wildmanx

An AI generates a game.

But it hasn't! This is just another step in the BS storm coming out of the latest AI hype. The language model has reproduced something that has existed before and was likely part of its training data. That's cool, but it's far from what's being claimed here.

We really need to get better at fact checking this stuff. And with "this stuff" I mean the output of LLMs and other AI frameworks as well as the claims about it. And with "we" I mean society as a whole and our industry in particular. Let's keep the hype in the drawer. The general population can be hyped up about sth, but we should know better, so instead of joining the hype, let's keep a cool head and educate people about what this is and what it isn't.

(Feb 2022)

> It hasn't aged well.

You’re wrong, it’s aged quite well.

Part of the problem is that there are many people in the field of security with overly strong opinions. This is not healthy. The field is full of know-it-all people, with if-only-people-were-not-as-dumb kind of attitudes. This is not helping anybody. Any not-as-strongly-opinionated bystander looks at this and has no clue whom to listen to, since so many people are strongly expressing 100% opposing views. Calling everybody else "dumb". This is not helpful to bring the field as a whole forward.

And sometimes that's what provides the edge over your competition.

This is a lie

Please check out what the word "lie" means and until then stop using the word.

Hint: The post you replied to did not claim something of which they know that it's not true.

I am paid to write code not sit in 10 meetings a day. If I raise the issue everyone gets all pouty.

Then it's time to move on. This is not a way mature adults should be handling such a situation.

A passive-aggressive reaction isn't mature either though, so maybe this is mutual and your company and you are actually perfectly in tune.

a concrete helmet is literally worse than useless.

Is it? It would still absorb energy upon impact. Which, in contrast to popular belief about bike helmets, is its main function. Yes, ordinary bike helmets will also just break. But in doing so, they absorb energy which then won't be absorbed by your skull.

There is no point in a perfectly stiff helmet. It would just transfer the energy right through and you'd crash your skull into the helmet, which doesn't help.

The stationed troops and machines were not nearly enough to handle the situation, see my sibling comments. You can't suppress protests of several millions with 300k not-so-willing troops and 5000 not-so-good tanks

I don't know how old you are or where you were at the time. I was there. In the GDR, in East Berlin. On the streets. And I can tell you, a few tanks and troops getting their guns out would have made major impressions on people.

It's not just a numbers game. You are greatly oversimplifying history here. Quite naively so, I might add.

It's a great achievement of history that Gorbachev made the Soviets keep their feet still and among many eastern Germans it's regarded as quite the miracle that this whole episode went down non-violently. Look around in the world in the last decades. This was the major exception, and Gorbachev was central to that.

Also, let's get the picture of the situation straight. He didn't just passively sit bunkered in in Moscow, letting things happen. He actively went out to meet leaders of other involved powers, including the German chancellor and foreign minister, Kohl and Genscher, which he outlived by a few years.

Transport them? What are you talking about? The Soviets had bases all over the place in Eastern Europe.

The case being discussed here, Eastern Germany, was just behind the iron curtain, remember? Soviet troops were at ~300 locations on the GDR territory, ~50 airfields, over 300,000 soldiers, over 4,000 tanks.

Playing devil's advocate .. it will also get the kids more disconnected from the natural environment that we're embedded in, after all. You know those statistics of so-and-so-many kids out of 100 don't know where beefsteak comes from, where milk comes from or how potatoes grow? Since all they see is the result in the store? This gets worse when the kids change from lack of knowledge to thinking plants grow in glass pots without soil.

Hydroponics may be fun for hackers. But it's a bastardized form of growing plants, that happens to work, but it's a terrible way to convey science (if not taught very carefully and embedded in a much larger picture with actual nature).

Schools should visit more farms instead.

While I appreciate the hacking potential here .. I fail to see the point beyond it. You won't save any money over buying the salad at an expensive organic store. You won't help the environment (all the plastic and heating and electricity for artificial lighting won't ever be compensated by you saving on pesticides). You won't scale this up to become self-sufficient.

Just get a garden and put the plants in soil. Sure, they may not be as "perfect". But you can have more of them to compensate. Sunlight and rain will do most of the work, use some home-made compost for nutrition, done.

So, what's all the fuzz? Feels like hydroponics, especially with all those 3D-printing gadgetry, is bringing you further away from nature, not closer.

