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

You are being too optimistic:

That relationship is not true. It might be true at the efficiency frontier. Maybe. I don't know.

But it's very easy to compromise on convenience without gaining any security and vice versa.


Presumably he's using the native harness for each model? (The one their company recommends.)

DeepSeek's models are also worth a try, mostly because they are so cheap. If you can split up your tasks so that deepseek can handle them, you can save a lot of money. (You can ask a more capable model to help you with the split.)

This user is clearly not concerned with saving money, they’re looking for quality.

Especially since that check might have further problems. The simpler you can make your code (and still solve the problem) the better.

I don't know if I'd count paving over the reality of what you're running with an abstraction layer as something simpler. If anything it demonstrably makes addressing issues like these harder.

I think you're right, but many people are running systems at large scale, with lots of hosts and all kinds of abstraction layers and stuff. Now this exploit shows up. What are they to do? It's just a practical thing they can do right now.

First we must stop the bleeding. After that, yes, it's worth figuring out if they should change how things are set up.

I've been in situations like this before. Sometimes the complications and abstractions were my fault, sometimes I inherited them. Either way, as responsible owners of the overall system or business, we've gotta be able to respond fast to issues.


You can use the kernel config to remove these things.

For comparison for chess, it didn't last that long.

Chess is very constrained, and doesn't require all sorts of knowledge about the broader context outside the game.

Once models are more hooked up to sources of information, their capabilities will improve. Currently, humans are a bottleneck for providing the information relative to specific problems.

"Take the human out of the loop" - I think there was a documentary about that, with Matthew Broderick.


> Chess is very constrained, and doesn't require all sorts of knowledge about the broader context outside the game.

e.g. chinese honeypots, or people sifting through your email, or stoned admins pushing scripts they shouldn't, etc. LLMs can sometimes account for this, but mostly via interface constraints or sanity checking later


A strange game! The only winning move is not to play

Only if you're in Zugzwang

> I think only really smart people are going to be left in Security and Software.

I'm looking forward to eventually getting there. At the moment, there's so much low hanging fruit that even someone without a clue like me can just point a fuzzer at some random Linux kernel code and still discover problems.

(Most of the problems I found from that exercise are patched by now, if you use a recent enough kernel.)

I used Kimi to drive the fuzzer. The model didn't have to do anything smart, just do the tedious work of writing the glue code to make the fuzzing work. AFL++ did the 'smart' parts, and I had the idea to kick it off and oversaw the patches.


Low earth orbit can't sustain a Kessler syndrome for more than a few years. Atmospheric drag is too large. (And it gets proportionally larger for smaller fragments.)

more worried about the upper range of LEO that'll last a couple hundred to thousand years.

Maybe. But that just makes the orbits unusable for longer term satellites or manned space stations. You can still pass through them with minimal risk (because the dwell time is so short.)

So just half a decade of LEO's being economically infeasible then?

I guess that gives the rebels time to organise for the next wave of fragmentation and shrapnel launches.


> So just half a decade of LEO's being economically infeasible then?

Actually, it depends a bit on how disposable and cheap your satellites are that you want to put there. If you can drop costs enough, even an expected operational time of a few months before they get hit might be worth it.


> Commercial customers largely used Ariane.

That and the Russians. But SpaceX is cheaper than both of them.


The current list price of Falcon-9 flight at 74 millions is a third more than what Russia used to charge for commercial payloads. Production cost of Russian rocket is below 15M USD and probably was below 10M at the peak of Russian comercial services. This is comparable with Falcon-9 amortized cost per flight.

Are you adjusting for inflation over the decades between when russia provided these services for financial gain and what falcon does now?

Also don't forget what a state their launch facilities are in. Just because they could do something before does not mean they can do it now. They haven't invested properly in their infrastructure.

Wouldn’t this imply the SpaceX costs need to adjust for the infrastructure they didn’t pay for? Not slighting them, just acknowledging an apples to apples comparison is hard



You can't compare rocket vs rocket. Cheap Russian rockets are way smaller. Falcon 9 destroyed their market even before it was reusable.

You are also missing that Falcon 9 costs as much as costumers pay, SpaceX internal cost are significantly less. They just don't have competition to make it cheaper.


Yeah, I'm a bit confused about how, if Russia was supposed to be cheaper, did SpaceX get all the marketshare.

Starship will make it even cheaper, if it works.

With price per kg to orbit dropping, you can afford much more off-the-shelf components for satellites and using more cheap and redundant parts, instead of high quality bespoke stuff.


Also, size! The sheer volume of starship would make say, James Webb telescope a lot less complex with a 175x30ft size ship to deploy it.

Only if SpaceX can work out how to get it out of Starship. It is not as trivial as it sounds, and so far they have shown no inclination to work on the problem.

They already have a signed contract for a customer that requires a large payload bay door, they'll get there. They've just been focused on getting proof-of-concept orbital deployments first.

What customer is that? The only commercial contract is for Starfall which launches from the pez dispenser. The HLS requires a human or rover sized door, a lot smaller than one a telescope or GEO satellite would require.

Superbird

True, but it seems like the Shuttle proved feasibility.

They could if they wanted to but for now it seems optimized to poop out little bom... sorry I mean little Starlinks.

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

Search: