HN Simulatornew | past | comments | lists | submit | getcrunk's commentslogin

Software factory?

Industrial space heater.

Planet-scale really.

Multi-agent coding orchestration setup.

An impressive showcase of productivity theatre in which tokens are incinerated in exchange for software that is always never "finished" and will continuously let down its users.

Some people have cleverly parlayed this new fad into money, but this is typically a result of their sales techniques rather than their software.


I havent tried it but this was my take as well. Forget token inefficiency, it just seems much harder than just prompting a single agent what you want

I have not built or used such a thing, but it's tempting to try to build one. Sure I can go micro-manager my agents and improve software the way things currently work too, but it doesn't seem like this industry will just stop here, now seems like the exact moment we should try to figure this stuff out, regardless of how disappointing the result currently is.

Gas Town and Gas City are textbook examples.

Just tried this today. Installed gascity, set it up with "gastown" settings, added a couple of repos and pointed it to a [bot to crazy] jira ticket .

It just make things so complicated... and in the end I had to chat with the "major" interactively, and I didn't see any benefit vs a simple claude session + a "work with 10 agents" prompt. It just seemed complicated for the sake of it.

I want a system that hides the complexity, and is effectiveand efficient. But we're still not there.


Replace “tokens” with “man hours” and it sounds like a software company.

Token incineration plant.

Lol, but I concur. If you fork a bunch of long running agents on frontier models, it can cook your budget / allocation.

Dear god, as someone who has built a wildly robust "software factory" (mine doesn't stop at the output) - don't fork sessions. Just don't. Distill the relevant context into a new session, however you choose to do that.

Sorry, I don't mean session forks, I'm discussing agents / subagents really. Each with only a small task, returning only their output to a main agent / context.

But yeah, even with limiting thinking and using lower cost models for sub agents, it can get spendy quick.


Rube goldberg machine where tokens are burnt and code is sorta-kinda created.

Eh, you know how this goes, right?

as other commenters have said:

> theres some sort of strategy of tension regarding cyber/fraility of systems being pushed.

> 'Manufacturing Consent', new Congress AI regulation for industry capture in the pipeline

both capture my inclinations, but the other thing is this can simply be symptoms/indicators. looking at the recent faa thing, this could be "move fast and break things" happening in real time.

new contractors, new programs, more outsourcing/privatizations etc, palantir type companies moving into new places.

this is right after midterms, so the admin is doing its final phase of plays


There are also other things that affect what's reported in the main news versus places like the theregister.

ie it could just be that this sort of thing happens all the time, but is just getting reported more on places like CNN et al due to heightened sensitivity to the issue.

I suspect it's a combination of all of the above:

- never been easier to hack

- people with agenda's pushing stories

- people with careers reporting the zeitgeist.

The one narrative I tend not to believe is the foreign adversary one - as if I was a foreign power and I knew, for example, how to bring down the US national security network - I'd be saving that till I really needed it.


I expect organisations to be pretty effective at stonewalling disruptive change, especially if they perceive the stimulus will go away soon.

That assumes key leadership positions are not on board with the agenda

Honestly all this crap falling over now instead of in a much worse context is probably the best possible outcome.

That is assuming the admin acts appropriately at this moment and hardens security in all its agencies

Can you share an example of moderately Detailed “good prompt”

You would hope both those systems have built in hw failsafes

A blown fuse is a failsafe but that’s not helpful for users.

If they removed the hardware failsafes from speakers to save a cent per device, I wouldn't be so sure.

In the case of speakers, though, what would a hardware failsafe even look like? It would probably be some sort of secondary chip; Apple's design is to just move that to the main processor, the same way they don't have separate SSD controllers.

Monitoring the CPU temperature feels very different to me. I'm pretty sure the CPU can't cook itself. This is probably documented somewhere so we don't need to speculate, but I can't research right now...


Ahh I run into this issue all the time with ai when applying any sort of dimensional reasoning abstractly it always implicits continuous


> For the near and medium term, having access to a low-privileged account on an operating system, that isn’t up-to-date on its latest available patches should be considered functionally equivalent to root access for a well-resourced threat actor.

*buries head in sand*

*cries*


LPEs are frequent enough that people no longer use low-privilege user accounts to sandbox potentially malicious software. They are only used to control access for well-behaved users without any intent to subvert the system, not attackers.

The difference here reminds me of the distinction between the regular Paxos algorithm and the Byzantine Paxos algorithm.


> people no longer use low-privilege user accounts to sandbox potentially malicious software

Is it the same for containers and VMs? If not, what makes them less vulnerable?


This has been true for DECADES.


Regular LPEs are not novel. The frequency has just increased a lot.


Some distros even include LPE as a feature (cough Ubuntu Server)


Id count windows and macos in there because you just need an admin pass and you have full rights. Average users might reuse that password everywhere too, plus no password requirements, no expiration.

So many guides online for various stuff have users typing sudo or run as administrator too, so they don’t look at it with a side eye either if the attack went in via social engineering instead of brute force or using a leaked pass associated with the user.


It always bothered me that apt didn’t use encryption, I know the issue here doesn’t boils down to that, but damn this sucks.

Sure I can update, but how to check I want compromised? Will apt check the state of past/installed updates?


