The analogy makes a lot of sense to me. Once you have an “easy button”, it’s hard to not use it. It’s sort of like when you’re at work and see the “quick workaround” effectively become the standard process.
I remember burning out on games because the cheats made them really fun in the short term, but afterward playing normally felt like agony.
I wish there were more options for “hints” instead of just giving you the walkthrough. I keep getting stuck in Subnautica, but I don’t want to just make a beeline to where I need to be.
Cheating always made games boring for me. I remember doing a cheat in Harvest Moon: A Wonderful Life to get all items, and it just evaporated any fun I had.
The best balance was a GameFAQs I printed out for Morrowind that just covered the first handful of quests of the game. Gave me tips for class and race selection, and just enough guidance to get my bearings.
I had a rule where I wasnt allowed to use a gameshark until I had already beaten the story mode.
So I guess the analogy there would be learn how to do the thing the old fashioned way and then only use AI as a tool to do it better.
One time I used a GameShark mid game for a Zelda game, I saved my file and came back later to find that by maxing out, I had ruined my save file and lost all my progress. I cried my eyes out and it took days to get back, boringly replaying the game.
I always wondered if that was an intentional lesson by the GameShark devs… ChatGPT seems designed to make you act against your best self interest.
I’ve had a similar train of thought. I work with a lot of people that have been doing their jobs for many years and know what they’re doing. They might benefit from an LLM since they already have the expertise to tell what to take or leave. A novice would benefit more long term from learning the hard way.
Continuing with that train of thought though, if someone has been learning and growing for years, is there really a point where it’s okay to stop, say “I don’t need to learn more,” and start relying on the easy method while their skills stagnate?
It is the externalization of internal mental processes, causing technological dependencies for even basic thinking on the subject it issues for. It is fundamentally the same as being dependent on a parent for answers, as a child. At some point the parent must force the child into independence to become capable of functioning, to build the infrastructure to answer its own questions by memorizing, and later discerning, the answers.
If we should regress to, or raise our children with, such a dependency, we will become enslaved to those who control these technologies, making useful thought into a subscription service. Technology is incredibly empowering but at some point it becomes a necessity and we are beholden to those who control such things, spawning a techno feudalism in which we are as tied to a corporation’s technology as serfs were to the lord’s lands.
The bits I used gamefaqs guides for (btw I love they are still there _) are rarely fun anyway. Mostly, it’s achievement grinding or 100%ing. If the game itself needs a guide to navigate it, I usually just drop it. If it fails at informing me about it’s mechanics that much it’s not for me.
Beneath A Steel Sky has a help system now you can refer to, and I ended up using it a fair bit. The solutions often just pissed me off though, as they rely on you remembering a one-off bit of dialogue you saw (or skipped) days ago in real time. or were just nonsense.
When I walk around the floor at work now I often see other devs on their phones while they wait for the AI to do stuff. People are getting disengaged are forgetting skills already - this is unsustainable.
Counterpoint, like, I can draw things, but I can’t draw people, but I have used AI to generate pictures of people that I can then trace to learn how to draw people, and because it’s a new person, and it’s something I’m in control of, I feel more encouraged to fire up Krita and work on my drawing.
I still suck, don’t get me wrong, but I have done more artwork since having access to AI art tools than I did for several years prior to that.
There’s just something about having an idea of knowing what the finished output is supposed to look like that helps me figure out how to draw what I’m supposed to draw.
And eventually I will be fully drawing my own stuff from scratch, thanks to using AI as a self-learning tool.
This argument says,“I use a tool as a tool.” Which is valid, using a tool as crutch when you don’t need one is wrong.
Like I sell AI art as a business.
I think that’s more than a fair point. AI is a tool, and I’m personally (tentatively) optimistic about what it will be capable of helping us with.
I think the distinction to be made in this case is when the use of a tool undermines the desired experience without us realizing it.
Like “I want to enjoy playing this game.” > (uses cheats for a little bit) > “Now I don’t enjoy playing it normally anymore.”
Or “I want to be able to think critically.” > (consults AI for everything) > “Now I have a hard time reasoning on my own.”
At first I was going to disagree and say “hey at least they are still looking up information, unlike most people” but then I did a 540° on that idea when I realized that I myself was a great example of how the OP is right.
I have been building things in my back yard like crazy this summer. I am currently working on a purpose-built little lego/craft tray for my wife to use in the house. I have gotten to plan out every detail in my head and sketching on paper, including convenient geometry knowledge like multiplying by the square root of 2 to find lengths for 45° supports or the good old 3-4-5 triangle for getting a right angle in a pinch. I have been able to discuss the table’s use with my wife to figure out the perfect features. It will be a little wooden table that’s ~2’/60cm wide like a TV tray but it will be held up by cantilever legs that are long enough and tall enough to hover the table over her lap with the footrest up. And it will have other features like little segmented bins for pieces/parts, and an instruction holder.
