AnotherGoodName

They are very popular amongst the elderly. Well specifically those of us setting the elderly up with a computer.

Mechanically straightforward since it’s all in one. Pretty much just the power cable, keyboard and mouse. Pretty unbreakable ux wise, reliably sleeps with little power draw and wakes up quickly without them questioning if it’s broken. Reasonably unobtrusive updates so they don’t go turning them off. Etc.

Australia has 'yous' for plural you. Yous is used exactly how you might think it'd be used.

At least you acknowledge some screen time unlike the puritans we have in this thread but i also suspect you're not in the most extreme of difficult parenting situations.

It's true and it's also true that parent's would be criticized from the sidelines for allowing kids, even from a very young age to sit in front of screens for hours on end when i was a kid.

I have a friend who didn't have a television until the age of 12. If you ask which kids turned out well adjusted and which didn't it wasn't the one's with no screen time anymore than the one's with unlimited screen time. There's some reality to life people here aren't accepting and i can't see an issue with some level of hands off parenting that this enables quite honestly.

I know people are going to attack with some extreme of "you allow your kid to be raised by a tablet" or some such but these views are all so tiresome. The reality is that majority of kids have always been allowed some form of screen time and todays screentime is going to at least be more interactive than yesterdays screentime.

A loaded question and I feel like people here including yourself are criticizing from the sidelines. Essentially the old "OMG you let your kid play on the phone!!! what a terrible parent!!!!". Although this isn't a phone, it's a device designed for kids with hopefully some apps that have a little more educational value than the typical phone apps.

The talk in this thread indicates absolutely no empathy for dual working parents and just repeats tired talking points. If anyone here looks back on their own childhood perhaps they actually had similar freedoms? Hell i used to play the Apple 2 obsessively from a very young age. I credit it as formative in a positive way.

You can argue that if you're both working and no one has 100% time to dedicate to raising the child they shouldn't have children but that argument leads to plummeting birth rates.

I say all of this as someone who doesn't have kids but I have at least some sense of how freaking hard it is and in the grand scheme of things a tablet designed for kids isn't the worst thing.

Tell me why an alternative that's at least better than a phone is so bad? If you want to argue a phone or any such device that keeps kids occupied for some time is unacceptable i think you're waaaaay out of touch.

Network hardware, especially high performance fabric is every bit as important as gpus and ram.

People seem to think Nvidia’s gpus are the be all and end all of enabling large models but there isn’t a large model out there that runs on a single gpu.

Broadcom sells things like pcie switches. They allow you to plug your in about a hundred gpus into one big switched pcie plane that can all communicate at full pcie bandwidth to each other. Nvidia has nvlink which is similar. You need this technology to enable large models and they sell for even higher margins than the gpus and ram.

And whole repositories with commit history that provides step by step training on how to code.

AI could probably write decently if every scraped book had all the revisions to create it embedded within or art had every brush stroke documented. As it stands programming is the first to be trained. The rest will take some more time.

Yes but it’s an arms race. You either give up or you use ai to compete.

>Because selling access to models will not be a good business

Let me go on a tangent to make a point. For some countries if you want a US Visa Consulate appt today they only open 3 months out and are booked constantly. They block you for 10mins on the 5th refresh.

It’s well known on crrtain communities that the only way to get one is with an ai agent that can patiently work around this and jump on the bookings with perfect timing or refresh within the bounds of the detection to nab a cancellation opening.

So we’re starting to head into a world where every single person needs an ai subscription just to deal with bureaucracy.

It’s a fucking fantastic business and i don’t understand how you could suggest it’s not good business.

You unknowingly substituted the name of another model when you’re otherwise talking about the competitor fwiw. Some if those in the reply chain didn’t even seem to notice.

That’s a little telling of the state of ai atm though isn’t it? It’s all pretty similar. The latest Opus/GPT/Gemini/Deepseek/Kimi on medium does indeed do the job

Yeah a lot of people bought into the idea of CUDA being a moat but we’re well past the point where it’s pretty trivial to jump to another equivalent stack on the software side.

Pytorch conference is happening today for explaining the coincidental timing

Might be a little controversial to say this but automated tests can be reviewed at a lower bar than production code.

I have no shame in accepting vast quantities of ai generated automated tests at review time without really paying attention to them. It’s better to spend more time on the core implementation.

So weird they allow advertising something so harmful at events with so many children is considered ok. It’s like the bad old days where cigarette companies targeted youth.

That's a very obsolete desalination technique.

A cup of water with salt added doesn't get noticeably warm since there's just not much entropy difference in salt water vs fresh water. These days you can reverse the process at very low electricity cost with reverse osmosis.

Fortunately the grid storage rollout is happening at a blistering pace too. Reported payback of investment for 2 years for many of the early rollouts (due to arbitrage between peak/off peak prices) and while that won't last it's still amazing economics.

On the flip side my nephews who don’t really code (i try to teach them but they are kids with everything that comes with that) go around stating things like ‘do you use ai lol!? It’s so bad!!’.

The ai discourse is reasonably fine on HN but as soon as you go outside of actual developer circles keep in mind that my young nephews and their ilk are posting memes and criticisms without real understanding. I hope they’ll catch on soon but for now be prepared to see a whole lot of ‘Linus using vibe code Linux going to crash lololol’ type of posts in the non-developer communities for a while.

