Hacker Newsnew | past | comments | ask | show | jobs | submit | reactordev's commentslogin

The big bang time relativity problem sometimes makes your brain hurt but this is amazing!

I’m so fascinated by the fact that we can look back through time by looking at these distant objects. I wish I went into astrophysics instead of engineering…


I went into astrophysics and came out very discouraged. The researchers actually pushing the envelope are 1% of academia and if you don't find a department with them, you are paddling in the open sea. There is an incredible amount of cruft in academia, not to mention how financially insecure that life is.

Truly, only those who think about nothing but (astro)physics can bear it.

I still love thinking about fundamental problems and upcoming research however. That will never be gone.


I realize my choice was definitely financially driven but in a future where that’s easier with AI, I’d like to focus on things that make my brain tingle.

I used to love engineering but with AI I feel like all the passion (learning things, making brain squeeze) is gone and I’m just managing another resource.

Don’t get me wrong, I like building things. I also like solving challenges and hard problems and I haven’t done that in a few years now.


what do the other 99% of researchers do?

Most research is boring incremental stuff, and very often you will find a dejected or disappointed individual that realizes this. The invention of relativity only made one scientist a household name. I guess everyone else that came before and after were doing nothing at all.

Everyone I know who studied astrophysics ended up in Fintech doing data science anyway. "illusion of choice"

I met a few in MarTech as well…

As if you needed another reason to switch to Linux

It means that you have no business being below FL180 or 18,000ft to enter this airspace.

Coming from groups that just pickup busses of people to murder, I wouldn’t be so sure that firing back at the US would be out of the question.

Murdering buses of people doesn't bring the full force of the US military on them. The difference is the risk not the depravity.

This is the answer. The cartels would have to be insane to poke that particular bear. They would get crushed like a bug. IIRC they murdered a single US undercover officer in the 90s and the retaliation was so bad that they themselves handed over the perpetrators.

> They would get crushed like a bug.

Much as I despise them, I'm not so sure that would be the case. I seem to remember folks saying the same about the Taliban, and the cartels have a lot more money and high-tech kit, than the Taliban.

Asymmetric warfare is a tough gig, on all sides.


I don’t think the technology matters nearly as much as the asymmetry. Iraq had better technology than the Taliban and their military didn’t last a week.

True enough, but the cartels are also experts at running what is basically guerrilla warfare, against each other. Not sure if the Mexican Army has ever tried to take them on. A lot of cartel soldiers come from the army.

You are right rationality is their strongest character trait.

How are they not rational? Violence is a tool. They operate an illegal business so they can’t sue other parties for breach of contract. They can't call the police if they are robbed or file an insurance claim for what was taken. Even the over-the-top violence has a rationale. They aren't punishing the victims as much as they are attempting to broadcast that there is a higher price to be paid than any gain from giving information, to reduce their future losses and enforcement efforts. It isn’t moral or ethical, but I wouldn’t say it is irrational.

Lots of organized crime around the world manages to operate without cutting all the limbs off somebody then arranging them like flowers in a "vase" made out of the poor soul's ribcage. The cartels take violence far beyond what is pragmatically necessary. Their system of crime breeds excessive violence and insanity.

This stuff mostly followed after the zetas. It was a very deliberate strategy to compete in a hostile landscape that others eventually copied to survive.

Marketing, if you don't know the answer it's always marketing

> How are they not rational?

It's the meth.


The cartels are incredibly rational - what they lack are morals and ethics

It's a business not an ideology.

Do you have much evidence of them behaving irrationally?

I would recommend reading the Freakinomics book or listen to their podcasts on drugs.

TL;DR: drug cartels are run like businesses. They are very rational. But, unlike your boss, their boss can also shoot you in the face if you annoy them too much


How did that full force of the US military work out in Vietnam?

Millions of dead Vietnamese.

In any case that was a war against a hardened, experienced, determined enemy fighting for its freedom from any form of colonial occupation, both as a formal military and as an insurgent force in South Vietnam.

I scarcely think the Mexican population would rise up in defense of the cartels here.


I think a lot of people would be cheering on the destruction of the cartels.

They'd probably quickly stop cheering as their own homes and families were destroyed as collateral damage, which is what would happen if the "full force of the US military" were deployed against the cartels.

The last time America invaded Mexico City it created martyrs. It's a fascinating story that they do not teach at US highschools lol.

The destruction of cartels would involve careful policing and corruption controls, the best American administrations have been bad at this. The worst... can barely put its pants on much less dismantle foreign organized crime. You can't shoot a missile at a cartel and poof it's just gone.

A non-aligned population will look out for their own interests and are aware that the attention of the US is temporary but the cuadillismo that lead to cartels are a durable cultural artifact.

