vzcx

Have you found LLMs to be good at metalinguistic abstraction? I have used them a little bit for lisp programming but never once have I seen them introduce macro.

Hardcaml compiles to verilog and then you use the horrendous vendor toolchains.

Nice solution and good easter egg find!

You can give the cell instances a stable name by setting GDS property 98, which I learned about from my reconnaissance of the puzzle author's github and sky130 visualization tool. This way I was able to spot check a pass over the netlist that broke up the regions into a hierarchical design.

I'd like to do a full writeup but haven't had the time.

This would need good image recognition, but maybe not so far out of the realm of possibility.

These GDS design files have a lot more structure to them.

Incredible amount of determination, but you really did make it hard for yourself!

You can install librelane to get the whole open silicon tool suite and the sky130 PDK. Circuit extraction can be done with magic. Going from a spice netlist to verilog netlist is pretty mechanical and not a hard transform to write. You almost immediately have something that can be simulated and a good baseline for further reversing.

LTCM was actually a lot more sophisticated, this fund's thesis was basically "number go up."

Count the OOMs!

As an American, I trust Chinese AI providers more than American.

I know your comment is about the absurdity of such a thing, and maybe the stupidity of the administration, but for the readers at home:

Fentanyl is actually a very simple molecule. The way it's (and other drugs) production is controlled is by controlling and monitoring the precursor chemicals. So if someone goes out and buys a lot of proprionyl chloride, they will pop up on the DEA's list.

This is a much more sane thing to do than hemming and hawing about how dangerous the AI chatbot is and placing yourself into a position superiority over your users, pretending you know better than them, assuming that you must protect them from certain knowledge for their own good, etc, etc.

> AI is dual use technology.

And? Computers are dual-use. Cars are dual-use. Telephones are dual-use. Freeze-dried chicken is dual-use.

Single-use, i.e. military only technology is actually pretty rare.

> This kind of posture is simply not tenable as frontier intelligence increases.

I reject the corpo speak that tries to brand these things as being "intelligent." They can be useful. But a language model cannot conjure a weapons platform from the ether no matter how "intelligent" it is.

Having a chatbot that talks to you about synthetic biology or nuclear physics is just not the same as being equipped to develop biological weapons or atomic bombs.

None of this will happen in the "neighbors backyard." You are exaggerating the threats to "democracy" while simultaneously invoking democracy to limit freedom of information. The suggestion that somehow the bad guys will get nukes if we let people access information is just absurd.

Society at large is not concerned about whether someone asks the chatbot about organic chemistry. They are concerned that they will be de-facto forced to interact with some shitty automated system to get by in life, like having to pass an AI-powered ATS to get a job.

They are tired of the hype and tired of idiots like Amodei being elevated to heights of power and influence. They are concerned that the things they love are being devalued. But they don't give a fuck if I ask an AI about genetically modifying viruses. This is a pet issue among some of the AI safety crowd.

So, yes, I am 100% fine with PRC-backed labs distilling Anthropic's models. I do not care about Anthropic. They have demonstrated that they are not on my side, and that they are at best ambivalent about actually empowering their users. I'm not a fan of the PRC either, but their distance makes them far less of a threat to me than companies like Anthropic and my own government.

Or... they just disagree with Anthropic's ethical stances and approach to applying them?

The best conmen have an abundance confidence in themselves.

Lisp is a family of languages. They share a common ancestry. That family is larger than Common Lisp, although Common Lisp is might be the last living descendant. While there are common traits among members of the family (parenthesized notation, certain special forms, symbol names), the presence of these traits does not itself define the family any more than having fins implies that an animal must be a member of the fish family.

A language is not only its syntax, but also it's semantics. Lisp is not just parentheses or even 'homoiconicity.' In addition to the surface syntax, there is a shared cultural understanding of what certain symbols mean, how they fit together into a system, what a symbol even is, and so on.

Given that, while there is some surface resemblance to lisp, it's plainly obvious to me that Clojure does not belong in that family. This is not a dig at Clojure! Clojure makes a lot of people happy and productive, and that's great. It's my third favorite language on the JVM. But when you take a closer look at some very core design decisions made in Clojure you'll find that they depart strongly enough from all the other languages in the Lisp family that it belongs to a distinct language family. A few of them:

- Reader doesn't intern symbols

- EQUAL symbols aren't EQ

- Introduction of vars

- Cons doesn't cons

- No numeric tower

- Backquote behavior

These things don't necessarily come up in day-to-day programming, so it might seem like quibbling, but these are fairly central differences in meaning.