For me, hydroponics really drove home how remarkable plants are: from a bin containing nothing but water and salts, a fully-formed plant emerges.

I feel this is a remarkably underappreciated feat of nature. Some people compare a plant seed to an SD card containing the drawings for a skyscraper. This is an unbelievable understatement. The seed does not just contain the genetic material describing the final object. It also contains information about the whole assembly process, and a very robust one in many cases. It also contains the information about all the tooling needed for the assembly process. And it also contains the initial basic tools to make all that happen. This is truly remarkable. Nobody would dream of putting an SD card into the ground and seeing a skyscraper grow. It's obvious that you need external tooling, knowledge, skill and materials (that have to be externally created from raw materials) for all of this to happen. Plant seed? Everything included.

What I find sad about this article is that it needed hydroponics for the appreciation for plants. Hydroponics is a very sterile and artificial way of growing plants, removed from their actual environment and only showcasing a single aspect of growing a plant. My appreciation for plants comes every time I step out into nature. Just take a step into a forest, you don't even need to move to see lots of wonders of nature around you. The complexity of all living things nature has to offer, from the molecular level all the way up to complex interactions in the ecosystem, that really dwarfs your k8s setup by orders of magnitude.

Tech people sometimes need to be a bit more humble. (To which this article contributes. I like that.)

If this law, all by itself, was perfectly enforced, the UK would bankrupt itself just on the extra spending in the prison system.

That conclusion ignores the interdependence of those effects. If people knew that it's going to be perfectly enforced, then almost nobody would do it anymore.

(Not that I'm a proponent of this particular law; just pointing out a flaw in your argument. Somebody wrong on the internet -- gotta reply!)

It saddens me how many smart people are working in such an unethical field.

The problem is that what's employable isn't actually computer science, but trendy industry nonsense...

You may be going a bit hard on the industry. I see where you're coming from, having a long academic past myself and seeing there is a lot of BS in industry. But there is also a lot of meaningful things in industry and academia is sometimes a little too arrogant about that.

Given that you went back to graduate school and are likely a bit frustrated with industry, I can see why you have that stance, but neither extreme is good. Neither is everything in industry just trendy nonsense nor is everything in it the holy grail. There are real problems being solved in industry. That's what produces systems that people actually use, be it mobile phones or server infra or network gear or embedded systems, which after all is the whole point of all this. There is also a lot of trendiness and lots of people who just hack things and don't have a clue. It's not black-and-white. Similar with academia. There is a real benefit to the fundamentals being taught and researched in academia. And there is a lot of nonsense there too, tons of paper produced just to produce papers and get the next grant and stay employed. (If anything, the gatekeeping in academia is way worse than in industry.)

Both worlds have their place and it's sad to see disparaging comments from either side about the other one.

Check out this paper:

https://homepages.inf.ed.ac.uk/wadler/papers/propositions-as...

Page 2 gives some background in how Turing's work fit into the problems that Church and Hilbert were about.

There are still some areas I have not covered, because they have never been of much importance for my work

I think that's a main difference. With a degree you tend to learn / get taught things upfront of which you don't know whether you'd ever need them, and you may not. You kind of trust, or hope, that what they teach you has some relevance and is useful.

Many subjects from courses I took during my degree I'd never have picked up "on the side" during my jobs that came after. But I can actually also confidently say that each and every course I took left some impressions/concepts in my head that later became useful in one way or the other. So for each course I took, had I not taken it, I'd be missing something. Of course there is opportunity cost: could I have used that time in some better way, i.e., even more efficiently? Not sure. There is also the question about the courses that I did not take because, well, there is only so much space you can cram courses into within 6 years :) but maybe there are diminishing returns as well, since many things tend to overlap.

Where do we find your podcast? http://localhost/ shows really weird contents, I probably have the wrong URL.

Sometimes I'm wondering whether this is the future, or whether we (as an industry teaching the next generation) are taking shortcuts that will hurt us in the long run.

Honest question. I don't have the answer or am insinuating anything.

