I think this comparison isn’t quite correct. The downside with carpentry is that you only ever produce one of the thing you’re making. Factory woodwork can churn out multiple copies of the same thing in a way hand carpentry never can. There is a hard limit on output and output has a direct relationship to how much you sell.
Code isn’t really like that. Hand written code scales just like AI written code does. While some projects are limited by how fast code can be written it’s much more often things like gathering requirements that limits progress. And software is rarely a repeated, one and done thing. You iterate on the existing product. That never happens with furniture.
There could be factories manufacturing your own design, just one piece. It won't be economical, but can be done. But parts are still the same - chunks and boards of wood joined together by the same few methods. Maybe some other materials thrown into the mix.
With software it is similar: Different products use (mostly) the same building blocks, functions, libraries, drivers, frameworks, design patterns, ux patterns.
> If we consider the prompts and LLM inputs to be the new source code, I want to see some assurance we get the same results every time.
There’s a related issue that gives me deep concern: if LLMs are the new programming languages we don’t even own the compilers. They can be taken from us at any time.
New models come out constantly and over time companies will phase out older ones. These newer models will be better, sure, but their outputs will be different. And who knows what edge cases we’ll run into when being forced to upgrade models?
(and that’s putting aside what an enormous step back it would be to rent a compiler rather than own one for free)
> New models come out constantly and over time companies will phase out older ones. These newer models will be better, sure, but their outputs will be different.
IIUC, same model with same seed and other parameters is not guaranteed to produce the same output.
If anyone is imagining a future where your "source" git repo is just a bunch of highly detailed prompt files and "compilation" just needs an extra LLM code generator, they are signing up for disappointment.
> If you look at the rapid acceleration of progress
I don’t understand this perspective. There are numerous examples of technical progress that then stalls out. Just look at batteries for example. Or ones where advancements are too expensive for widespread use (e.g. why no one flies Concorde any more)
Why is previous progress a guaranteed indicator of future progress?
If AGI doesn't happen, then good. You get to keep working and playing and generally screwing off in the way that humans have for generations.
On the other hand if AGI happens, especially any time soon, you are exceptionally fucked along with me. The world changes very rapidly and there is no getting off Mr Bones wild ride.
>Why is previous progress a guaranteed indicator of future progress?
In this case, because nature already did it. We're not just inventing and testing something whole cloth. And we know there are still massive efficiencies to be gained.
For me the Concorde is an example of how people look at stuff incorrectly. In the past we had to send people places very quickly to do things. This was very expensive and inefficient. I don't need to get on a plane to have an effect just about anywhere else in the world now. The internet and digital mediums give me a presence at other locations that is very close to being there. We didn't need planes that fly at the speed of sound, we needed strings that communicate at the speed of light.
It is unfortunate that things like this become politically impossible because older people are one of the most reliable voting groups out there.
I will always be bitter that older voters chose Brexit by a large margin, in opposition to the younger voters who will actually be around to feel its long term effects. Not taking that into account in voting feels wrong but there’s no politically palatable way of addressing it.
Didn't the UK just start allowing 16 year olds to vote, which presumably helps offset the impact of older voters? I remember not getting around to voting in my first election (USA, Colorado). The outcome was George W. Bush being elected president, who favored policies not well-liked by younger people at the time.
This assumes parents would vote in the interests of future adults. In my experience, parents are quite happy to vote against future adults, even their own. Housing policy is the most obvious example.
Actually the US maintains a Senate and Electoral College because of slavery, and refuses to abolish them for (supposedly) any and every other reason. These systems allow whites in less populous states to exercise outsized power.
>It is unfortunate that things like this become politically impossible because older people are one of the most reliable voting groups out there.
They become politically impossible because they're not a front burner issue for anyone so the only people who are driving the issue are extremists who want the criteria set at like 10 whereas normal people want it at like 5 on some arbitrary scale of extremity so whenever it goes up for public consideration it gets shot down. You see this across all areas of mundane policy.
The movie was a satire, the novel was earnest. If you arent willing to sacrifice everything for democracy, then why should you have a voice? I am with Heinlein 100% here.
