> I like that it begins as Asteroids, turns into a countdown race, borrows ideas from Nintendo, Bomb Jack and Pang, and then actually ends after sixteen waves that can be learnt by heart
The thing about using AI to write articles for you is, because it's tuned to be enthusiastic about everything, it doesn't know what's boring and uninteresting. Shameful.
If you wrote this, apologies. Blame Pangram in having high confidence that most of this is AI, my spidey sense going off because it sounded strongly like AI to me before I checked Pangram, and most importantly my lack of imagination in what people finding interesting.
No worries at all. It's a wild time we are living in, so it's good to be sceptical and question everything. I always try to keep in mind that at the other end of the wire is (usually!) another human.
It's a game design postmortem; probably most interesting to game developers and designers and/or people who have played the game it is describing. I imagine that it's not very readable if you've not played the game, or have no reference to game design.
I find this quite amusing, because my blog turned 30 years old last week: it was started in 1996 which is 27 BC (Before Claude).
I'm sorry that you find the topics uninteresting, the rate of my output challenging, or my enthusiasm for my own work grating, but I'll keep writing anyway.
My goal is to simply write about the stuff I am doing, creating, solving…because it's fun! I don't expect anybody to clap, but you are correct that I am incredible—so thanks for the compliment!
Author here. It leaves me despondent that I can plough so much time into a project, a game, and a blog post, have someone else find it interesting enough to submit to HN, and then see the first response dismiss it as LLM-generated text—from someone who admits they didn’t even read it. Believe it or not, it wasn’t.
You need a static IP address for this to work is the downside, and depending on where you live and who your provider it it can be difficult and/or expensive.
You can programmatically update DNS whenever your dynamic IP changes. One issue though is that some residential ISPs prohibit webhosting in their terms.
you can go ipv6 only, any good isp will give you a static /56 for free. practically none of your users have ipv4 only devices when every major os has been dual stack by default for like 15 years. if your isp cant give you one its time to switch as soon as you can.
You can get a Hurricane Electric free static v6 tunnel.
They have poor reputations and are blocked from streaming sites and so on. But when you're the server, that doesn't affect you.
Note that you need a static v4 and DMZ because the tunnel protocol is a very simple one - presumably because they run it on giant routers. It just puts a v4 header in front of the v6 header. No TCP or UDP.
Ehh my ISP at home is still ipv4 only. The amount of ipv6 capable connections only just passed 50% worldwide a few months ago.
I don't think ipv6 only is feasible yet unless your audience is exclusively in Asia where ipv6 uptake is much higher due to them running out of ipv4 years ago
Are you claiming that privacy can never be a prerequisite for freedom and/or justice?
It's trivially easy to see cases where freedom+justice+innovation can conflict with each other (it's even trivially easy to see where they can conflict with each other specifically for innovations involving the reduction of privacy, ye olde panopticon.)
So it's also trivially simple to understand that at some point you're gonna have to pick one over another. And note that freedom is the first word in that list.
> The EFF's mission statement supports them prioritizing freedom and innovation over privacy.
What about the sentence right before the mission statement? [1]
> The Electronic Frontier Foundation is the leading nonprofit organization defending civil liberties in the digital world. Founded in 1990, EFF champions user privacy, free expression, and innovation through impact litigation, policy analysis, grassroots activism, and technology development. EFF's mission is to ensure that technology supports freedom, justice, and innovation for all people of the world.
> The EFF's mission statement supports them prioritizing freedom and innovation over privacy.
For as long as I've known about EFF (which is less than a decade, I admit), EFF has never seemed to prioritize "innovation" over privacy. As for freedom, privacy often is a prerequisite.
But you could. There is no law of the universe that is going to stop you. Words are something randomly made up by humans.
> it's just plain wrong.
Again, words are completely made up, so it can't really be wrong in the traditional mathematical sense. It could be misinterpreted, perhaps. Of course that is dependent on how you've chosen to randomly make up "wrong".
