Hacker Newsnew | past | comments | ask | show | jobs | submit | dmitrygr's commentslogin

"No CarPlay" is approximately where I stopped reading -- and I am literally their dream target audience - enough money to buy a second car, easily able to afford one of theirs, have solar on top of my house making my electricity cheap, like techy things.

They will either learn or ... not, I guess. I know I am not the only one. Nobody in my family would buy anything techy without my advice (and a modern EV is basically an iPad on wheels, so it qualifies as techy), and I will never ever give "yes" to a car without CarPlay


yeah, seems extremely weird to not have it, it is a must for newer cars, especially ones that are EVs. I wonder why they don't have it, maybe a beef with the phone makers...

Who implements these idiotic policies? We do! Politicians could not code their way out of a paper bag! Giving up is not the solution. Refuse to do it. Make ID pass for a full-white jpeg.

I hereby remind you of a bet you lost: https://news.ycombinator.com/item?id=39186555 :)

my contact info is in my profile to arrange settlement


Because that is where the users with the money are.

Any real estate agent will tell you what people actually want. Ask one who's been around for a while, in an unofficial setting over drinks. Ask what questions people ask. Ask what they follow up on. Really dig into it. You'll realize that while nobody likes commuting, a commute is the price one pays for those other things people want that you'll hear about over those drinks. It'll give you a lot to think about, i promise.

This is 1000% the truth - nobody passes up a house because there's a similar one further away.

> I wish we had a way to build a durable chain of custody into these technologies

Do you? Consider for a moment all the dissidents and protestors who would be ensnared by their own devices then, with no "it was all ai" defense available?


I do - I think the videos and pictures that protestors smuggle out become less powerful if the state can dismiss them as fake and while most of us will remain skeptical of authority the more easy it is to fake something the more people you can convince of your falsehood.

I don't think the lack of a durable chain of custody really provides any protection - that protection needs to come from a strong legal system and social contract to protect whistleblowers. If you're thinking of, as an example, an Iranian smuggling out protest footage, they're already taking an extreme risk and have a state using numerous tools to try and track them down - but the lack of a durable chain gives a wide area of authorities to cast doubt on the truth.

I think your question is interesting to ponder and I think there are arguments in both directions - but my mind keeps coming back to the tank man photo being smuggled out of China and how much more difficult it would be in the modern world for a single image to carry such weight.


> I don't think the lack of a durable chain of custody really provides any protection - that protection needs to come from a strong legal system and social contract to protect whistleblowers.

That social contract is quite a bit of a hit&miss if you look at countries across the globe. Same for the strong legal system. Other concerns aside, does this not make the whole approach a non-starter?


I don't think it does but I do see the counter arguments. There have been prominent publicly open political dissidents and they do often suffer from assassination attempts. I think if you're considering political dissent the potential cost is a major factor in that decision. I have not had to make that decision personally so I am not an expert here - but it might be important to consider the value to those people of knowing your evidence can be proven true no matter what the authorities say.

I don't think the lack of a durable chain of custody really provides any protection

creating a fake video about a protest is going to upset those that want to hide the protest just as much.


> The generated code is not very efficient. Even with all optimizations enabled, it outputs less efficient code than GCC with all optimizations disabled.

Worse than "-O0" takes skill...

So then, it produced something much worse than tcc (which is better than gcc -O0), an equivalent of which one man can produce in under two weeks. So even all those tokens and dollars did not equal one man's week of work.

Except the one man might explain such arbitrary and shitty code as this:

https://github.com/anthropics/claudes-c-compiler/blob/main/s...

why x9? who knows?!

Oh god the more i look at this code the happier I get. I can already feel the contracts coming to fix LLM slop like this when any company who takes this seriously needs it maintained and cannot...


I'm trying to recall a quote. Some war where all defeats were censored in the news, possibly Paris was losing to someone. It was something along the lines of "I can't help but notice how our great victories keep getting closer to home".