  The Battle of Culiacán, also known locally as the Culiacanazo and Black 
  Thursday, was a failed attempt to capture Ovidio Guzmán López, son of Sinaloa 
  Cartel kingpin Joaquín "El Chapo" Guzmán, who was wanted in the United States 
  for drug trafficking.
  
  Around 700 cartel gunmen began to attack civilian, government and military 
  targets around the city, despite orders from Ovidio sent at security forces' 
  request. Massive towers of smoke could be seen rising from burning cars and 
  vehicles. The cartels were well-equipped, with improvised armored vehicles, 
  bulletproof vests, .50 caliber (12.7 mm) rifles, rocket launchers, grenade 
  launchers and heavy machine guns.
https://en.wikipedia.org/wiki/Battle_of_Culiac%C3%A1n

It was never used, there.

Pretty badly for both sides

I don't really think you thought through that one. It sounds like what your saying is that the Vietnamese won and thats the outcome that matters. It does matter but that isn't the issue - it is the cost that everyone is talking about: the amount of destruction that was brought upon the country and people was terrible.

The distinction is those are cases where they are murdering Mexican citizens. If a cartel murdered a bus of people in America I suspect most any administration would retaliate in some form.

Dude, Americans are getting kidnapped and murdered in Mexico all the time. The cartels don’t care your nationality.

If the administration strikes cartels first, they may find it egregious enough to do what they refused to do in the past…

I don’t rule out any options when it comes to murderous organizations.

*EDIT* This isn’t me saying don’t go to Mexico or that Mexico is unsafe either. Out of the tourists that visit from America, 0.001% see violence or are kidnapped or anything negative. If anything it would be petty theft near cruise ports and resort towns that would be the biggest culprit of crime for Americans.


“Dude”, murdering a us citizen in Mexico is different than murdering an entire bus of people on US soil.

You say it’s happening all the time but then say it’s .01%.

Looked it up myself, maybe 40 to 300 people annually. Hard to discern how many of those are pure tourism vs visiting family. I suspect you have a greater risk visiting family, especially if it’s a border town.

13.5mm US citizens visit d Mexico in 2024 so .00002% got kidnapped. I bet that number is even lower when you separate pure tourism vs dual nationals or similar going back home to visit.

The point is any action taken on US soil in a large capacity would be seen as an attack by any administration.


I never said “In the US” guy

Of course things happen sometimes. But, the cartels typically do not want to mess with Americans, particularly in tourist areas, because that brings heat they don't want. It's literally bad for business.

I think the GP was referring to buses on US soil rather than Americans on buses in Mexico.

Cartels only strike their own on US soil…

You’re missing the point. Absolutely cartel violence impacts all types of people in the US and Mexico but large scale brutal violence that is usually saved for Mexico since unfortunately the Mexican federal government does not have control in most of the regions.

There is a huge difference between a one off gang killing in the US and someone taking a whole grey hound bus and burying the bodies in the desert.


Which is why I bring up their affinity for going after busses of people, because they have, in Mexico…

The world does not stop at the Us border.


> Dude, Americans are getting kidnapped and murdered in Mexico all the time

Dude, can you put some numbers with a citation behind that? Then we can extrapolate a risk ratio and see if it really merits the "all the time" claim.


who are we (the US)? People who wantonly murder people on fishing boats, etc.

I’m not saying our cartel is any better…

Your use of "our" makes me wonder if the people of Mexico see the drug cartels as "theirs".

Merely pointing out that the US administration is operating like a cartel now a days.

I doubt Mexicans see the Mexican cartels as “theirs” in the same way. Cartels have only been interested in paying off politicians and (as far as I’m aware) weren’t interested in being politicians. However, our politicians here… would LOVE to be Cartel members and make millions it seems. Because they definitely don’t give a shit about law and order.


This is different.

See, Drug cartels over here operate with the blessing and favor of our president. They are tightly connected.

If a cartel dared to ground a US flight. The US government would have a "free pass" to break all hell loose in Mexico, and Sheinbaum wouldn't have a way to stop it.

She doesn't want that in any way, so the message to the cartel bosses would be to be very careful in that respect.

Sure, there have been US citizens killed within Mexico here and there, but those can easily be attributed to local violence. And as retribution, Mexican government sends a couple of wanted criminals to the US.


Yeah, if a cartel actually used anti-aircraft weapons on a US passenger plane in US airspace? It wouldn't even matter if MAGA or the Democrats were in charge. The US would collectively lose its shit and spend the next 10 years and several trillion dollars retaliating against the cartels. The media would be ecstatic, because it would give them a decade of story arcs, starting with "our brave troops in uniform" all the way through to covering the eventual quagmire and anti-war protests. By year 6-8, editorial columnists would be writing columns reconsidering their initial support for the war.