I think it's sensible to consider "Clojures" as a distinct language family. Rhere are languages like Fennel and Hy that are clearly Clojures, and I'm sure there will be more.

Since there are other decent dynlangs out there, the benefits are less about the surface language aspect, and more about the process of getting to a solution. These are the system aspects of lisp that enable an incremental development workflow with a lightning-fast feedback loop. While there are important language features that support this, it's ultimately a "whole is more than the sum of its parts" effect.

I'd recommend checking out videos of people developing systems in lisp. Off the top of my head:

- https://www.youtube.com/@CBaggers/playlists

- https://www.youtube.com/playlist?list=PLTA6M4yZF0MzsMlNL0N67...

If you are interested in machine learning, check out Gabor Melis's library: https://github.com/melisgl/mgl. It's not an area I'm super familiar with, so I can't speak to it's feature set, but I believe he used it to win a machine learning competition some years ago.

I don't think anyone's written a transformer or diffusion model with it, could be a fun challenge.

lmao, brilliant scientist Elon Musk is NOT. A closer comparison would be general Groves, someone who can get the team and resources in place so the work can get done.

Grammarly used SBCL in production:

https://www.grammarly.com/blog/engineering/running-lisp-in-p...

It also sounds like they were very much a polyglot shop.

I don't think that's at question in this case. It was decided in Massachusetts v. EPA that EPA is required to make that determination itself. Three years after that decision, EPA basically answered "we think it is."

The court isn't bringing any of that up in this opinion. They are looking only at whether Congress specifically delegated EPA the power to regulate emissions by means of "generation shifting". The majority opinion was "no, they didn't."

Excuse me, sir, flamebait is classified as a destructive device. I'm going to need to see the NFA tax stamp for this post.

Isn't it a good thing that armed citizens and the professional warrior caste like police and military -- who are also really just other citizens -- work together?

If you are asking why the glorious revolution hasn't happened despite having an armed to the teeth peasantry, then the answer is simple: the peasants are still getting their bread and circuses.

yes.jpg ?

We are armed and dangerous, we we want to be. Not to you of course, but to anyone who threatens to cause you death or great bodily harm.

How about a thank you for being prepared?

Right, right. Guns are absolutely about power. What is interesting to me about this difference between us is that I really, really want people to have that power: I think it is very important that regular people have this "violent potential" as you put it. Having the potential is obviously not the same as deploying it, but having it still matters very much.

Most people are good, trustworthy, and deserving of this power. A few people are not, but taking power away from regular people really won't do so much to help them. In fact, at the end of the day, all it does is force them to become like children, dependent on someone else -- or really, a special class of person, since people absolutely should work together -- for their own security.

"Society" is really nothing more than these people: those that choose "cooperate". It's not the same thing as the "political system" or even the "rule or law". Some people choose "defect", and view the cooperators as prey. Correct me if I misunderstand, but it seems to me, that out of fear of defectors, you want to take power from cooperators and give it to a sanctioned, "regulated" class of people. Presumably because you think you would be safer.

I think you can imagine why this would be ill-considered. If you are concerned about "equality", it's hard for me to imagine a more unequal state. We've basically arrived again at a medieval distinction between peasants, warriors, and priests. Who keeps your warriors in check? The unarmed peasants? The priests with their "regulations"? I'm guessing you are more of a priestly type. I would caution you, your magic is not as strong as you think it is, and it doesn't seem to work at all on the bandits.

It's a bad idea to dox gun owners, but I personally would be very pleased to learn that my neighbors are well armed. I already know a lot of them are. Perhaps you would not be. But if not, consider that you might need better neighbors or maybe a better relationship with them.

It can't even do that for long, as the difficulty adjusts downward to compensate.

Tether has the veneer of bank equity from the balance sheet "attestations" it publishes, but I think a better analogy would be that tether issues casino chips for the decentralized casino that encompasses the tether-based offshore exchanges and defi universe.

There isn't any real visibility into the balance sheet, but I'd guess most of the "collateral" boils down to margin loans issued at the exchanges, and since the house has the edge on those platforms, their ship stays afloat.

Make Калининград Königsberg Again.