Last year I tried using an LLM to make a joke language, I couldn't even compile the compiler the source code was so bad. Before Christmas, same joke language, a previous version of Claude gave me something that worked. I wouldn't call it "good", it was a joke language, but it did work.

So it sucks at writing a compiler? Yay. The gloriously indefatigable human mind wins another battle against the mediocre AI, but I can't help but notice how the battles keep getting closer to home.


> but I can't help but notice how the battles keep getting closer to home

This has been true for all of (known) human history. I’m gonna go ahead and make another bold prediction: tech will keep getting better.

The issue with this blog post is it’s mostly marketing.


Can one man really make a C compiler in one week that can compile linux, sqlite, etc.?

Maybe I'm underestimating the simplicity of the C language, but that doesn't sound very plausible to me.


yes, if you do not care to optimize, yes. source: done it

I would love to see the commit log on this.

Implementing just enough to conform to a language is not as difficult as it seems. Making it fast is hard.

did this before i knew how to git, back in college. target was ARMv5

Great. Did your compiler support three different architectures (four, if you include x86 in addition to x86-64) and compile and pass the test suite for all of this software?

> Projects that compile and pass their test suites include PostgreSQL (all 237 regression tests), SQLite, QuickJS, zlib, Lua, libsodium, libpng, jq, libjpeg-turbo, mbedTLS, libuv, Redis, libffi, musl, TCC, and DOOM — all using the fully standalone assembler and linker with no external toolchain. Over 150 additional projects have also been built successfully, including FFmpeg (all 7331 FATE checkasm tests on x86-64 and AArch64), GNU coreutils, Busybox, CPython, QEMU, and LuaJIT.

Writing a C compiler is not that difficult, I agree. Writing a C compiler that can compile a significant amount of real software across multiple architectures? That's significantly more non-trivial.


Frankly, I think you are exaggerating. My university had a course that required students to build a C compiler that could run the C subset of SPECint (which includes frigging Perl) and this was the usual 3 month class that was not expected to fill in 24h of your time, so I'd say 1 week sounds perfectly reasonable for someone already familiar. Good enough C for a shitton of projects is barely more complicated than writing an assembler, in fact, that is one of C's strong points (which is also the source of most of its weaknesses).

I really, really don't think so, but you're welcome to try :-)

> I can already feel the contracts coming to fix LLM slop

First, the agents will attempt to fix issues on their own. Most easy problems will be fixed or worked-around in this manner. The hard problems will require a deeper causal model of how things work. For these, the agents will give up. But, the code-base has evolved to a point where no-one understands whats going on including the agents and its human handlers. Expect your phone to ring at that point, and prepare to ask for a ransom.


Claude is only a few years old so we should compare it to a 3 year old human's C compiler

Claude requires many lifetimes worth of data to "learn". Evolution aside humans don't require much data to learn, and our learning happens in real-time in response to our environment.

Train Claude without the programming dataset and give it a dozen of the best programming books, it'll have no chance of writing a compiler. Do the same for a human with an interest in learning to program and there's a good chance.


Claude contains the entire wisdom of the internet, such as it is.

> I can already feel the contracts coming to fix LLM slop like this when any company who takes this seriously needs it maintained and cannot

Honest question, do you think it’d be easier to fix or rewrite from scratch? With domains I’m intimately familiar with, I’ve come very close to simply throwing the LLM code out after using it to establish some key test cases.


Rewrite is what I’ve been doing so far in such cases. Takes fewer hours

  >Basically PIO smells like a wart to me. I genuinely don't know who wants it. Regular hackers aren't sophisticated enough to use it productively and the snobby nerds have better options.
what are you blathering on about, sir?

driving complex displays with no spu use: https://dmitry.gr/?r=06.%20Thoughts&proj=09.ComplexPioMachin... (my work)

pretending to be memory stick and sd card at dozens of mhz as a slave to a sync bus (my work)

ethernet: https://github.com/kingyoPiyo/Pico-10BASE-T (not my work)

68k bus slave (my work)

usb host https://github.com/sekigon-gonnoc/Pico-PIO-USB (not my work)

all on a $1 chip


> what are you blathering on about, sir?

Please don't.

I mean, I applaud your work. But let's also be honest (in the "tough love" sense): those are all toys with significant limitations that preclude anyone shipping any of them on an actual device to an actual consumer. I mean, your SOC (maybe a $2 one) surely already includes a SPI master and USB host!

Actual interconnects that solve real market problems have big boring spec books and competing implementations and silicon vendors. The application for PIO is basically limited to "I have to connect to this crazy old junk and no one makes the part I'd otherwise need".


Having dealt with the errata sheets for microcontrollers with all those fancy IO devices that solve real marketing problems etc, I'd kill to fix those problems with a software upgrade.

if you find a SoC for $1 that has 2 Ethernet ports, and a usb host on it, while also having two cores and supporting 32MB of RAM you'll surprise me. rp2350 does all of the above for $1

Have you tested out the 2x PSRAM configuration; if so, have you written anything up about it? :) I've thought about a configuration like that myself but haven't committed to any hardware yet.

yes i have tested it and no i have not written about it (yet). it does work and with the new ISSI 16MB chips, it gives 32MB of continuous memory!

Awesome - I might just have to try it.

How are you handling startup? The approach I had in mind was putting boot flash on the second channel with a separate CS pin from PSRAM and configuring that in OTP; any idea if that would work?


Yes, sort of. Basically a tiny loader in OTP that enables nCS1 on some pin, copies code from there to ram on nCS0, reconfigures nCS1 to point to second ram. This does break usb flashing. Use swd.

Is there no way to do that without code in OTP? The bootrom is supposed to scan for an IMAGE_DEF on both memory banks, and can use nCS1 as specified in FLASH_DEVINFO. Unless there's something I'm missing, that should be sufficient.

I have not disassembled the production RP2350 rom (last i looked at it was ~year before release) so i do not know if it does indeed check nCS1 address space, but even if it did, which of the 4 possible pins would it configure as nCS1? surely trying all would be dangerous (possibly breaking customer logic attached to this pins that does not expect them to wiggle at boot). This seems like it could break customer designs so I would be worried if this did happen. I could be wrong. You can read out the rom with a debugger and look, of course :)

FLASH_DEVINFO (in OTP) has fields to set how large M1 is and what nCS pin to use for it. It's off by default.

Yes, you still have to write to OTP - but a few bits of configuration is a lot easier than an entire init routine.


ok, sorry, was trying to reply with regard to "no otp" as i understood your comment. i see i misunderstood. then yes it'll work

the solution is, of course geometric mean. 1KB = sqrt(1000 * 1024) B

>When people complain about protesters getting in the way and being noisy and generally being inconvenient, they are bemoaning effective protest. That's a constitutional right.

Since you did not exclude it, i will assume that by "being inconvenient" you mean all sorts of things done nowadays as parts of protests, like blocking roads and such... Thing is, it is not nearly as clear cut as you might think.

> Congress shall make no law respecting an establishment of religion, or prohibiting the free exercise thereof; or abridging the freedom of speech, or of the press; or the right of the people peaceably to assemble, and to petition the Government for a redress of grievances.

There has not yet been a SCOTUS test on whether impeding others' free movement is considered peaceably assembling. I expect we'll see such a test soon. You indeed do have a right to petition government, and assemble peaceably, it is not clear that you have a right to inconvenience unrelated persons whose only fault is living in the same town and trying to get to school to pick up their kids while you block a road.

Inconveniencing unrelated persons is not nearly as clearly legal as you seem to claim it to be.


> Inconveniencing unrelated persons is not nearly as clearly legal as you seem to claim it to be.

It seemed to be OK for guests at diners in the 60's. But - I'm sure that's different somehow. It always is.


lack of enforcement != legality

Guidelines | FAQ | Lists | API | Security | Legal | Apply to YC | Contact

Search: