I'd be a lot more comfortable with gun ownership in the US if we could require every owner to have military-level firearms training and certification.
HN user
kelnos
Core developer at Xfce, maintainer of xfdesktop[0], xfce4-notifyd[1], and now building xfwl4[2].
I was at Twilio for 10 years (left in 2022), and worked at several other companies before that.
[0] https://gitlab.xfce.org/xfce/xfdesktop
[1] https://gitlab.xfce.org/apps/xfce4-notifyd
[2] https://gitlab.xfce.org/kelnos/xfwl4
You can personally interpret the US constitution however you want, I suppose, but the only interpretation that matters is that of the Supreme Court. And they have held over and over that 2A includes a right to many different kinds of firearms.
I feel like you're the one who is poorly informed?
(For the record, I am in favor of strong gun control laws, but your statement just ignores reality.)
Are handgun deaths predominantly murder, or are they mostly accidents? If the latter, then I don't think it's odd to go after AR-15-type weapons, which usually end up killing people during mass shooting incidents.
The difference is that Chinese companies and their government are much more intertwined than their American (or Western, in general) counterparts.
It seems much more reasonable to look at actions a Chinese company takes through the lens of what their government wants, than it does to do the same for a Western company.
Agreed, but the Chinese government has much more of a say in what their companies do than is common in the West. It seems perfectly plausible to me that the Chinese government wants US AI labs to fail, and might direct some/all of their own AI companies to release their model weights.
Did the GP actually say that? I think you're putting words into their mouth.
They specifically referenced the actions of the Chinese government, as well as mentions of IP theft. I don't think that covers 1.5 billion people. More like a few thousand or tens of thousands?
I mean, I expect Google open-sourced those projects because they see economic benefit to themselves in doing so, not because they are good-hearted.
Chromium is an especially silly example to use: they benefit by controlling the web platform, and open-sourcing Chromium has allowed them to get their web engine into many other browsers.
The issue with stories like yours is that they don't really tell us anything. The incidence of racial profiling is low enough that some people, such as yourself, will never have a problem, ever. But that doesn't mean there aren't problems, and that they don't crop up often enough to be something we should be really uncomfortable with letting happen.
In Latin America, things are considerably more professional, and the government officials don't act like they hate you and just want you to go away.
Depends. Someone I know had a terrible experience crossing into Belize on a land border. Was held there for two hours while they "verified" something. Seemed like a power-tripping agent.
SFO has fingerprint machines in all lines, including the citizen lines. And in the Global Entry lines, even.
I agree with you that US citizens aren't supposed to be fingerprinted on entry, but I am absolutely unsurprised that it happens sometimes.
It's hard to say what their motivation is.
Feels pretty easy to me.
They want to turn LLMs into a commodity, and watch the US AI labs crash and burn.
There will still be plenty of customers who will pay them to host the models and run inference, even if the weights are open and others can offer competing products. (If necessary, the Chinese government can ban use of foreign inference services by Chinese citizens and businesses to give their own companies a domestic monopoly.)
When their models equal or surpass those from the Western AI labs, they can even stop releasing weights for new models, and keep all the inference revenue for themselves.
Meanwhile, they're still manufacturing much of the hardware that everyone in the world needs in order to run datacenters (see also: Spolsky's "commoditize your complement" essay).
Beyond that, it's a soft-power play. As the world keeps looking at the US more and more skeptically as an ally and superpower, Chinese companies releasing weights for competitive models is a way for China to look better and more world-minded.
It may be available to us, in the form of a git checkout of their public repository.
It's true that we don't know exactly which, and whether or not there are any other changes added on, but that is the case for every single release of Claude Code that has used bun.
A charitable interpretation is that the version of bun inside Claude Code corresponds to some recent state of their public git repository, so if you clone that, you more or less get the same thing.
I don't see what it would buy Anthropic to have some sort of secret internal fork with special private code in it.
Your attitude suggests you are unserious about the future of the human race.
If the cookie in this analogy is crack cocaine, then yes, I'd say it's immoral.
That suggests the system working as it should. They present terms for use, you don't like them, so your don't use it. One of their other products has terms you're ok with, so you use that product.
Good, fine. This is an example of trusting the company to honor their own terms, not the opposite.