What do I mean? Well, I also acquired all that knowledge, but through a formal university CS degree. Before starting my degree, I was also "just" tinkering and hacking. I had been "programming" for like a decade before that, different languages, solving some problems that I found interesting or practically useful. During my degree though, I got all the details. I had full courses on hardware (transistors and upwards, all the way to CPU architecture), on mathematical logic, on compilers, on programming language paradigms, on algorithms, on complexity theory, on formal methods, in-depth on C++, Java, Prolog, computer networks, databases, cryptography, real-time systems, statistics, and a few more. This also allows me to in-depth describe a "nand to tetris" process, but obviously my education has been much broader than that.

Was it necessary to take that 6-year degree? Would that have been equally well covered by a bunch of Coursera online classes and then a few years of work experience? I really don't know.

I have a small sample set of colleagues who are self-taught and though ambitious are sometimes lacking broader perspective and in-depth understanding of CS foundations (that includes my boss), and of colleagues who went through a similar degree and do have a similar perspective as me, some of them not really remembering what was taught (or not caring), so those samples go both ways. I don't really know anybody without a formal degree who has the full fundamentals groked though. That would let me to conclude that the degree is a good thing, but frankly my sample is too small to tell.

A solution is to have somebody in each FE team with a "that's what the masses use" device. Screen-wise, CPU-wise, RAM-wise, network b/w wise. And let them use it as a primary dev device.

Doesn't sound sexy, no. Compensate them royaly. And make clear that they are superstars just by having this work environment. Then keep listening to what annoys them with whatever you are building, and let them fix it.

GP may have meant that this is useful outside of bisecting something in the main branch.

For example, you may want to get your tree in one of the intermittent states and play around with it. Run some tests, try how some object behaves, to better understand the different commits comprising the PR. I do this quite a lot.

since 7 doesn't increase cleanly, I probably have to custom-code the final update to 100 manually,

What do you mean with "not cleanly"?

bar.update_smooth(i * 100 / 7)

Once i is 7, this will be equal to 100. No matter your rounding mode. You don't need a CS degree for that formula.

I've used quoted searches now and then in the past, as recently as last week or so, and it definitely did not work. There were no usable results, none of the results shown did contain the exact words. Is there a bug somewhere?

Hm? The nukes dropped in august 1945. Nazi Germany had long surrendered by then. In part because the Red Army had come all the way to Berlin.

Unfortunately, having a big name from MIT on your grant application as co-applicant significantly increases your chances for funding. You get the money and do the majority of the work, they get to claim disproportionately much credit for the work. Win-win? (Doesn't sound like it, but that's how it works..)

Exactly! What we are discussing here is an ad-hoc solution. That $4 device will work for how long? A week? A month? Then we're up for the next one. A continuous stream of aid is necessary to keep this solution going. So, it's not sustainable, assuming we don't consider permanent aid a sustainable solution. (I don't.)

That's not an answer to my question.

I meant, how sustainable is it to keep depending on foreign aid? Obviously dirty surface water that does not suffice for the whole population is not sustainable. But in the long run, creating new dependencies on aid can't be the solution. Any long-term plan that involves foreign aid is a bad plan.

There are not many people around anymore who have witnessed it and can tell the difference.

And it's a good example of my point. Once thing that's for sure is that the relative importance of "western allies" vs. USSR in their role of ending WW2 does not match reality, or what you'd have perceived on the ground. USSR had over 10 million dead soldiers (and the same in addition in civilian deaths). That's about a factor of 1 compared to the western allies in the European theatre. The eastern front was huge, bloody, deadly, and USSRs role in fighting back and steamrolling over German troops into Berlin was crucial in ending all that. The popular narrative nowadays is that the US ended the war, most movies and war documentaries are about the western front, etc.

Of course Stalin was a dictator and what happened during the cold war on the eastern side of the iron curtain and what's happening now in and from Russia is terrible and unacceptable. But without having witnessed events as they unfolded, it's much easier to be buy into today's narrative of the situation back then, which is heavily skewed.

The market is tapped. Just not by the U.S. / western countries.

Indeed. The article links to the MIT News article https://news.mit.edu/2022/solar-desalination-system-inexpens...

The three involved researchers from MIT are grad student Lenan Zhang, postdoc Xiangyu Li, and professor Evelyn Wang and "four others" which are not mentioned by name in the beginning of the article.

At the bottom of the article, it's mentioned that Zhenyuan Xu from Shanghai Jiao Tong University in China is one of those "four others".