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

However, unless you are a super-programmer, it's very easy to introduce subtle bugs. Software I write has hit this occasionally, someone somewhere does something like cast an int to bytes to do some bit-twiddling. Checking your whole codebase for this is incredibly hard.

My modern choice is just to make clear to BE users I don't support them, and while I will accept patches I'll make no attempt to bugfix for them, because every time I try to get a BE VM running a modern linux it takes a whole afternoon.


You are correct, honestly, I couldn't disagree more with th article. At this point I can't imagine why it's important.

It's also increasingly hard to test. Particularly when you have large expensive testsuites which run incredibly slowly on this simulated machines.


Static code. Also, very fine details of the machine because less important.

When emulating a NES (for example), you really have to emulate every register, how registers change, and also weird effects like instructions that take longer to read or write values, as games rely on that stuff. Once you have modern systems where much of the code was originally C, it becomes less important to ensure every register has exactly the right values when a subroutine finishes in most cases, you can rely that (most) of the code follows standard calling conventions.


You really don't. Most of the time the code will be doing normal things. So you can have other code that also does the same thing.

>and also weird effects like instructions that take longer to read or write values

The percentage of a code in the ROM that depends on such things is a small percent.


The problem with NES emulation (having written a fairly bad one) is it can be hard to predict when you need to be careful, particularly in the presence of self-modifying code and interrupts. If it was easy to figure out, emulators wouldn't have to keep such careful track of this stuff either.

They won't for coding and images, but they will socially. Everyone I know who has invested in home AI use is mostly using it for 'things that might get you banned/limited'.

I'm quite impressed what is possible with just 12 to 16 GB of vram in terms of image generation.

Part of the problem is the courts. There are many laws now about protecting babies, in this case of they'd done nothing and mother and baby both died, they could be arrested for that.


Medical consent is a house that I consider holy but I draw the line at 'societal good'. This woman was at worst a risk to herself and the degree of violation here (besides the insults!) is revolting at so many levels. They had zero respect or consideration for her, their paper reality appears to have been more important to them than the flesh and blood human that trusted them enough to put herself in their power. And then they proceeded to grossly violate that trust. This is 'Terry Schiavo' in reverse, another case that had me wondering why people don't have respect for other people and allow them some dignity. There the courts were required to allow the hospital to stop a treatment they knew would never result in a patient recovering or even improving, and yet the parents tried to use the courts to force the hospital to continue to keep her alive against the wishes of the husband and the hospital.

And that took 15 years...


> they could be arrested for that

Is this real? Here in Argentina someone had a esophageal endoscopy a few years ago. The hospital used a high pressure air tank to insert air in the esophagus instead of a low pressure one. Her esophagus broke and she died for internal bleeding. She [the victim] was very well connected, an air TV journalist and a member of the council of the city of Buenos Aires (~3MM habitats, capital city). IIRC the only penalty was a 3 years sentence in parole the full time.


I've always wanted something like this board, buttons which can light up (preferably a few colours), to use to make games. Anyone ever found such a board which is hackable / programmable?


Novation Launchpad used to be exactly that: you send MIDI CC messages with proper values and you can light up the grid (with different colors).

Did that a few years back, i guess this might still be possible


> used to be

Looks like they are still around? https://novationmusic.com/launchpad

Also seems to be in stock locally.

The device that I think popularized that design (citation needed) was the Monome (https://monome.org/) that looks like it is also still around and it has (always had?) some kind of open source license (https://github.com/monome).


Thanks for the heads up! I have been out of this loop for quite a while, glad to see it's still there!

Yes Monome was this initial inspiration then followed by many, but it was way more expensive and niche IIRC


There's custom firmware project for the Novation Launchpad.

I had a fork of this at one point where I made GoL for it and had mapped the lit cells to a MIDI output.

https://github.com/mat1jaczyyy/lpp-performance-cfw

I haven't tried this yet, but there's also an open source grid-controller project.

https://github.com/203-Systems/MatrixOS


Aliexpress has a lot of cheap "smart gomoku board" now, they are cheap rgb-lighted membrane button grids O(13x13) with firmware playing games like Go, Reversi etc. (https://he.aliexpress.com/item/1005010089221164.html, https://he.aliexpress.com/item/1005011840606955.html ...). These proved to be surprisingly fun with children, the tactile interface is neat, and they're a good middle ground that helps learn the rules of games (has 1-on-1 modes + solo against several levels of "AI") but not as brain-rot distracting like a tablet.

I haven't tried reverse-engineering one yet, I hope at least some of these are hackable? Conway's life would certainly be one fun use, but custom games are totally something I'd want to make too!

Are these actually membrane or are there some switches there? I assume membrane, if only based on the dirt-cheap price. Anyway in combination with lighting up + sound the feedback feels immersive enough, it's definitely more fun than poking a tablet.


https://www.adafruit.com/category/280, they're ready made from 4x4 to 16x8 but in theory you can just put more modules into an enclosure.


One problem here is it's very easy to overtune to a past problem set -- even accidentally. You can often significantly improve performance just by changing your random number generator seed until you happen to pick the right assignment for the first few variables of some of the harder problems.

It would be interesting to take the resulting solver and apply it to an unknown data set.


yess. loads of space for further exploration here. there is an attempt to keep things as general as possible in the expert.md file, but hard to mitigate overfitting fully. however, changing the seed will not get you much further with all else in the solver constant. unless you try a number of seed that exponentially scales with the size of the problem


Absoultely, I had a 2 -> 3 code base I'd mostly given up on, and Claude was amazing. It even re-wrote some libraries I used without py3 versions, decided to just write the parts of the libraries I needed.

It does much better with good tests. In my case the output was a statically generated website, so I could just say 'make the same website, given these inputs'.


So do you think we should remove all laws on selling cigarettes, alcohol and guns to children, make it the parents job?


Do you think we should mandate fridges should have locks if they can contain alcohol?


It required custom glibc patches, and getting videoes to work required some kernel stuff as well.

This is a combination of getting stuff merged upstream, and removing the need for some more specialist features.


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

Search: