TehPers

If anyone’s curious:

$60/$49 = ~1.22, or ~22% more from 2021 to 2026 (~4.5%/yr) $49/$32 = ~1.53, or ~53% more from 2001 to 2021 (~2.7%/yr)

Time periods are vastly different of course (5y vs 20y) but inflation has accelerated drastically since the pandemic (specifically 2021).

None of this is representative of spending power, of course.

It’s nice to see that business execs can be automated now though. Well, except the part where they openly bribe politicians I guess.

Maintaining projects, keeping them up to date and fixing bugs etc really isn’t difficult for them.

They tend to bloat codebases over time. It takes someone who understands the code to know how to use them correctly, and it often takes about as long to direct and correct them as it takes to just write the code, at least based on what I’ve experienced at work.

The step I’ve seen a lot of people skip when using these coding agents is learning what the code actually does, which includes learning all the libraries and functions being used rather than looking at a call and going “yeah that seems right” because it looks like it’s doing what it should. Without that, it’s impossible to maintain a codebase over a long time, LLMs or not.

Like Name 10 AI games for example I dare you.

I could probably find 10 just by going to recent releases on the App Store or Play Store, but I don’t really subject myself to those stores.

Anyway, on the topic of LLMs used in modding: they are. There are a lot of mods made primarily with LLMs. Having recently installed mods for several games and worked on a modding project locally while in an active modding community, there are mods that even admit on the mod page itself that it’s basically written entirely by AI.

This is usually for mods that are primarily code, though. I don’t know how prevalent AI-generated assets are, honesty.

Personally, I don’t really have an opinion on it. I’m not going to dictate to someone who I’m not paying how they should work on their own project. I may choose to avoid them if they generate their whole mod, though.

Also, in the past, I learned a lot from writing my own mods, and I only got that experience from doing it myself. It’s a shame, but people generating their mods will likely not get any of that experience themselves and will forever be dependent on LLMs as a result.

The code is the only thing that gets more available but even with that you need to know how code actually works. What the hell is code and what does it actually do. Their is books on C++ and even the creator of it said you don’t actually need to know the code fully to work with it. Ironically Code bases have become so bloated that you need to use AI to actually explain what the hell are you looking at.

Many mods for Starfield (which is Creation Engine as well, though later) are AI-generated. They generate either Papyrus or Lua (if they use SFSE) usually. Assets are another story of course, but I don’t think many of these AI-generated mod creators actually know how the code works.

Honestly, I have no idea what’s going on in conservative spaces, but they’ve been taking quotes out of context for well over a decade at this point (if not longer but I haven’t paid attention much longer than that). It’s entirely possible he said that and they’re quoting it, though the conclusion I’d have without more context is that he’s referring to something like binary partisanship, not gender. edit: skimmed past the quote with context lol. Ok, good job Talarico? I’m genuinely surprised.

But yeah, it’s going to spout nonsense. Even ignoring the potential for foul play, it makes shit up even with its “harmless” responses. My point was more that it at least lowers the barrier for people who would have done no research anyway, and those people often tend to lean right.

Using ChatGPT (or any LLM) for this purpose is obviously incredibly flawed and very likely to lead towards bias (both intentional by the service provider and unintentional from the training), and in a world of mass misinformation, a chatbot is going to be no better at filtering the facts from the bullshit than a person would be, and we know how good people are at that.

But if there’s one silver lining: anybody who did no research before who now does this at least is exposed to more perspectives, hopefully. One would imagine this is a possible gateway to further self-directed research.

300 hectares — the equivalent of almost 500 football fields

Thanks for the context. As someone who lives in the US, I wouldn’t have been able to put that to scale otherwise.

I keep reading RoR as “Risk of Rain” and I have to admit that I’d be far more interested if their developers said something than if DHH did. DHH is pretty much just a wannabe Elon Musk, and nothing he says is worth the bytes of storage it takes to hold them.

Yep, tokenmaxxing is a thing at some companies now. It often aligns with the same people competing to see who can generate the most lines of slop code (losc, which is different than sloc of course).

It wasn’t long ago where people understood that every line of code you write is a line of code that could have a bug in it. The good devs still know this, but many are also being forced to generate shitty code just to appease their shitty management now.

As an aside, I have yet to see a case where tokenmaxxing generates a profit for either the consumer or the provider. It just reeks of the same “market capture at all costs” mindset that have plagued SV startups for so long.

It’s not a question of if anyone holds the copyright, but a question of whether the models themselves and their outputs are derivative works of the data the models were trained on.

If they are derivatives, it may be possible that the licenses of that training data place additional restrictions on the generated code or forbid derviative works outright.

Speaking practically, for a random “FOSS” project that isn’t very big, this is unlikely to ever come up. Courts would have to rule in favor of the training data owners, plus they would need to issue DMCA takedown requests and/or sue to have the content removed. That probably won’t happen.

However, you should probably question any project where the developers don’t understand the code well enough to write it themselves. I don’t know if that’s the case here, but that tends to be the case with vibecoded projects.

Also buried in the article is that they’re killing off RSS feeds (unless you moderate that sub I guess).

I haven’t touched Reddit in a long time now, and I have no plans to go back. But if I still did use it, killing old Reddit would have me leaving entirely. Their “new” interface was atrocious on release and I assume it’s far worse now that they’ve ramped up hard on ads.

They also have been showing video ads occasionally to nitro subscribers, which kills any value nitro could have had.

There are almost certainly bugs. Like any game, give it a few weeks (or months) before playing it unless you want to bugtest it.

You can switch to an older version of the game in Steam if you want by going to the game properties in your library.

From my understanding, only federal criminal charges (and convictions) can be pardoned anyway, whereas I believe this is a state civil case (a lawsuit).

Ok I had to look up what the federal limit is here in the US and it’s apparently 10gal/min (~38L/min). Your pumps are just firehosing it then considering your fast one is more than triple the flow rate of the fastest pumps we can have here.

Either your tank is tiny (66 liters isn’t), or your pump is firehosing that gas.

Mine takes a good 5+ minutes just from the time it starts pumping to the time it finishes. It’s a >20gal tank though so I don’t fill it very often.

From an article linked in this one:

Via WSVN, Bermudez claims he’s “not against video games, but to go out and get a video game where – which glorifies not only our first responders getting shot, but also pedestrians and residents – it’s nonsensical.”

Republican sheriff Cordero-Stutz similarly points to GTA’s crime-centric themes, adding: “At a time when law enforcement is working every day to reduce violent crime, prevent robberies and car thefts, combat drug trafficking, and build stronger relationships between our deputies and the communities we serve, I believe we need to seriously consider the message we send when our own government promotes a fictional identity centered around many of those same crimes.”

Remember, kids. Guns don’t kill people. Games about guns kill people.

I’m not driving under the influence, I’m travelling under the influence.

These people are just sovereign citizens with a lot more money.

For what it’s worth, modern component libraries compile the styles to CSS classes anyway. I don’t know why they choose to go through the convoluted process they do to generate those classes, but the user usually just sees classes at least.

The article seems to conveniently hide the rampant corruption between Flock and local governments. Not that Wired has an obligation to include it, of course, but it’s still relevant context as to why everyone hates Flock.