That's the dream. But I am quite sure Framework is looking into it, if not only because Windows on ARM is gaining traction, which means developers need an ARM machine to develop. I can imagine some would want two identical machines, one with X86_64, one with aarch64.
I tend to agree but you can't deny that the widespread use of LLMs means that codeberg will expend more resources overall for projects taking advantage of them. And its members have spoken. We might disagree about the conclusion they came to and even how they went about it but their stance is clear now so we might as well just move on.
Yeah but if it's about the hardware resources, why not just establish an option to ban projects that use them unreasonably for their projects relevance? In their explanation they say each project will be evaluated individually and the whatever-AI-allegations are not the primary reason to eventually ban.
They could have established a direct way to protect hardware resource with an option to ban users that cause unreasonable load. Maybe not great but understandable considering they are an e.V.
It is already like that. They aren't hooking up an LLM detector to the ban button. They are saying this is grounds for human-triggered enforcement action.
Simplest answer is that there's no representative test case in the test suite yet.
Unfortunately the problem of compiler testing is a very challenging one IMO. You can't test it exhaustively, or perhaps if you did write such a suite it would take longer to execute than is feasible.
> we all are just lucky enough that most messed up optimizations will break _something somewhere_ early enough to not get merged
Your compiler is only as good as the set of code people routinely use it on. Rust has exploded in popularity but it's still much less popular than C, C++.
You know that feeling when you work on a feature for weeks or months and then something comes along and the feature is no longer needed or the project is cancelled?
It's a pretty frustrating experience -- was it all for naught? Maybe it's useful to vent about it a bit.
I definitely had this feeling early on in my career, but it did flip around somewhere around halfway through.
"We're not shipping? Well, that's a bummer, but also, what a relief! If building it that was this hard, I can only imagine how bad shipping it would've been; now we can delete that code and with it all of the maintenance we would've had to commit to for years."
The personal attachment just had to go eventually. It proved not to be terribly helpful or healthy anyways.
Ok, but then honestly, spending 40+ hours/week in an office, doing work that's neither enjoyable nor useful doesn't seem like the best way to spend that time.
It also feels like willfully abandoning the bit of agency you still have if you don't even try to understand why the world around you works like it does.
>40+ hours/week in an office, doing work that's neither enjoyable nor useful doesn't seem like the best way to spend that time.
For almost everyone, working is not the best way to spend their time, it's just how they can afford the stuff they do the rest of the time. Obviously it's preferable if it's useful and/or enjoyable, but they're not necessary qualities.
>It also feels like willfully abandoning the bit of agency you still have if you don't even try to understand why the world around you works like it does.
Odd. "Agency" usually refers to the ability to exert will. Understanding would not seem to contribute towards that.
Yep 90% of companies don't matter and don't affect things. Only 10% of people do. That's just the way the world works. There's no way to know before you start so just live while you can enjoy.
I worked at a company whose product was truly boneheaded. Without giving too much away, it’s the kind of technology that would have been useful if we lived in a world where smartphones weren’t being carried around by literally everybody.
I knew this, but took the job because I was burned out and knew I could spend a year or two coasting and padding my resume with some interesting things.
I came to the conclusion that the company was a grift, but at least they took care of their employees and included them in the profit part of it.
We had startup perks that were basically paid out in cash when the pandemic hit. The “gym” perk became $500 in cash which could be spent on anything vaguely fitness related, like an Apple Watch. The commuter benefits rolled into our accounts which gave me free tolls for years afterward. Instead of taking all the money, they cut us in.
So yeah, maybe frustrating if you expected your startup to be successful, but that’s so often outside of the control of any engineer. It’s always a crap shoot. Get your best offer and make the most of it. You can do resume driven development even in the shadiest of firms.
This attitude actually becomes the only viable strategy once you realize that the majority of the economy in the majority of the world is connected in some way with some kind of graft. Even honest salt-of-the-earth sewer workers or something are probably pawns in some larger fraud scheme. The "economy* is in some sense one large scam.
> You know that feeling when you work on a feature for weeks or months and then something comes along and the feature is no longer needed or the project is cancelled?
I would have thought most people would grow out of having this kind of feeling after their 1st job. But I also definitely work to live, not live to work, so YMMV.
So much of what I’ve worked on in my career has proven to be utterly ephemeral. I’ve learned not to dwell on it too much, in part because one of software’s great strengths is its malleability[0].
However, I was quite surprised a few weeks ago, on a client project, to find in one of their repos a chunk of example code that I’d worked on 22 years ago.
[0] Being real, a lot of the ephemerality actually stems from questionable commercial decisions, working on the wrong thing, etc. But some at least is a legitimate result of evolving markets and needs.
Same same. I don't expect any of my product code to survive for very long.
I suspect some of my open source contributions will live a long time. Not my personal projects that I make open source just in case, but the (very small) contributions to fix things in the dark shadows of established projects with longevity. Some of that will become obsolete and hopefully be removed, and some might get refactored eventually, but if the project is older than my career it's may well last beyond me.
Our work is extremely ephemeral by nature. Many times I only value the communication that happened and the shared understanding that was built. The code is an .. artifact. Nice to hang on the wall.