The above 2^n is just the extreme. There’s always baby steps to that direction. If you want you can point an agent at any known nlogn algorithm (eg. Sorting) and quickly get similar improvements in a similar way. We’re about to see an explosion of ‘complexity bound of x broken!’ as people discover what the creator of complexity theory warned us about. Namely that there’s never hard limits in complexity theory and you can always speed up any time complexity with a cost in space complexity. There never was some impassable nlogn boundary for any algorithm.

Sure but 2^n space complexity to get O(1) time complexity works for all n even infinite. Thats the end state of Blums speedup theorem in either direction and it doesn’t limit to any bounds. I just stating a limited of bounds above for the sake of giving understanding. Complexity theory doesn’t care that 2^n space complexity doesn’t work when n is infinite anymore than it cares that nlogn also doesn’t work when n is infinite. It’s not the point.

It’s unintentionally making a mockery of OpenAIs results by highlighting Blums speedup theorem in action. Namely that you can always reduce time complexity with a small move into space complexity.

That unintentional mockery of a repo is 262million lines of such moves and growing. I don’t think for a second anyone involved has a clue about complexity theory though or they’d just jump to the obvious “O(1) time complexity using an O(2^n) lookup table to cover all possible n bits of input” which is the final end state of Blums speedup theorem.

I'm quite lucky in that i have the right to live and work in America, Australia, New Zealand, UK and the EU. I've also thought about this a lot as i near retirement (working in tech in the USA atm, it's been a fun ride but definitely time to think about options).

Australians don't like to acknowledge it, i guess to avoid being flooded since house prices are high, but it's the easiest pick by far. A western nation which rises as China rises due to sheer resource wealth and stable government.

Currently it's already >3x the median wealth per capita of the USA. Slightly lower on average but medians the real way to judge for the typical (non-0.01%) person.

Exports more than it imports.

Government revenues have been far ahead of spending for the past few years. Can you believe that?! A western government not just going into ever increasing government debt!

An extraordinary rollout of solar and batteries (dwarfing China's attempts at solar on a per capita basis) to the point where you can sign up to plans where power is free during peak daylight hours and the government will pay you $3k to install a home battery to store that free power.

Basically no tariffs so you can buy an electric car for USD$15k. Which goes well with the above point.

I look at the USA, UK and EU and people seem to have given up trying to vote in a better government? They don't seem to care about importing more than exporting or government spending exceeding revenue as election issues worth voting on?

Never infinite if you have bounds on integer size. For n bits of input it's always possible to have a 2^n space complexity lookup table that's O(1) speed since 2^n covers all possibilities of n bits.

You don't need to go to these extremes regardless though. For small speedups take an nlogn algorithm and just move a very small part of the inner portion out in a similar tradeoff. Eg. "Every nth input is now part of a lookup." If you look at the openAIs <nlogn results there was a series of small improvements of this type (we take this part out, precomputing it, and likewise with this part and that small portion, aha! <nlogn!). I finally loaded the link above and it's 262million lines of such tricks in fact.

The site is taking forever to load but complexity theory doesn't really have the hard walls at things like O(nlogn) that people seem to think.

You can always always tradeoff time vs space. It's actually a fundamental part of complexity theory called Blums speedup theorem by the guy who invented complexity theory no less.

As an example you give me an nlogn solution and i take a portion of the inner loop, put it in a lookup table (who cares if the lookup table is 2^n in size, that's not what we're measuring here since we're specifically measuring time not space) and suddenly i have something better than nlogn since that lookup table is O(1).

That's not to say there's some merit to all of this. It's just that people shouldn't be as astonished by these results as they are. It's complexity theory. There's literally no such thing as a hard wall where something is nlogn in time and can't get lower. It can always get lower. But you need both the time and space complexities together to reason about this.

An invitation for a semantic debate online on a minor point along with a random linked in post stating something about said debate?

No thank you says pretty much everyone.

It's the opposite isn't it? You assume consciousness in other things which may have it. Higher order animals, other humans etc.

Imagine if we actually demanded that you prove your consciousness before we believe you.

We used to schedule a ‘16yr planning’ meeting for our friday afternoon whiskey tasting and were once praised by an exec for our diligence and forethought

Local models cost more imho.

I acknowledge all the pros of local models but even if i maxxed out the 50 tokens/s over a month there’s no way I’m ever repaying the hardware costs. It’s not what local models are for.

So a strange article to talk of financial trap of tokens and then advocates for local models. A pay by the token plan capped at the same 50tok/s would always be cheaper per month than the amortised cost of the local model. Assuming you compare small model to small model.

A vpn to my home country to get around geoblocking. The best television there is the public broadcaster and free to stream but geoblocked outside Australia.

I setup an old router on the home side so switching on the vpn is just changing the wifi.

The reason to use a $3 vps rather than a dedicated vpn: the dedicated vpn services get blocked since geoblocking knows the big vpn providers, they don’t always have a config for your routers supported vpn types which ai can setup for you in seconds, the dedicated server can do a ton more for you when you need it too (eg. All the other stuff listed here).

I honestly think there’s a lot of fuzziness possible in complexity theory because of things like this. Yes you can skip some portions of a calculation and rightfully so by the current established formalisation of complexity theory but i think under another formalisation we’d probably see these nlogn^0.999999 cases become more clearly nlogn.

It was both actually.