When people work on projects like these, it is often a communal activity, for the love of a shared mutual interest. Friends are made, technical expertise is gained and shared, people learn from the code and from each other, and ultimately something fun is produced. We are a social species and it feels good to help other people, and to feel supported in our efforts by others. It gives us the feeling that we matter, what we’re spending our time on matters, that our interests are not merely selfish, that we are not alone in the dark.
In any ecosystem, different specialists fill different niches, and you can’t have a vibrant ecosystem without all of them. For projects like these, the place where the reverse engineers finish is where the mod tool developers take over; where the mod toolers finish is where the artists and game designers take over; where the artists finish, the fans and beta testers take over. Some fans might even try their hand at helping with one of the other parts, and learn how to become good at it, and now you’ve got a pretty good stew going.
LLMs poison the entire ocean, kill all the coral and destroy all the ecosystems. This is true in RE, but it is also true in OSS. If you try to do something human, the moment it becomes interesting to someone with tokens for the yes-sir machine, they will steal all that momentum for themselves. Contributing to and actually using the human version, then, becomes something one would only do out of principle. That’s a very shallow pool to try to survive in compared to an entire ocean.
The actual physical reality of the status quo of reverse engineering and recompiling is that the vast majority of stuff never gets the treatment because they aren't lucky enough to catch the attention of one of the few magical individuals with extreme skill and dedication willing to spend worth hundreds or thousands of hours of their time on them. The consequence is rot.
The status quo is thousands of pieces of software, including but not limited to video games, are lost every year, with no feasible way to recreate the systems and conditions that are needed to run them, and no way to update and repair them to be able to run again.
An estimated 88% of PS2 games are already lost forever. There are thousands of people in the world right now who have in their possession the last copies of digital media with no way to play them and nowhere near enough skill to crack and preserve them themselves.
The alternative to REA isn't hand-made free-range organic recomps, it's nothing.
No, not at all. I myself am not a “the means is the end” kind of person. If I just wanted to puzzle out solutions with nothing to show for it, I’d do the crossword. But, in my experience, caring only about results leads to myopic short-term solutions that create bigger long-term problems. For instance, if you’re in pain, I can “cure” that by shooting you in the head. A quick positive result for getting rid of the pain—but at what cost? For thinking work, AI seems to be shaping up to be a lot closer to a bullet than an analgesic.
All I was saying is that human motivation does not exist in a single dichotomy, and that ecosystems that develop around mutual effort create benefits beyond the mutual goal. You can’t get all those benefits without humans doing the work, but there’s no extrinsic motivation for humans to do the work, so very few will bother, and the ecosystem dies. (I hope it wasn’t ‘load-bearing’.)
Anyway, the benefit of reconstructed source code is not even about preservation, it is in enhancing the user experience beyond mere preservation. We’ve been preserving software using emulation alone just fine for quite some time. If you are actually worried about preservation, you might consider how much the Wayback Machine is getting blocked these days because of AI scrapers. Or how an AI agent might apologetically wipe out some irreplaceable archive. Or how nobody in the future will probably be able to care about preservation much as the one-two punch of climate change and AI descend on us. It’s always those damned long-term consequences from short-term thinking, you know?
This is certainly not true. Where did you read this?
also, emulators exist, it's not recomps or nothing.
On the one hand, I feel like there was this nice little niche of hyper technical people that had formed nice communities. Then AI bros came in took all the attention and resources.
On the other hand, I have significant concerns that this is a mechanical transformation that will blow up for everyone and undo the last 40 years of fair use progress.
I think your question would make sense if these were the only two reasons. Someone who just wants to make sure their old games can run would see these tools as a good thing. People who just enjoy reverse engineering by hand wouldn't care this exists, or maybe even, a helpful tutor.
So perhaps then, the sentiment being as negative as it is, is revealing something about the hobby: that many participants, considered the metagame of being the first to do it, to be "smart enough" to have figured it out, the "street cred" if you were, is what they are lamenting, but they just either lack the words, or aren't being honest with themselves.
This is a very ingenerous interpretation.
I am a hobby hiker. My mountain hikes are so easy that basically anybody could do them: no street cred involved, it’s not even worth to speak about them. But it is important to me that most goals are only reachable by walking and putting in the effort: if a company offered a 20$/mo service that brings subscribers to the endpoint, carried in a luxury litter by robotized carriers, I would feel that my hobby has been devoided of sense.
As someone else who hikes: why? I do not understand. Do you not enjoy the hobby of hiking for the sake of hiking?
The views will be just as beautiful, the air just as fresh, and the feeling of moving my legs just as energizing, regardless of how others may decide to get to the destination themselves. Why would the experience of others have any impact on my own experience of a trail?
There is something about a place requiring effort to visit that makes it special, and it loses that quality when it can be trivially accessed. The air isn’t quite as fresh when there’s a parking lot there.
Rather its the impact of others on, not their experience of, a space that would impact my own experience, for better or worse.
At every turn it seems like a lot of intellectual pursuits are just being voided by this technology.
I've heard the "one level up" argument in software with the supposition being that the work's still there, it's just becoming more abstract... but even in software, this doesn't really track.
The injection of this technology into not just intellectual but artistic pursuits (since there are now commercial AI image and music generators) should be a red alert.
What is there to live for once everything you once loved to do has been optimized for maximum revenue? Where every time you do something for fun, someone will come out of the woodwork when you show it off to either tell you AI could've done it quicker, or that they made an AI version with a flashier UI or something?
if it brings you joy, what does it matter how someone else does it?
If you run for fun, what does it matter that someone else does it faster, or in a paid exhibition optimized for max revenue?