Please, let's not do this.


Wholeheartedly agree.

And you were casting spells at Relic. Bedazzle spells as young gamers played your games and grew up to become artists and engineers…

Remember your audience and not just the product. Homeworld shaped me in ways I couldn’t even tell you.


Reaganomics has never worked

> Reaganomics has never worked

Depends how you look at it.

Trickle down economics has never worked in the way it was advertised to the masses, but it worked fantastically well for the people who pushed (and continue to push) for it.


> it worked fantastically well for the people who pushed (and continue to push) for it.

That would be "trickle up economics", though.


Sure, because it all trickles into their pockets.

Today. Ask again in 6 months. A year.

People have been saying this for multiple years in a row now.

And it has been getting more true for years in a row.

Disagree entirely.

If you state “in 6 months AI will not require that much knowledge to be effective” every year and it hasn’t happened yet then every time it has been stated has been false up to this point.

In 6 months we can come back to this thread and determine the truth value for the premise. I would guess it will be false as it has been historically so far.


> If you state “in 6 months AI will not require that much knowledge to be effective” every year and it hasn’t happened yet then every time it has been stated has been false up to this point

I think that this has been true, though maybe not quiet a strongly as strongly worded as your quote says it.

The original statement was "Maybe GP is right that at first only skilled developers can wield them to full effect, but it's obviously not going to stop there."

"full effect" is a pretty squishy term.

My more concrete claim (and similar to "Ask again in 6 months. A year.") is the following.

With every new frontier model released [0]:

1. the level of technical expertise required to achieve a given task decreases, or

2. the difficulty/complexity/size of a task that a inexperienced user can accomplish increases.

I think either of these two versions is objectively true looking back and will continue being true going forward. And, the amount that it increases by is not trivial.

[0] or every X months to account for tweaks, new tooling (Claude Code is not even a year old yet!), and new approaches.


Using a LLM to program is simply another abstraction level. Just how C was to assembly.

I feel like the nondeterminism makes LLM-assisted programming a different sort of concept than using a compiler. Your prompt isn't your source code.

Fortran to Assembly.

Six months ago, we _literally did not have Claude Code_. We had MCP, A2A and IDE integrations, but we didn't have an app where you could say "build me an ios app that does $thing" and have it build the damn thing start to finish.

Three months ago, we didn't have Opus 4.5, which almost everyone is saying is leaps and bounds better than previous models. MCP and A2A are mostly antiquated. We also didn't have Claude Desktop, which is trying to automate work in general.

Three _weeks_ ago, we didn't have Clawdbot/Openclaw, which people are using to try and automate as much of their lives as possible...and succeeding.

Things are changing outrageously fast in this space.


> Six months ago, we _literally did not have Claude Code_.

Claude Code came out a year ago.


Vulkan is by far the most powerful and the most pain in the ass API I've ever worked with. I agree on every point you just made.

Isn't the idea that 99% of people use a toolkit atop of Vulkan?

Like, these days game devs just use Unreal Engine, which abstracts away having to work with the PS5 / PS4, DirectX 12, and Vulkan APIs.

I imagine unless it's either for A. edification or B. very bespoke purpose code, you're not touching Vulkan.


> Isn't the idea that 99% of people use a toolkit atop of Vulkan?

This idea creates a serious chicken-egg-problem.

Two or three popular engine code bases sitting on top of Vulkan isn't enough 'critical mass' to get robust and high performance Vulkan drivers. When there's so little diversity in the code hammering on the Vulkan API it's unlikely that all the little bugs and performance problems lurking in the drivers will be triggered and fixed, especially when most Unity or Unreal game projects will simply select the D3D11 or D3D12 backend since their main target platform on PC is Windows.

Similar problem to when GLQuake was the only popular OpenGL game, as soon as your own code used the GL API in a slightly different way than Quake did all kinds of things broke since those GL drivers only properly implemented and tested the GL subset used by GLQuake, and with the specific function call patterns of GLQuake.

From what I've seen so far, the MESA Vulkan drivers on Linux seem to be in much better shape than the average Windows Vulkan driver. The only explanation I have for this is that there are hardly any Windows games running on top of Vulkan (instead they use D3D11 or D3D12), while running those same D3D11/D3D12 games on Linux via Proton always goes through the Vulkan driver. So on Linux there may be more 'evolutionary pressure' to get high quality Vulkan drivers indirectly via D3D11/D3D12 games that run via Proton.


You might be unaware of this, but Vulkan Video Decode is slowly but surely replacing the disparate bespoke video decode acceleration on almost all platforms.