Dwarkesh Patel has AI/ML guests on his podcast. BoorishBears may have been referring to the Jensen Huang episode where they discuss TPUs: https://youtu.be/Hrbq66XqtCo?t=982
Specifically Dwarkesh couldn't understand that GPUs are not enough: it's GPUs plus multiple ecosystems to leverage them at massive scale during training vs inference.
Instead of giving China open access to US controlled chips and creating a misalignment between labs that want to train a model on whatever is best, and hardware manufacturers that need labs to suffer the growing pains for their new ecosystems built from scratch... we removed the option from the board and now they've beat the growing pains decisively, with a speed that reflects the non-optionality.
I don't listen to Dwarkesh but I'm aware of who is and his influence. I was baffled that he could not understand it...Don't know if he had his own agenda or just not intelligent (which is scarey for someone with influencec), but I sensed the frustration in Jensen Huang for something that is fairly obviously.
The same scenario happens all the time when the US takes away something from China and China doubles down, gets into survival mode and then beats the US.
The Chinese ecosystem has not caught up; in fact, it's falling further behind, due to export restrictions on semiconductor manufacturing equipment. Even if America sold China all the chips Nvidia wants to, the CCP would still develop chips as quickly as possible as a matter of supply chain security.
While some years might pass until they will really catch up, that does not prevent them to find workarounds for their weaknesses.
For example, they recently have demonstrated a supercomputer faster than any of the US supercomputers.
Unlike the recent European supercomputers, which like the US supercomputers have been built by buying racks from HPE Cray, because China was not allowed to buy such things they have developed their own custom CPUs, designed in China, which have surpassed in throughput the AMD GPUs used in the fastest US supercomputers.
The Chinese CPUs match in memory bandwidth per socket the latest AMD MI355X GPUs (8 TB/s), while being significantly faster than the older AMD GPUs installed in the US supercomputers.
While the purpose of the new Chinese supercomputer is mainly for scientific/technical computing tasks that need high FP64 throughput, the CPUs used in it also have high enough BF16/INT8 performance and memory bandwidth and interconnection bandwidth (1.6 Tb/s directly from each CPU socket) to be able to train any big LLM.
So the evidence does not show China falling behind, but at least in certain directions they are already exceeding the performance of what they have been forbidden to buy.
For something like training a big LLM, the only disadvantage of the current Chinese devices is a lower energy efficiency, of only 65% to 70% of that of the best NVIDIA GPUs.
However that is not really a problem for China, as they have abundant cheap energy.
Moving forward requires two parties with two very different financial incentives to cooperate in a way that harms the incentive of the other: the manufactures making the chips, and the companies using the chips to build AI.
Even if the Chinese government tells them to prioritize homegrown solutions, the A effort and players at labs are going to be on pushing the frontier, and the B effort/players goes to solving the teething problems
Asking adults for beer doesn't scale, code sharing can. If you want to crack down on code sharing, you'd have to start surveilling who is signing up to what.
> If you want to crack down on code sharing, you'd have to start surveilling who is signing up to what.
Not at all. If you want to find who shares codes at scale, you just track that person. If someone sells hundreds of thousands of codes online, you can just buy one from them and you have proven that they do it.
Again, if age verification is honestly about helping the kids, I don't think it has to be 100% efficient.
> If you want to crack down on code sharing, you'd have to start surveilling who is signing up to what.
Why? One code for one user account per site. If you're paranoid about privacy rotate codes and accounts weekly. As long as you can purchase the codes with cash in IRL stores the privacy impact is minimal.
So this solution is fine for proving your immigration status, getting employment or renting a house, but it's not good enough for browsing porn(to be a tiny bit flippant)?
>>If you want to crack down on code sharing
Right now, all the kid has to do is grab their parents passport while they are not home or asleep, scan it on their phone and they are in. It takes 30 seconds.
With the codes they would either need to convince their parents to generate a code for them, or find someone online who will - which of these solutions seems less prone to abuse to you?
Again, let's not let perfect be the enemy of the good.