Do it because it brings you joy, there is no external dependency.
> Where every time you do something for fun, someone will come out of the woodwork when you show it off
See, this is where things get tricky, either the activity brings you joy, and/or the sharing of it brings you joy, and/or the showing off of it brings you joy.
For the first two, it is irrelevant what someone else has to say or do about it; ignore the noise. For showing off, yes, some activities are no longer worthy, so change is needed to find those that are.
Hopefully for a hobby, or for anything that you love to do, showing off is not a major criterion for why you do it.
That aside, I've been lucky enough in my career to work alongside, mostly, people who have a personal interest in using software as a tool, and knowing their toolchains very well, be that a language, a command line, a system architecture, you name it. If you found something cool to do with a tool, you'd share it to disseminate the knowledge. That culture is very quickly dying off, I feel, because LLM-driven work encourages siloing to a degree I don't think gets discussed enough.
Lastly, using LLMs to write code is its own type of exhausting. I thought I could keep up my hobby by mentally checking out of work and working on projects by hand in my free time, but babysitting the LLM, reviewing the massive corpora of code and documentation it puts out (including those put out by colleagues), and the insane volume of context switches this technology encourages in the workplace makes me want to stay the hell away from computers alltogether if I can help it.
This example is actually very clarifying - perhaps almost no one actually runs for fun?
The cases of a niche culture destroyed by dilution are so common that there are essays around it (e.g. Geeks, MOPs, and sociopaths in subculture evolution), so it seems to be a very normal and human thing to happen. Blaming the individual members for failing to feel differently is not going to fix that.
If the goal was to show off your accomplishments on social media, sure.
Taking that away compounds with the crisis of meaning we already have to make people feel more alienated than ever.
I must think of all the people who poured countless years of their lives learning all the C++ bullshit..
This is a very weird false dichotomy. It almost certainly was both of those for everyone who made any meaningful progress. You have to care about the end result to stick with a project to the end after you've finished the fun parts, and you have to enjoy most of the process of getting there for it to be a hobby. The journey and the destination matter, not one or the other.
I haven't followed this closely the past few years, but it used to be pretty obvious you were listening to a VST compared to a real instrument in actual playing. When you compare recordings, sample based ones sound awesome as long as you just try a single note, but they were obviously missing something in the sympathetic resonances (sounding worse than my upright piano which is perhaps a 20th percentile piano), and the more physics based ones had not so great tones.
I've actually been hoping that someone would try to apply deep learning to piano VSTs, even emailed a few researchers, but no reply. If something new has come up I'm not aware of, I apologize.
When it comes to real instruments, Roland ships physical modeling in some of their instruments. They are a mainstream company.
So I think physical modeling is there but harder to get right than recording samples of piano keys being pressed. The results can be been convincing and it's nice that it requires little memory compared to sample based techniques.
It's to the point where you virtually put microphones around the virtual instrument and it behaves (according to the person who does it for a living) creates an intuitively correct sound.
Edit: and to be clear this was a demo from before the LLM times so it's something built the traditional way.
You can bring this distinction back to recomps, too -- one area you can't use a vibed recomp over an emulator is things like TAS's[0], since the TAS community has strict rules about staying compatible with the original hardware, including doing things like sending the TAS inputs into an actual console to verify it works properly. A recomp is likely going to have minor (or major) compatibility issues that would prevent this from working. But it won't matter for 99.99% of players.
It’s interesting to dwell on what ‘we’ appreciated before versus what we appreciate now. Since LLMs lowered the bar, it feels like raw human effort has come to be held in higher esteem.
Are we going to re-evaluate what we were impressed with in the past based on that new metric? For example, will accomplishments by experienced engineers be devalued in comparison to accomplishments by outsiders? Will the software engineering equivalent of ‘folk’ or ‘outsider’ art become more valued? There will be a natural instinct to distrust outsider contributions because, well, how did they get so good?
Are some metrics of human effort going to be de rigueur going forward?
Since it's not educational as it used to be, if you built your own version of n64 emulator number 10143 you got something out of it you learnt some interesting stuff, learnt software development, hardware etc.
Now we are literally all just taking a magic machine pouring money on one end getting half baked output the other but no progress is being made on money is being wasted. Since even if 1 person had built it everyone would have had access already.
This is the same wave of slop coding/vibemaxxing we had a few months ago where everyone was building product X for the 1 million-th time without any actual progress being made.
Math has a similar problem now, before if 1000 people tried to solve a problem they all grew from that experience, the LLM is not growing/learning anything from this effort.
All we are doing is burning resources for our pleasure, I say that as someone who has now written his own compiler, vcs and browser stack, with vibe coding, though largely I would claim I have used and tried to read the code, clean it up, to learn from it where different things break.
But even this feels absurd waste of resource imagine all of us collectively burning 200$ per month plans which assuming a at cost pricing we are all burning 200$ of electricity each month to do what exactly? Some folks are on their 10th subscription that means most people are burning 1000-2000$ of ultra cheap energy generally sourced from a mix of gas/oil turbines burning away at big data centers these days.
We could all solve it by making a site.. cough github.. and not duplicating work.. cough..
But somehow because it's easy no one seems to care.
I am hardly against software progress, and if AI gets better we will have better software (because most people write horrible software most of the time) so I am happy about it, to an extent that I can be, with my profession feeling somewhat challenged.
But given I have always worked as a fixer of broken stuff it's very much also feels fine to me. Either way I am very much against these 200-500 dollar plans that allow you to burn hundreds of dollars of energy for little productive gain duplicating the same work 100th time.
I am sure software developer using 100-200$ plans just for work can get more out of it. And it's probably making up for the effort they would have put in their work otherwise.