Your argument boils down to "they've done something I find objectionable, so that means everything they say must be lies".
I'm not comfortable with how these models were trained. I have quite a bit of open source code out there, and I personally see such training as copyright and license laundering.
But that's not how the law sees it, and I grudgingly accept that, regardless of how I may feel, and I don't let my feelings on the matter make me think irrationally when it comes to whether or not these AI companies honor the terms they provide.
Sure, they might be breaking their promises, training on our data when they say they won't. But I do think they most likely aren't, and that it would be corporate suicide if they were and it ever came out.
... and nothing of value would be lost.
I don't think the GP said anything about their own practices, just their impression of the majority of devs.
Maybe turn down the temperature on your vitriol a bit?
Your anecdote does not make GP's comment "patently untrue". It's just a counter-example, and we don't know how prevalent your scenario is compared to GP's.
(And I agree with the GP. I'm fairly cynical about most developers' security stance and threat model. Source: my own usage patterns.)
As an American, what? A heat pump is a heat pump. An air conditioning unit is a subset of heat pumps (that is, a heat pump that can only run in one direction).
I'm just as confused as you are about the temperatures quoted though. I'm most comfortable around 19C or 20C. At 23C I start having trouble sleeping. At 26C I'll be openly sweating.
My wife and I cook often, but we'll usually go out to eat (together or separately) once or maybe twice a week. We love having several good restaurants within walking distance, with a variety of cuisines.
We also love having a small but well-stocked grocery within walking distance.
I grew up in the suburbs, and while yes, they do vary wildly in density, many will not have these walkable features.
Neither?
Amazon sounds like an absolutely terrible place to work, and I'm not convinced it's going to be stable or long-term lucrative for someone just entering the job market now.
I don't think a masters in CS is going to buy you flexibility and money in the long term. Specializing further is the opposite of flexibility. Staying in school longer carries an opportunity cost, and a masters doesn't increase your salary much when it comes to software.
(And if you're aiming to be a Big Deal at an AI lab eventually, a PhD would probably be what you want.)
Do the masters if you think you'd enjoy that particular experience. Go out and find a job of you think you'd enjoy that more.
If that's the case, I don't really get the purpose of this. It's presumably not a useful system for day to day computing. The main reason I could see someone wanting to build this would be as an educational exercise, and using an LLM to do it completely fails at that.
Collectively, yes, but not individually.
The problem is that the electorate tends to not understand the concept of second-order effects. For example, a college graduate in the arts might, directly or indirectly, generate more economic activity than someone without a college degree, regardless of the difference in income level of people in those two buckets.
Typical attempt by the right to disrupt higher education. Hopefully this garbage gets challenged in and struck down by the courts.
(The person you are replying to is one of HN's mods.)
I have lots of smart gadgets that also have physical switches. It's convenient to be able to control them in more than one way, from more than one location.
Anyway, why are you commenting here if you're not into this sort of thing? Feels like you're just trying to stir up an argument.
Absolutely, but just as it's not ok to enter someone's home just because they forgot to lock the door, it's not ok to exploit access at your old employer because their offboarding process missed something.
I do the same as GP does; I don't want there to be any chance that my former employer has forgotten to revoke access to something, so I make sure to clear out anything that might remain on any device that I don't return to them.
Who knows, maybe another former employee will decide to steal from them around the same time I leave, and me having access credentials on a personal device, even if I haven't used it, might arouse suspicion.
This is a fine example of coming up with some "plausible" numbers to support a conclusion, but all of the "plausible" numbers are actually wildly incorrect.
A new ambulance will probably run 2-4x what you quoted. Gas/maintenance will not be comparable to a regular automobile; it will be significantly higher. Paramedic wage (though there will probably be more "cheaper" EMTs out there than paramedics) seems right, maybe a little high for some markets. $12k for liability seems shockingly low; I wouldn't be surprised if you're off by an order of magnitude.
Regarding salaries, you're not just paying for the people in the ambulance. You're paying for dispatch, operators, support staff, administration, etc. You're also paying them 24/7, not just for the times that the ambulances are actually making trips.
You also left out the cost of medical supplies and equipment, both of which need to be maintained and replaced.