Vulkan is mature. It has been used in production since 2013 (!) in the form of Mantle. I have no idea why all the Vulkan doomsayers here think it still needs a half-to-whole decade to be 'useful'.


>”hardly any Windows games running on top of Vulkan”

I run all my windows games on Vulkan.

https://www.pcgamingwiki.com/wiki/List_of_Vulkan_games


280 games over 10 years really isn't impressive (2.5x less than even D3D8 which was an unpopular 'inbetween' D3D version and only relevant for about 2 years). D3D12 (890 games) isn't great either when compared to D3D11 (4.6k) or D3D9 (3.3k), it really demonstrates what a massive failure the modern 3D APIs are for real-world usage :/

I don't think those lists are complete, but they seem to show the right relative amount of 3D API usage across PC games.


I’m just pointing out that Vulkan is supported on all major modern engines, internal and public. Some also go so far as to do DX12 (fine, it’s a similar feeling API) but what’s really amazing is taking all of those games that run on OpenGL, DirectX, etc and forcing them to run on Vulkan…

Proton is amazing and Wine project deserves your support.


What deserves our support is pushing for native Linux games, not helping to keep Windows status quo by making use of Wine and Proton.

It is no different from being happy to use Amiga, Atari, Nintendo, PlayStation, Spectrum, C64, NeoGeo emulators on Linux.


Video games are entertainment. In the old days you inserted a cartridge or optical disc into a physical device. You play the game, finish it and then move on. They are always self contained experiences with a custom UI independent of the OS.

In the best case, explicit Linux support does not affect the experience in a positive or negative way. In the worst case, explicit Linux support means the game can't be played anymore.


> They are always self contained experiences with a custom UI independent of the OS.

Not really, as each OS and hardware provided different capabilities that made some games only possible in specific platforms.

Additionally depending on the platform, some ports were great, others were money thrown into the garbage bin.


Many people need something in-between heavy frameworks and engines or oppinionated wrappers with questionable support on top of Vulkan; and Vulkan itself. OpenGL served that purpose perfectly, but it's unfortunately abandoned.

Isn't that what the Zink, ANGLE, or GLOVE projects meant to provide? Allow you to program in OpenGL, which is then automatically translated to Vulkan for you.

Those are mostly designed for back porting and not new projects. OpenGL is dead for new projects.

I do all my new projects in OpenGL and Cuda since I'm not smart enough for Vulkan.

> OpenGL is dead for new projects.

Says who? Why?

It looks long term stable to me so I don't see the issue.


DirectX 9 is long term stable so I don't see the issue...

No current gen console supports it. Mac is stuck on OpenGL 4.1 (you can't even compile anything OpenGL on a Mac without hacks). Devices like Android run Vulkan more and more and are sunsetting OpenGLES. No, OpenGL is dead. Vulkan/Metal/NVN/DX12/WebGPU are the current.



It’s also almost 10 years old and not a current gen system.

Owners of Switch 2 beg to differ.

And secondly that doesn't matter when we look at the amount of sold units.

Valve can only dream to sell a quarter of those numbers.


The aforementioned abstraction layers exist. You had dismissed those as only suitable for backporting. Can you justify that? What exactly is wrong with using a long term stable API whether via the native driver or an abstraction layer?

Edit: By the same logic you could argue that C89 is dead for new projects but that's obviously not true. C89 is eternal and so is OpenGL now that we've got decent hardware independent implementations.


Wasn't it announced last year that it was getting a new mesh shader extension?

I don't see the point of those when I can just directly use OpenGL. Any translation layer typically comes with limitations or issues. Also, I'm not that glued to OpenGL, I do think it's a terrible API, but there just isn't anything better yet. I wanted Vulkan to be something better, but I'm not going to use an API with entirely pointless complexity with zero performance benefits for my use cases.

That is the approach Google is taking, by making Vulkan the only 3D API on Android.

See Android ANGLE on Vulkan roadmap,

https://developer.android.com/games/develop/vulkan/overview#...


No.

There are literally dozens of in-house engines that run on Vulkan. Not everything is Unreal or Unity.


>Like, these days game devs just use Unreal Engine

This is not true in the slightest. There are loads of custom 3D engines across many many companies/hobbyists. Vulkan has been out for a decade now, there are likely Vulkan backends in many (if not most) of them.


Why not, they do it in real life…

You’re just scratching the surface here. You’re not mentioning agents exfiltrating data, code, information outside your org. Agents that go rogue. Agents that verifiably completed a task but is fundamentally wrong (Anthropic’s C compiler).

I’m bullish on AI but right now feels like the ICQ days where everything is hackable.


Guidelines | FAQ | Lists | API | Security | Legal | Apply to YC | Contact

Search: