases.
The noise of the industry churn and frenzy don't obscure that much, at all.
1270018080•7h ago