I read this thread and found it interesting, but the mafia boss argument seems weaker than the simple desire of all businesses to have commoditized inputs and avoid vendor lock-in. The oil and gas guys who are friends with the boss are politically secure, do they really need to worry about usurpers? This poor stifled engineering nerd vs. big dumb mafia jock narrative seems a little too indulgent of the audience's biases...

Much crazier things have happened than revitalization of national industry. Whether their leadership has the intelligence and will to do this is of course another matter.

Honest question: if you are nvidia, why not publish the interface specification for your device? If there are silicon bugs, publish the errata and the workaround. I've seen SOC manufacturers do similar. What's different about GPUs?

Well this is interesting: my perspective is that we are actually already living in the age of the bureau and the technocrat, with only some vestigial democratic traditions, and we have been since 20th century wars. Burnham (who is well worth reading, in my opinion) called this the "managerial system". Of course, it still calls itself "democracy" despite not really being democratic at all. In fact, whenever you hear the word "democracy" -- a word which tends to have positive valence in society -- in the official press, you can substitute "managerial system" for a more neutral term without loss of meaning.

That age might be coming to an end, but it's less about some vague general mood of the 2020s and mostly due to proliferated networked communications. This allows people to find their frens exchange ideas and coordinate actions. Oh and everyone starts to see how transparently duplicitous the official lines are.

We perceive all this as "populism." From Arab Springs to meme stocks to Donald Trump, the phenomena are clearly a form of network dynamics. Is it bad? Is it good? Well... it's certainly disruptive, and is probably both good and bad in different cases and at different times. Which are which depends on your point of view.

But disruption of any kind is certainly bad for the managers, so the natural reaction is precisely what you suggest, effectively, sinification: tighter control ("moderation") of online communications, suppression of opposition parties. Policy drafted by career mandarins. The CCP has done alright, so I suppose if (big if) they can execute it well it might buy them some more time.

But unless you are a party insider, I'm not sure why you would throw your lot in with them.

I recently sent a low-fee, nearly instantaneous payment over the lightning network with exactly zero trusted intermediaries. LN may or may not be ready for wider-scale adoption, but it was incredibly cool. It gave me that same feeling of power and possibilities as when I first learned to program a computer. I'm not sure if I can convince you but there is actually some there there.

If, as I predict, volatility goes down as the network grows and bitcoin becomes used for more things, so does average return. In other words, the "deflation" rate reaches some small but steady level. The days of number go up will be over, bitcoin would be something boring that goes up only slowly over decades. Like the foreign exchange of a virtual country, its market price vs other currencies would be determined more by balance of trade than by currency speculation.

Of course, that future is far from certain. Buying bitcoin now is like buying an option that pays out if that future comes to pass. The price of an option is very volatile, but it is never zero until it expires. Bitcoin might wax or wane in popularity, but it will never expire.

This is all that the words "speculative asset" entail: they are risks that work out in some possible futures and don't in others. So it doesn't bother me that some people bought bitcoin and $10 and are now sitting on a nice stack. That bitcoin could have easily gone to 0.01 today. They took that risk. Why shouldn't they reap that reward? Without those early market participants in bitcoin, wouldn't be even more highly concentrated? My advice to such whales would be to spend their bitcoin on business investments that help spread the coins and bootstrap the network. Bitcoin-related businesses that pay employees in bitcoin. There are still bold people out there who want to live in an alternate future and try to make it real.

Of course, I don't think everyone should get into speculation... It's got a pretty high attrition rate even among the professionals. But you know, oddly enough, it's when inflation rates are higher that the more people are forced to be speculators just to stay in place.

Being on the cutting edge means that most of the time, what you are doing just doesn't work.

I've been listening to old timers tell stories about working at MOS technology back in the 70s. MOS was rather vertically integrated, with chip design and fabrication feeding all the way to consumer products. So chip designers dealt with buggy manufacturing processes, board engineers got buggy new chips hot off the wafer, and software guys got buggy dev boards. A lot of time was spent trying to go back and fix bugs at the lower layers, sometimes all the way down to manufacturing defects. It might sound crazy today, but for a time they could do things that very few other companies could.

"It's slow", "it's buggy", "it's inefficient"... all of these are ways of saying "I can't, I won't". Expressions of fatigue. They don't lead anywhere.

"I'll make it fast," "I'll make it work," "I'll make it efficient"... all are expressions of will. This is the attitude that the builds the future.