almostdeadguy

> I believe that Silicon Valley possesses plenty of virtues. To start, it is the most meritocratic part of America. Tech is so open towards immigrants that it has driven populists into a froth of rage. It remains male-heavy and practices plenty of gatekeeping. But San Francisco better embodies an ethos of openness relative to the rest of the country. Industries on the east coast — finance, media, universities, policy — tend to more carefully weigh name and pedigree.

I believe I read that 27% of the founders in the YC Spring 25 class went to an Ivy League school and 40% previously worked at a magnificent 7 company. I'm not saying this is any worse than the east coast, but so much for name and pedigree not mattering.

Northern California is what it always has been: the barrier wall of manifest destiny, where instead of crossing the ocean the pioneers and all subsequent generations stayed to incubate the same incentives, and have been relentlessly in pursuit of the next gold rush. Gold, yellow journalism, semiconductors, personal computing, SaaS, crypto, AI, etc. It's the sink drain attractor of people looking to improve their fortunes in one way or another, but almost always around some kind of bonanza of concentrated opportunity. The concept of it being "meritocratic" is a rephrasing of ideology that's always existed about the region: you too could get rich here. But I don't really see any difference in the networks of power that exist in SV as do the rest of the country.

I grew up in the bay area and am far happier living outside it. I'm happier to be in a place where art and the humanities are valued instead of cast aside as immaterial or silly or a distraction. I'm happier to live in a place where people have varied interests instead of orienting their life around whatever the prevailing Big Thing is.

> So the 20-year-olds who accompanied Mr. Musk into the Department of Government Efficiency did not, I would say, distinguish themselves with their judiciousness. The Bay Area has all sorts of autistic tendencies. Though Silicon Valley values the ability to move fast, the rest of society has paid more attention to instances in which tech wants to break things. It is not surprising that hardcore contingents on both the left and the right have developed hostility to most everything that emerges from Silicon Valley.

I see some positive aspects as to more inclusive definitions of autism and neurodivergence, but I hate that we're at the point where "trying to get rich at all costs" is now perceived as autistic (and let's be clear: using mobile gas turbines that get people sick to generate power for AI is not "autistic"). Greed is not autistic, but of course the ideology of SV is that nobody actually cares about money there. Why else would they have apartments without furniture and piles of pizza boxes. It must be the autism.

> While critics of AI cite the spread of slop and rising power bills, AI’s architects are more focused on its potential to produce surging job losses. Anthropic chief Dario Amodei takes pains to point out that AI could push the unemployment rate to 20 percent by eviscerating white-collar work. I wonder whether this message is helping to endear his product to the public.

The animating concern of developing AI since 2015 has basically been "MAD" applied to the technology. The Bostrom book mentioned later in this article was clearly instrumental in creating this language to think about AI, as you can see many tech CEOs began getting "concerned" about AI around this time, prior to many of the big developments in AI like transformers. One of the seminal emails of OpenAI between Musk and Altman talks about starting a "Manhattan Project for AI". This was a useful concept to graft the development of these companies onto:

1. Firstly, it's a threat to investors. Get in on the ground floor or you will get left behind. We are building tomorrow's winners and losers and there are a lot of losers in the future.

2. Secondly, it leads to a natural source of government support. This is a national security concern. Fund this, guarantee the success of this, or America will lose.

On both counts, this framing seems to be working pretty well.

I mean I suppose you can continuously add "critical feedback" to the system prompt to have some measure of impact on future decision-making, but at some point you're going to run out of space and ultimately I do not find this works with the same level of reliability as giving a live person feedback.

Perhaps an unstated and important takeaway here is that junior developers should not be permitted to use an LLMs for the same reason they should not hire people: they have not demonstrated enough skill mastery and judgement to be trusted with the decision to outsource their labor. Delegating to a vendor is a decision made by high-level stakeholders, with the ability to monitor the vendor performance, and replace the vendor with alternatives if that performance is unsatisfactory. Allowing junior developers to use LLM is allowing them to delegate responsibility without any visibility or ability to set boundaries on what can be delegated. Also important: you cannot delegate personal growth, and by permitting junior engineers to use an LLM that is what you are trying to do.

I think it’s a fair point that google has more stakeholders with a serious investment in some flubbed AI generated code not tanking their share value, but I’m not sure the rest of it is all that different from what engineer at $SOME_STARTUP does after the first ~8monthes the company is around. Maybe some folks throwing shit at a wall to find PMF are really getting a lot out of this, but most of us are maintaining and augmenting something we don’t want to break.

This going from the front page to effectively buried (cant even find it in the first 10 pages) within minutes is not a great look for HN.

Why are we pretending like his comments about any of these things are as a neutral observer instead of as an investor cheerleading his investments? Why should anyone take Gary Tan seriously as a futurist?

Most of the money flowing to the big players is from tech giant capex, originally from net cash flow and lately its financed by debt. A lot of these investors seem to now essentially be making the case that AI is "too big to fail". This doesn't at all resemble VC firms taking a lot of small bets across a sector.

I love TwixT, I discovered the Bookshelf Games version of this recently (a friend of mine's parents had an old copy at their beach house). Also a big fan of Alex Randolph's Ricochet Robots.

Dieter Stein's games are also supposed to be wonderful abstracts (Urbino, Fendo, Tintas) though I haven't had a chance to play those yet.

Actually what he said was "The MAGA Gang is desperately trying to characterize this kid who murdered Charlie Kirk as anything other than one of them and doing everything they can to score political points from it." and it's been taken for granted that what he meant is that Tyler Robinson is MAGA, but that's not strictly what he said.

Waymo would not fight against "people getting rid of their cars", many people in NY who use the incredible public transit system would like to see more car-free streets, which they absolutely would fight against.

The ideology of SV will only get more dangerous the more government contracting becomes ingratiated into the business models there. Extremely scary in our world of Palantir, Anduril, NSO group, etc.

A funny account of spending time with him by Rick Perlstein, who wrote an amazing series of books on the history of conservatism: https://prospect.org/power/2024-04-24-my-dinner-with-andrees...

All the #[derive(Clone)] does is generate a trait impl of Clone for the struct, which itself can be bounded by trait constraints. It doesn't have to know that every use of the struct ensures generic parameters have to/don't have to be Clone. It doesn't have to make guarantees about how the struct is used at all.

It only needs to provide constraints that must hold for it to call clone() on each field of the struct (i.e. the constraints that must hold for the generated implementation of the fn clone(&self) method to be valid, which might not hold for all T, in which case a Struct<T> will not implement Clone). The issue this post discusses exists because there are structs like Arc<T> that are cloneable despite T not being Clone itself [1]. In a case like that it may not be desirable to put a T: Clone constraint on the trait impl, because that unnecessarily limits T where Struct<T>: Clone.

[1]: https://doc.rust-lang.org/std/sync/struct.Arc.html#impl-Clon...

I love the implicit and totally baseless questioning of motives in this reply.

Very funny to re-title this to something less critical.

> I don't think "acts of war" mean much in real statecraft; there's no referee, things are what people say they are and outcomes are determined by military and economic power. But anyone going down that path has to recognize the hole Iran dug for itself here. They didn't have to do any of this.

I'll preface this by saying I understand fixating on a small detail, and as evidenced by this thread, I readily engage in that. I also think the thing about the fatwa against nuclear weapons is a little silly, it seems like there are incredibly obvious, rational reasons for Iran to want a nuclear arsenal.

To back up to Oct 2024 (to make a different point, I'm not trying to take us further down the rabbit hole), I think it's worth pointing out how arbitrary it is to choose this moment in time as a point where Iran "dug a hole for itself". Presumably a moment of intervening agency that breaks from what came before and after. It's a vantage point that has no real significance to the broader conflict, doesn't tie to the beginning or end of anything significant. It's unclear why that moment in particular is where Iran could have set us on a different course, and why we should consider jettisoning the rest of historical baggage that lead up to that moment. And it has the whiff of being chosen arbitrarily to exculpate (or sideline any notion of) the United State's involvement in this conflict. It's the kind of detail I expect to see fixated on CNN, without any mention of events like Israel's former invasion of Lebanon, the impact of the Nakba and the One Million Plan on the surrounding Arab states, the Dulles brother's lead coup in Iran that deposed a secular, democratic leader, etc. Not that you even really need to go back that far, there's plenty of events proximate to 10/1/24 that lead to Iran launching missiles, like Israel staging a land invasion outside Lebanon.

I don't think that's what you're doing (I'd rather not speculate on why that moment is significant to you, and would be curious to hear your own take), but I want to explain why discourse like this becomes touchy. For some of us millennials, our defining political experience was seeing the United States become an incredibly sore loser via a problem of our own making (and infuriatingly, we apparently learned nothing from the consequences of funding the Mujahideen). We are obviously, also a victim of our own circumstances, no less than Iran, but we are also an agent of incomparable power in world events. And for many of us it became clear how carelessly, callously, and selfishly that power is wielded and how quickly we victimized ourselves and were unwilling to tolerate criticism. Aaron Sorkin wasn't even able to make the movie that depicted how much of a problem of our own making this was: https://en.wikipedia.org/wiki/Charlie_Wilson%27s_War_(film)#...