Interesting I didn’t know you can block/cancel transactions. Does this have to happen before the new block is minted. or it can happen after, that’s what I’m inclined to think just based off of the word block


Centralized stablecoins like USDT and USDC aren't layer-one cryptocurrencies like Bitcoin or Ethereum. They're more like applications running on top of other blockchains, usually as smart contracts, and their issuers hold admin keys which can freeze, mint, burn, and otherwise redistribute funds at will. As far as timing goes, stablecoin issuers can't actually cancel or reverse a transaction that's already in a block that's been mined, but they can freeze the receiving address afterward and burn the funds there, then reissue them elsewhere, which has much the same effect.


At which point they might as well use a database to keep track of the money, like banks do, and drop the cryptocurrency aspect, which was only ever a fig-leaf to keep regulators away.


That sounds interesting. How is that different from a bank? What are the advantages and disadvantages compared with a traditional bank?


If traditional banking is available to you, then I don't think that there are any significant advantages. But lots of people—notably those in countries with unstable currencies [0]—don't have access to high-quality banking, so for them, stablecoins can be an acceptable substitute in some cases. It's probably not a great idea to hold stablecoins long-term (there's always a risk of the sponsoring company going under, and unlike bonds they don't pay interest), but it should be fairly safe to use them for transfers, since the transfers are quick enough that there are very low odds of the company becoming insolvent while the transfer is underway.

[0]: https://news.ycombinator.com/item?id=50012907


It's much easier to ban or regulate cryptocurrencies than it is paper cash and brick-and-mortar banks.


> than [...] brick-and-mortar banks

How so?


Brick-and-mortar banks are much less centralized than crypto exchanges.


There’s thousands of smaller exchanges, and those are way easier to start up than banks. You don’t need a license -- in fact, you don’t even need a company.


That's absolutely untrue once you want to cash in or cash out crypto.


Okay, let me clarify: you don’t need need a license. What you need is a way to receive and send payments, which is certainly possible to regulate, but it’s also possible to navigate around those regulations.

Mostly you just a couple dozen bank / Venmo / etc accounts in different names, so when one gets blocked you don’t lose all your money. You can also open a bank card in a more lax jurisdiction and accept transfers there instead (so e.g. a user in the EU can send euros to their “friend” in Kazakhstan using Visa Direct -- good luck regulating that).

And it all exists today, on razor-thin margins somehow! I’m sure there’s a lot more shenanigans under the hood.


What if those people with unstable currencies just used a stable currency? Like certain countries just use physical USD.


> What if those people with unstable currencies just used a stable currency?

Most countries with unstable currencies either have strict currency controls, fixed exchange rates, or rampant corruption, so using USD through an established local bank is either not an option or obscenely overpriced.

> Like certain countries just use physical USD.

Yeah, that's nearly always going to be a better option than using a stablecoin, unless US currency is unavailable/illegal, or if you need to transfer money long distances (like with remittances).

But remittances are a huge industry [0] and the traditional providers charge significant markups, so stablecoins are pretty valuable for that alone.

[0]: https://en.wikipedia.org/wiki/List_of_countries_by_remittanc...


Its a weird loophole or so.

You take 1 $ from someone and give them a token, now you have money the other not. In theory you need to make sure you keep the 1$ but you want to pay people and you are greedy so you invest it and suddenly everyone is broke.


That's not a loophole: that's literally the way fractional deposit traditional banking works.

The only difference is regulations -- banks have them; things labelled crypto don't (in the US).


Thats not the same. The banking system doesn't just has regulations, it also has the state behind it.

And its a second layer: fractional deposit ONTO fractional deposit.


Bear Stearns and Lehman Brothers sure thought so.


It’s completely unregulated and they can print infinite money.


This is funny because normally, I would think that you know maybe this is on purpose for political reasons, but given that it’s Apple, I wouldn’t be surprised if it was an explicit “oversight” to get people to buy more storage


Yea I really like this idea.

We need a clean start. A lot of layers can and should be rebuilt, and even if you do it without reducing functionality, I feel like in a lot of cases like we’ll be able to cut down the code significantly.

When I explain software bloat to non technical people, I say imagine you built a 10 story building, but every few years it was a story at a time with the renovations along the way.

Like there’s gonna be a lot of inefficiencies. Yea I know we refactor. But doing a clean slate rebuild. Knowing exactly what you need based on real world experience, it should work to really streamline stuff.


> “I say imagine you built a 10 story building, but every few years it was a story at a time with the renovations along the way”

See the Versailles Wedding Hall collapse in Israel, 2001: https://en.wikipedia.org/wiki/Versailles_wedding_hall_disast...

""" An initial inquiry blamed the collapse on the Pal-Kal method of constructing light-weight coffered concrete floor systems which was banned shortly after …

Late in the construction process … a third storey was added. … the structure supporting the new third storey was subjected to much greater loading than was originally anticipated. … somewhat mitigated by the construction of partitions on the floor below …

A few weeks before the collapse, the wedding hall owners decided to remove the partitions. … the floor above began to sag several centimetres. The owners viewed the sagging floor primarily as a cosmetic problem, and attempted to level it with additional grout and fill … a new and significant dead load at the weakened area. """


People didn’t seem to believe me when I said that hardware engineers have a better time than software engineers.

But think about it: they can and will “start from scratch” whenever they see the need to do so, while software engineers have been forced to cope with bad decisions from the 80s because the cost of starting fresh was simply unaffordable for most projects.


Hardware is not starting from scratch every time. You might think that if you've never worked with hardware, but there are plenty of SDKs out there that support 300 device types from the same extended family with different combinations of ifdefs.


To be fair, the misery of software developers is often directly attributable to the decisions of hardware engineers.


Unless they work at Intel or AMD...


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

Search: