My immediate thought. Except not "vanilla" YAML, but a safer stricter subset (iirc some people published a spec about it): no implicit conversion, no norway problem, etc. If only this gained actual traction.
The JSON in the article is a bit, let's say, heavy on the different objects and does not try to represent anything useful with most keys. All the things like `greaterOf`, `sum`, etc are much better expressed as keys than `{"children": [{"type": "greaterOf", ...}]}`.
Basically something that feels an reads like "freeform" yaml, yet that has an actual spec.
"- yeah I have a macbook" "- what, an air?" "- no a macbook" "- ..?" "- the one in colors, not the one-port 12 inch one from 2015 but you know it just released!"
This already happened in 2015, they probably don't want for it to happen again.
Honestly the 8GB is not really an issue. As opposed to basically every other computer on this price range, Apple puts real storage in their machines, making a well-tuned swap simply transparent. I'd also bet they have very performant hardware engines for memory (de)compression.
A few years ago, my parents asked me for a laptop for my sisters, for university use. We targeted this price range. It's shocking but pretty much all laptops from Dell, HP, etc come with some form of eMMC storage. And I'm not speaking about the other specs like display or build quality. We ended up buying second-hand M1 and M2 macbook airs, and both I and my sisters are very happy about it.
(also, as the "tech support guy" of the family, I'm oh my so happy about them not running windows)
The SSD in the Neo only manages around 1,500 MB/s in sequential benchmarks, it's not an impressive drive.
> It's shocking but pretty much all laptops from Dell, HP, etc come with some form of eMMC storage.
I just went to Dell's website and picked a random $400 laptop and it had an NVME SSD. The $650 Dell 14 Essential also is NVME. Both of which are M.2 so easily upgraded, replaced, or have data recovery done on them. The only eMMC options I'm seeing are the $300 Chromebooks? Which is no where close to "pretty much all laptops." In fact it'd be "pretty much none of the laptops"
> The SSD in the Neo only manages around 1,500 MB/s in sequential benchmarks, it's not an impressive drive.
That's sequential, not what you want for swap, but already a good start. I agree that it's not impressive, but already leagues ahead of a SATA SSD. And for swapping a 8GB machine it's more than enough (when the swap pattern is sequential though): you swapped your whole system memory in 3 seconds, which is impressive.
> The only eMMC options I'm seeing are the $300 Chromebooks? Which is no where close to "pretty much all laptops." In fact it'd be "pretty much none of the laptops"
Then it's good the situation improved, genuinely! Less e-waste being on the store shelves. Pretty sure windows is nigh unusable on eMMC. And yes, those were sold alongside chromebooks, but at a markup of a "real computer" despite having roughly the same internals.
Another thing that could impact, though, is availability in different markets. I am in France, and the offerings are perhaps worse than in the US? (quite likely, in fact). Add to that the usual price markup where US companies tend to do, at best, 1 USD = 1 EUR, and we get worse machines for the equivalent price range.
> you swapped your whole system memory in 3 seconds, which is impressive.
As a user a 3 second hang is unusable. Also, critically, swap consumes the life of the drive. Since the Neo's isn't user-replaceable, a 3-5 year lifespan before death is actually a non-trivial compromise, although time will tell on that one I suppose.
Should be fast enough to swap in a browser page I guess. Overall you're right that it's the wrong device for memory hungry applications, but it's not the target audience.
Two years ago, I've had Factorio crash once on a null pointer exception. I reported the crash to the devs and, likely because the crash place had a null check, they told me my memory was bad. Same as you I said "wait no, no other software ever crashed weirdly on this machine!", but they were adamant.
Lo and behold, I indeed had one of my four ram sticks with a few bad addresses. Not much, something like 10-15 addresses tops. You need bad luck to hit one of those addresses when the total memory is 64GB. It's likely the null pointer check got flipped.
Browsers are good candidates to find bad memory: they eat a lot of ram, they scatter data around, they have a large chunk, and have JITs where a lot of machine code gets loaded left and right.
I think the most salient point about Factorio here is that its CPU-side native core was largely hammered out by 2018, most of the development since then has been in Lua or GPU-side. The devs could be quite confident their code didn't have any unhandled null pointers. That's not really the case for Chromium or (God help us) WebKit.
The harm is well above (even that is arguable), but the probability of getting harmed is so much lower it's not even comparable.
Alcohol? Yeah one or two too many drinks and you're in the hospital getting your stomach vacuumed. Driving? Blink and you run over a kid. Internet? You can spend evening and nights over there and not be harmed in any way.
We have full generations of kids that can now be studied about the effects of the internet, starting with millenials. I won't pretend the Internet is a better place now than it was when I was a teen, but it appears to me the "dangerous" things are more focused and concentrated (at least for kids): social networks. It's still a minefield, but with leagues between two mines.
Porn has always been the topic touted for children safety, because it's scary and resonates with conservatives and religious people. Access to is is roughly the same today than it was then, and arguably less dangerous today because the dirty stuff is hidden deeper, thus less likely to stumble upon.
But other than porn, the thing that changed the most is social networks. Addiction, bullying, etc. Facebook 15 years ago was a not serious place. The equivalent today is the best place to get roasted by fellow kids and bullied 24/7 while not being able to get off the hook. The damage is psychological, which is insidious, but not systematic. Not every kid will get bullied, not every kid will be addicted to the algorithm(tm), etc.
In the end, education plays a bigger role than simple age verification. Stimulate your kids, give them things to do other than doomscrolling, and get them on the dark corners of the internet to give them curiosity about the world and un-sanitized stuff (hacking in all forms, etc).
Heh that's already what parental controls do (granted, the website don't report the content, and it's based on blacklists), but they are trivial to bypass. Even the article mention it:
> The child can install a virtual machine, create an account on the virtual machine and set the age to 18 or over
It's precisely how I worked around the parental control my parents put on my computer when I was ~12. Get Virtualbox, get a Kubuntu ISO, and voilà! The funniest is, I did not want to access adult content, but the software had thepiratebay on its blacklist, which I did want.
In the end, I proudly showed them (look ma!), and they promptly removed the control from the computer, as you can't fight a motivated kid.
That's assuming the parental controls allow the kid to create a virtual machine. And then that the kid knows how to create a virtual machine, which is already at the level of difficulty of getting the high school senior who is already 18 to loan you their ID.
None of this stuff is ever going to be Fort Knox. Locks are for keeping honest people honest.
We could argue on the technical feasability all day, as non-kvm qemu does not need any special permission to run a VM (albeit dog slow).
I honestly don't really agree on the difficulty, as if this becomes a commonplace way to bypass such laws, you can expect tiktok to be full of videos about how to do it. People will provide already-installed VMs in a turnkey solution. It's not unlike how generations of kids playing minecraft learnt how to port forward and how to insatll VPNs for non-alleged-privacy reasons: something that was considered out of a kid's reach became a commodity.
> None of this stuff is ever going to be Fort Knox. Locks are for keeping honest people honest.
On that we agree, and it makes me sad. The gap between computer literate and illiterate will only widen a time passes. Non motivated kids will learn less, and motivated ones will get a kickstart by going around the locks.
> We could argue on the technical feasability all day, as non-kvm qemu does not need any special permission to run a VM (albeit dog slow).
That's assuming the permission is for "use of kernel-mode hardware virtualization" rather than "installation of virtualization apps".
Notice that if the kid can run arbitrary code then any of this was already a moot point because then they can already access websites in other countries that don't enforce any of this stuff.
Would that make the LLM (or the company who made it) liable under the DMCA for showing someone how to work around a digital lock that controls access to a copyrighted work.
It might be Fort Knox just fine at some point, when computers will require a cryptographically signed government certificate that you're over 18, and you can't use the computer until you provide it.
I promise there are people who can't figure out how to do it.
And again, the point of the lock on the door where you keep the porn is not to be robustly impenetrable to entry by a motivated 16 year old with a sledgehammer, it's only to make it obvious that they're not intended to go in there.
Depends on how much people want the hidden content. People in Eastern Europe, regular people, noch tech wiz kids, know how to use torrent and know about seed ratios etc. At least it was so ca 5 years ago. People can learn when the thing matters to them.
Regular people want to get things done, the tinkering is not a goal for them in itself and they gravitate to simple and convenient ways of achieving things, and don't care about abstract principles like open source or tech advantages or what they see as tinfoil hat stuff. But if they want to see their favorite TV series or movie, they will jump through hoops. Similarly for this case.
a kid who can install Linux, or set up an ssh tunnel to a seedbox, is a kid who doesn't need to be told by the government what he or she should be watching
I'd actually argue that's exactly the kid who the government is there to tell them what they shouldn't be watching. The government is never really there to restrict the incompetent, they are pretty good at doing that themselves.
There's an ocean of difference between your device changing behavior based on a flag set by individual sites and your device using a blacklist set by some list maintainer - the main difference being that the latter is utterly useless due to being an example of badness enumeration.
Conversely, an app using native toolkits (at least the Windows ones) will have better chances of running fine under Wine. I've recently had the (dis)pleasure of trying to run some .Net monstruosity with Wine, and oh my got did it not work for obscure reasons.
But overall yeah, from a compatibility perspective, nothing beats Electron. I'm not sure we'd ever get an official Discord client on Linux otherwise.
Are you really exposed to those concepts for daily Zoomer usage? I mean, you can spend your whole normie life using an Android phone never going to the file manager.
(fwiw it's been a while since iOS also have those concepts)
Sam Altman is lying by omission. It’s been confirmed [1] that OpenAI agreed to "lawful" use of the models. Since it’s the DoW, they can make pretty much anything lawful by invoking "national security".
The JSON in the article is a bit, let's say, heavy on the different objects and does not try to represent anything useful with most keys. All the things like `greaterOf`, `sum`, etc are much better expressed as keys than `{"children": [{"type": "greaterOf", ...}]}`.
Basically something that feels an reads like "freeform" yaml, yet that has an actual spec.
reply