All this to say, if you feel ganged up on, I get it and sympathize. I think you have good intentions here, and I'm sorry if I came in too hot. Some of us are just nauseated by our position in the world and culpability for all this harm, and are constantly frustrated by the hegemonic political discourse that is adamant to deflect criticism and prioritize American exceptionalism above all.

I think this is all common knowledge? A brigadier general of the IRGC was in the room with Nasrallah when he was killed IIRC. I might quibble with "militia", which has legal ramifications I'm sure you're aware of. Maybe there are conspiratorial cranks who deny that's what Hezbollah is, but I'm not sure I know any people who would disagree with this. I believe Khamenei called Hezbollah and Nasrallah "his children" with pretty clear connotations being conveyed after Nasrallah was assassinated.

I'm not really sure what point you're trying to make here. Do you think it's not public knowledge that Iran finances and orchestrates Hezbollah? Or is this the terrorism card? Maybe it's different when the US supports al-Zenki et al (or the many years of financial support and hands-off permission Israel gave Hamas and then encouraged Qatar to give Hamas).

I think Hezbollah has done some truly disgusting things and I have no delusions about the ill deeds of Iran's regime. But I think it's extremely fair to say that Israel is an even more egregious perpetrator of murder, displacement, and apartheid and Iranians have every right and reason to see themselves in solidarity with Palestinians and the people of Lebanon.

After Israel assassinated Haniyeh in July and launched an air strike on Beirut that killed 30+ people, some of which were civilians. Keep going on with this game though.

> Whether this understanding of engineering, which is correct for some projects, is correct for engineering as a whole is questionable. Very few programs ever reach the point that they are heavily used and long-lived. Almost everything has few users, or is short-lived, or both. Let’s not extrapolate from the experiences of engineers who only take jobs maintaining large existing products to the entire industry.

I see this kind of retort more and more and I'm increasingly puzzled by it. What is the sector of software engineering where we don't care if the thing you create works or that it may do something harmful? This feels like an incoherent generalization of startup logic about creating quick/throwaway code to release early. Building something that doesn't work or building it without caring about the extent to which it might harm our users is not something engineers (or users) want. I don't see any scenario in which we'd not want to carefully scrutinize software created by an agent.

I don’t disagree with this at all, psychological safety is important and abusive jerks don’t belong on any team. It’s also important to try to understand if someone is being intentionally rude and callous or simply being blunt/potentially being misunderstood.

Being kind and knowing how to deliver disagreement in a constructive way is important. Human beings aren’t perfect logical subjects immune from emotion (thankfully). Hopefully a good manager can take steps to address this or help mediate conversations where such a person might struggle. But places where “playing the game” (like withholding valid concerns with a decision) is a criteria for success are not long for this world

I regard this the same as the crypto/NFT people saying hate the practitioners not the technology. The only things people seem to do with the technology is to create more and more trash.

Or alternatively it's being used to grift the 1% of instagram users willing to buy vaporware products: https://x.com/KrangTNelson/status/1928446282323894780

I seriously doubt most of the money being made on this stuff is through consumption. There's a big pot of money by tech giants being used to commission and promote this kind of work to attract a user base.

Sturgeon's law is generous to the output of LLMs.

The "just a tool like any other" perspective can't account for how much talentless slop is in practice created by AI.

Look at the work spotlighted by Google to promote their new Flow tool: https://x.com/GoogleLabs/status/1925596282661327073 . This is a garbage imitation of a Guy Ritchie film or a Jose Cuervo ad (maybe more the latter).

Instead of being a tool for creatives it has empowered a number of grifters to churn out more and more "content" bypassing any concern about craft and formal restrictions that help generate creative work. The work that is most often created with the help of AI is not creative, it's a bland, tasteless simulacrum of creativity.

I think most people using these frameworks are aware that DI is just automated instantiation. If your program has a limited number of ways of composing instantiations, it may not be useful to you. The amount of ceremony reduced may not be worth the overhead.

I meant "singleton" in the sense of a single value for a type shared by anything that requires one, i.e. a Guice singleton ( https://github.com/google/guice/wiki/scopes#singleton ) not a value in global scope. Or maybe a single value by type with an annotation, the salient point is that there are values in a program that must be shared for correctness. Parameterless constructors prohibit you from using these (unless you have global variables).

Calling these "singletons" maybe created confusion here, I'm more talking about singleton in the sense of a value that is created once and shared (i.e. injecting a value as a "Singleton" via something like Guice). I'm not arguing that you need to have values in global scope, I'm arguing that parameterless constructors prevent you from using shared values (actually _unless_ you have singletons in global scope).

It often is not enough. Singletons frequently need to be provided to things (a connection pool, etc.).