Do you get your major feeling of success primarily from work, or is it the rest of life? I don't think I need to spell what is viewed as the proper long term winning strategy here. I know folks who ride the work part, and let me tell you - unless you land a stellar employer and exceptional boss, its recipe for a lot of miserable days and worse, for decades. Its also a symptom of what I would call 'unfulfilled life', but thats my personal take please don't get offended quickly.
With that mindset (or work-to-live or whatever you can call it), these things are just an afterthought. That after-work climbing session and that weekend meeting with friends or hiking trip in the mountains with kids mean world to me, and I fully indent to keep that mindset till retirement and continue with it further. If it means I won't get into top 1% or whatever I am fine with that, QoL is firmly above that and career rat races are meaningless (and fruitless) ego polishing / insecurities managing exercises.
Look at the cryptocurrency and Bitcoin economies for an example. Instead of being a democratic mining economy where spare cycles are used, only companies which invest capital to find semiconductors from the latest process node combined with facilities and inexpensive electricity benefit from mining.
Only the next Standard Oil / Amazon / Google will benefit from the people-free economy.
There are quite a few machines connected to the internet right now with no owner. Boxes forgotten over time and power consumption not enough to matter.
If we keep on the trajectory of energy usage and computation, in 50y you might have the same with smarter models. Also, a virus could have its own bitcoins to rent compute and work for more.
Ownership is as much a social construct as Money or The Economy. Do with that what you want.
> There are quite a few machines connected to the internet right now with no owner. Boxes forgotten over time and power consumption not enough to matter.
They are on somebody's land, and unless they run completely off of solar or turbine then they are still drawing someone's electricity. Their access to the internet also must be mediated through a SIM card or a wire to someone's network. With no dispute from other possible owners, the owner of the related land, electricity, and network access (far and away these will all be the same party) is the owner of the internet-connected device.
Sure it's a social construct, but billions have killed or been killed for money, land, and ownership of assets. Social constructions at their extreme are just abstractions of power which determines life and death. Something being a social construct does not make it invalid or ignorable. All of life is a series of intrapersonal relationships built atop social constructions. I'm sorry but I find pointing out something being a social construct only serves a somewhat naive/juvenile purpose as an easy way to make a statement look nonsensical.
Some of us in Texas are all too familiar with the problem of balancing load with generation, the risk of a cascade failure causing a slow restart.
During winter storm Uri, they did a duty cycle where we only had power available for ~6-12 hours at a time on the days it was available. This was apparently to avoid that very problem.
So far as I know, the obvious mitigations like winterizing NG generation and/or peering with neighbor grids have not been performed.
It's not like the article is indirectly related to the subject of the conference. It is critical of the government, but not in the "human rights abuses of this administration threaten us all" way (though even that seems reasonable to discuss).
Is it your position that if an article is critical of a world government it must not be discussed at a scientific conference? Or even "you should expect to get ejected from a conference if you criticize the host government"? Because believe it or not, that's not been a problem in the USA prior to Trump. And it runs contrary to how science should work.
If Adria Richards and PyCon can make a such fuss about a joke [1], then other conferences can do the same about current political events.
I personally wouldn't mind or better said I wouldn't be offended because the topic might be boring for me.
This is a false equivalence. "What constitutes appropriate humor" versus "Government silences dissent." Just because the Diabetes conference cited their code of conduct doesn't mean that's actually what this is about.
This is what people seem to get mixed up about the First Amendment to the US Constitution. These scientists were removed from the conference because they were highlighting the scientific role to push back against government censorship. Not because it wasn't germane to the conference, but in furtherance of the censorship itself. The US Government participated here indirectly via its chilling on scientific discourse.
Comments in this thread suggesting that "this is just some private actors" are mistaken. This is absolutely the consequence of the Trump HHS policymakers decisions.
I was under the impression that the First Amendment means that the government can't arrest you for what you say, not that it lets you talk about whatever you like at a conference organized or supported by it. Nevertheless I understand the urge of those guys to have their voices/ideas heard.
For what it's worth, ADA has released this press statement [1]:
> As a 501(c)(3) organization, the ADA has safeguards in place to ensure that it complies with all IRS regulations. This includes maintaining a strictly nonpartisan environment at all organizational events and functions while engaging across party affiliations to advance our mission. We have always, and will continue to welcome scientific inquiry, respectful dialogue, and diverse perspectives in the pursuit of better outcomes for people living with diabetes and obesity.
One of science's most critical roles is to inform policymakers. And if they can't do that job effectively then it's right and just to point out the problems preventing it. Scientific conferences that fear critiques of the government chill new scientific publications.
It's not like they were handing out "Trump sleeps during press events" posters. You should read the article they distributed, it's very relevant to the conference attendees.
> One of science's most critical roles is to inform policymakers
Yup. From their article: “DCCT/EDIC revolutionized the approach to treating people with type 1 diabetes, establishing standards for glucose control and resulting in improved quality of life along with clinically significant reductions in the risk of diabetes complications and major adverse cardiovascular events.
After 44 years, it continues to provide new insights, including showing that in adults with type 1 diabetes, neurodegeneration is likely the result of non–Alzheimer disease mechanisms. DPP/DPPOS, which enrolled people with prediabetes, demonstrated the benefit of intensive lifestyle intervention and metformin in reducing the risk of developing diabetes. These findings led Congress to approve an amendment to the Social Security Act to establish the Medicare Diabetes Prevention Program and provide lifestyle intervention services for eligible individuals.”
reply