Having since read more about the author I'm pretty sure you're right the novel was earnest, but honestly it read as excellent satire when I didn't know it wasn't meant to be (and I read it prior to seeing the movie). Would recommend.
It's worth noting that in the book "service" is heavily implied to be primarily military in nature yet Heinlein purported years after the fact that in the book's canon "95 percent" of citizen service was actually civil. I think it's debatable whether or not this was his intention all along or a retcon to fit his more, ahem, liberal worldview that emerged as he aged.
I also loved Verhoeven's film adaptation but he straight up admitted that he didn't finish the book before making the film, which was itself based on a Neumeier (of "Robocop" fame) script called "Bug Hunt at Outpost 7" that bore only superficial resemblance to the book. He made the same mistake as many others in casting the book as fascist merely because of its militaristic elements when it's clearly not. On top of lacking many essential elements of fascism (a dictator, a directed economy, suppression of political dissent, etc) there are also several spots that veer into philosophical treatise to espouse the opposite. The flashback scene involving Rico's professor talking about how a society is obligated to raise its children correctly (and how it's society's failure if they end up delinquent) is a perfect example - "the system is the problem" hardly reads as far-right.
This is all to say that I think Heinlein was more interested in exploring a concept of reciprocal responsibility between a citizenry and its government. The militaristic aspects of the novel as regards a distant, dehumanized enemy and the dominance of the fight over all other aspects of life are far more alarming in my opinion.
I think people should be able to get up to 3 votes:
1. Veteran
2. Property ownership
3. Having children.
If you dont hit 1 of those criteria, you dont get a vote. You need skin in the game. Letting anyone vote is why “tax someone else, give me things” is such a popular platform. Politicians should have to hit maybe 2 out of 3.
Property ownership seems like a pretty transparent way to disenfranchise the poor. In what way does a renter not have “skin in the game” compared to a homeowner?
I am none of these. I'm in my late 50's and have been paying income tax since I was 16. Sure, rescind my voting rights ... I'd like all my 40+ years taxes back please then.
Having children? Why not consider instead: teacher, healthcare professional, municipal worker, civil engineer, volunteer ...and all of the many other roles that make society. Being a parent isn't the only indicator of caring for others.
You're right about military eligibility, but also that you shouldn't make calls for a nation which you will never see. Doubly so if one does not have children. No skin in the game, no alignment of incentives, no moral right to choose.
Even moreso when you consider basically the whole generation relies on leeching off the young and have continued to capture an ever-increasing proportion of public spending across the western world despite owning an outsized proportion of both real estate and wealth overall.
The book does address that, in that the federal service is universally available (and even the blind, deaf, or crippled would spend their time performing some job, even if it eas "counting the hairs on a caterpillar by feel".
Well, "capabilities" is carrying a lot of weight there. One of the main objectives is to design it for unrestricted modding to accommodate all of the wishlisted features, but "out of the box" the default game mode will be 1:1 in mechanics with some QoL improvements. The inspiration is mostly for designing systems in a way that can be easily reconfigured or extended to behave in other ways. We hope that by the time we reach feature parity, people will have already built some mods to do things that were impossible with Civ3.
As mentioned above this was started by Civ3 modders, and we all have our passionate reasons for preferring it over other entries, but you're not wrong that doing this with a 3D engine would be a whole `nother ballgame. There are actually Civ4 and Civ5 remakes underway which have both opted for 2D implementations.
Right, but they don't. Not to mention a significant portion of the target market are children whose brains are still developing.
Smoking is a vice. Anyone can stop smoking any time they want. But it was still incredibly popular. Government regulation put warning labels everywhere, tightened regulation to ensure no sales to children, provided support to quit. And then the number of people smoking plummeted. Society is better off for it.
"Anyone can do it" is an ideological perspective divorced from lived reality.
A clear difference here is that Apple creates the TV+ shows and they don't create the News+ content. And I really don't think they want to get into the news content creation business.
The problem is that people don't use it. I imagine it's a chicken/egg thing, the audience on News isn't big so it isn't worth the publishers time catering to an entirely new format, the News experience is crappy so the audience doesn't grow.
They could have insisted that everyone use their format but I suspect publishers would just refuse. It's not exactly in a publishers interest to help boost a middleman between their content and readers.
I'd be really interested to see what Apple's approach would be if they used more web technologies (since that's what publishers are using today anyway). Even just a webview with disabled JavaScript would get a ton of the way there in terms of performance. They have WebKit engineers in house that could probably tweak it even further.
This is actually the trajectory of both Apple News and iAd before it, which is what started out providing the ad service for Apple News. Apple would like to do a high quality solution, and then keeps relaxing their standards when there's not enough buy-in from the content providers. They were forced to allow the non-curated news formats to have sufficient content.
I wonder why they don't just prioritize the ~500 most popular of those content providers that are feeding them sludge articles, and write (AI-generate, even) logic to manually parse and transform said sludge into their format?
It'd be a big one-time lift; and of course there'd be annoying constant breakage to fix as sites update; but News.app could always fall back to rendering the original article URL if the News backend service's currently-deployed parser-transformer for a given site failed on the given article. It's make things no worse and often better than they are today.
I can't imagine it's a great deal for publishers. It's probably why NYT, Economist and other prestige publications aren't on it. (Save for Atlantic, New Yorker). I. Assuming they use the Spotify model ( paying commissions on articles per reader)?
The 10000ft perspective on AMP was correct, the lived reality was awful. And the technical implementation used can't be divorced from everything that surrounded it: Google's place in the industry with regard to search engines, ads, etc.
In this specific example there is a very big difference between producing a format for use in a first-party app vs trying to replace standards for content used across the web.
> And the technical implementation used can't be divorced from everything that surrounded it: Google's place in the industry with regard to search engines, ads, etc.
I mean... sure it could have? There could have been an independent "AMP Foundation" that forked the standard away from Google and owned the evolution of it from then on. Like how SPDY was forked away from Google ownership into HTTP2.
Are we remembering the same Factbook? It had summary statistics for every country and some brief blurbs about their history, climate, economy, etc. Strictly speaking yeah it generated some legitimacy to publish a resource like this and I find it hard to believe the CIA can't scrape a few quarters together to keep it running, but most of it's value is sentimental.
Soft power includes positive perception. Every time someone learns that GPS is completely paid for by the American government and then freely available to the rest of the world, that shapes perception.
The Facebook being quoted by so many school kids worldwide was a cheap softening of how the world perceived the CIA and America. Now how valuable that is isn’t clear, but when something is that cheap it doesn’t take much to be a net gain.
I had something similar to this talking globe[1] when I was a kid and it was amazing for raising my geopolitical awareness. You tap on a country with the pen and it tells you the name and some facts about it. Even if I hadn't learned anything, I had fun pressing "Azerbaijan" over and over because 10-year-old me thought it was a funny spelling and pronunciation.
You completely misunderstood what everyone was talking about. The point is to make people in other countries do what we want them to do.
American diplomacy, foreign policy, spying, soft, and hard power etc is obviously primarily targeting non Americans here.
Thus like most things the CIA does this is targeting foreigners or foreign influence, though of course direct impact on Americans is a nice bonus. We don’t want young Americans looking up facts on a Chinese or Russian website.
You might be underestimating the reach, you've got schoolchildren around the world using it as it's usually the most convenient source you're allowed to cite for this data
As an anecdote example, I've never ever accessed said Factbook, but I've heard about it enough times to remember that such thing exists and that USA govt. is collecting a relatively objective fact list. So yeah, it was a tiny bit of soft power of sorts. It showed that USA cares about outside world, in some way at least.
PS: and I live in Eastern Europe, far far away from the USA.
I grew up outside the US. I have a distinct memory of using the Factbook for homework assignments and being told it is a reliable source of information. That shapes people's perceptions of the US and the CIA from a young age.
Or maybe a conscious decision, as neoconservative Robert Kagan writes:
"President Trump has managed in just one year to destroy the American order that was and has weakened America's ability to protect its interests in the world that will be. Americans thought defending the liberal world order was too expensive. Wait until they start paying for what comes next,"
The administration is dispensing with the institutions of soft power. I don't think it's the main goal so much as a consequence of their worldview. Soft power is essentially worthless to people who have no interest in maintaining a facade of international cooperation.
And 'soft power'? Like lying about stats and using it for propaganda? Otherwise its just objective and someone else can do the work. For some reason I never attributed it to the US or CIA.
Under the current administration it wouldn't surprise me if they decided in their last budget cutting meeting to indiscriminately erase everything with the wildcard "fact" in the project's name.
“Soft power” refers usually to credibility. The point of the Factbook is to be a credible public resource for an entity that would otherwise not have much.
Credibility is not what soft power means, though they are related. Power is the ability to get other people to act in your interest. Hard power is when that is done through immediate, direct economic or military coercion. Soft power is everything else.
Credibility is the core currency of soft power, whether one views its ultimate goal as manufacturing consent or fostering genuine cultural attraction. Without that perceived reliability, the indicator "soft" loses it's meaning.
>Credibility is the core currency of soft power, whether one views its ultimate goal as manufacturing consent or fostering genuine cultural attraction.
Not sure its worth dissecting this, but there is a lot of grey area in your claim of the meaning of Credibility. (Credibility and cultural attraction? Pretty sure these have little correlation. Dictators can make creditable threats.) Further, its a debatable claim that there is a 'core currency' of soft power.
As a contextualist, I am not going to die on this hill for your personal meaning of Credibility. But I can attest that your conviction in your claim is stronger than any International Relations Realist practitioner would make.
You can make propaganda without lying, by choosing what metrics you value over others for example, by adding them or omitting them or implying whether a stat increasing is positive or negative.
Also choosing which methodology is the "right" one to measure a specific number.
There are lots of ways to measure ethnic groups, the size of the capital or the unemployment rate. If you publish the numbers you get to choose which one suits you best, you just have to be globally consistent
Having friends means that you can build bases where if you ask nicely, rather than having to invade. It prevents those friends from undermining you in a lot of cases. It makes them help you when you need, e.g. to get your hands on someone plotting attacks against you. It makes them more likely to trade with you under advantageous terms. I am sure you could think about at least a dozen other cases in a couple of minutes.
Soft power is spending pennies to convince other countries to do your dirty work.
I think it's instructive to compare the U.S. and Soviet stances in Europe after WW2. To maintain a military presence in Eastern Europe, the Soviets had to rely on repression, coercion, and occupation. This was expensive and fragile and eventually fell apart. The U.S. was openly welcomed into Germany and other countries in Western Europe. This was the value of "soft power."
Among the countries that host US bases, how many had to accept it under the threat of force, invasion, or occupation? I would guess Japan and Germany (initially). Look at the map here: https://commons.wikimedia.org/wiki/File:Foreign_bases_2.png . Brute force was not a facto in the vast majority of them.
Some people mentioned the dollar as the global reserve currency, but there's also the use of English as the global lingua franca, the US being the largest global destination for talent and investment, and countries (previous) willingness to make sacrifices or deal with the US on less-than-perfect terms out of a sense of shared culture.
Some people really do think of soft power, propaganda, shady covert operations, etc. as something "the other guys" do (China! KGB-Putin!), but assume the US is somehow above all that.
Basically a neoconservative-esque sentimental view of the USA as "the good guys" on "the global stage" (although many would rightly recoil at the comparison to neocons).
Pretty sure OPs point is that Tesla has shitty results, has a CEO that lives “any publicity is good publicity” as a mantra and the company valuation is through the roof.
They're good things right now for Elon and Tesla. Whether they will continue to be is up in the air. But, it definitely works as a strategy, at least temporarily.
Code isn’t really like that. Hand written code scales just like AI written code does. While some projects are limited by how fast code can be written it’s much more often things like gathering requirements that limits progress. And software is rarely a repeated, one and done thing. You iterate on the existing product. That never happens with furniture.
reply