It’s a great activity for numerous reasons. It gets me outside, it gets me physical, it gets me interacting with my wife and excited to give her the finished product, it gives me opportunities to practice new skills/tools, and it engages the senses as well as the mind while I spend hours in a calm almost meditative state and not seeing anything that’s happening on my phone (though it will read texts to me through my earbuds).
It’s a pretty funny look. I’m wearing a big round brimmed sun/fishing hat that looks almost like Gandalf’s but without the pointy top. From the outside the sound of the scene is 95% the sound of falling water and birds chirping, interrupted by the 5% of the time spent actively cutting or planing some wood. But if my earbuds are in my ears, they are blasting my playlist of various high-tempo Thrash and Industrial Metal songs! (at 45-50% volume. I’m responsible here, lol)
So if I take all that and compare it to some schmuck who pulls up ChatGPT and types something like “design me a sturdy two foot wide table, create a list of the pieces I need and the cuts to make them, and generate detailed assembly instructions with pictures.” Yeah you might still get a functional table but your life has missed out on the vast majority of the potential benefit of the activity!
This is the way I started looking at these tasks once I really internalized the whole “life is about the journey, not the destination” thing.
I’d love to see a picture of the tray once you’re done with it!
sure thing!
busy weekend for us but there’s no way I don’t finish it tomorrow. (right?)
The stuff I’m making right now is all just pine, with flat surfaces and 90 degree corners like you might get from ikea. But with visible wood grain and built so that you can dance on it or use it to hold the biggest aquarium you can find.
I need chat gpt to summarise this post
I read the whole thing, it’s basically “sometimes the journey is the reward.”
A bit long winded but correct.
This summary is approved by the completely normal human author.
I’ve recently been obsessed with a streamer called AboutOliver. He played Minecraft for the first time about a year and a half ago, played his entire first season with no wiki or external knowledge, got a little tour of the community server (which he 99% forgot at the time Season 2 rolled around) and is now on Episode 75-ish of season 2. Still no wiki, no guides. He has figured out some crazy things about the game (which I won’t spoil), but is also completely clueless about some super basic features.
It’s been incredibly inspiring to just watch him figure things out, because he is exceptionally inquisitive and methodical by default (I think he’s a phd candidate in Astrophysics irl?). Made me realize the point of a game shouldn’t be to produce the optimal output, but that struggling and finding things out is exactly the point. Incidentally, that mindset also noticeably boosted my performance at work because I’m now one of the few people who will happily continue to tackle a programming problem over and over again, even if there are no helpful guides on it.
Long story short, here’s a link to watch the supercut of Olivers Season 1 Playthrough: https://youtu.be/ljemxyWvg8E
The total season 1 supercut is about 6 hours iircOR, if you are insane, here’s the link to the full-episode playlist: https://youtube.com/playlist?list=PL68V5Cxs_CvTpTY9o7KJ75nLPqlCRxze0
It’s 50 Episodes á 3-5h, great as background noise when doing something else.Ha! I watched him play Outer Wilds and it was perfect. It is the ideal game for someone like him because this game is all about exploring. But please play the game before you watch him play and don’t research anything beforehand or during playing.
Well…
But considering in modern Minecraft you already have a crafting book that says how to craft any item it’s not as needed anymore as before.
In the early times I believe it was to either know the recipe or to look iz up on the web.
I have to force myself to not fall into the trap of trying to play a “perfect” game and instead to just let happen, what happens. Blundering through content and accepting temporary setbacks is more fun than following guides or save scumming.
But it also depends on game design:
With bg3 I missed a one of a kind item in act 1, a staple dnd item (ring of protection) that I was locked out off because I did quests in the “wrong” order. that gave me some anxiety, after which I started checking the wiki page before starting a new zone, which eventually sucked the fun out of the game, after which I abandoned my first playthrough.And then I found a mod that randomizes all loot, so I can just let happen again what happens, without that anxiety of losing access to unique loot because of game design.
I also fall into this trap semi regularly, a happy medium I have found is a missable items guide that doesn’t tell you how to play or where to go but it does tell you “make sure you get item X before going to place Y as that’s your last chance”
It means I can be happy to play sub optimally knowing that if I really want I can go back and collect anything I missed later.
This has been quite good for Clair obscur
You got upset because you missed a +1AC item? There’s so many much better items in that game I’m surprised this one matters so much.
I totally recognise playing the perfect game angle though, depending on the game I look up collectibles ahead of time, so that when I find the area I know there’s one nearby.
Nah, the knowledge that I could be locked out of unique items is what caused anxiety, not what I was actually locked out off (though I do think it’s a really good item for a ring). I played act 1 as a blundering fool, at the end of act 1 I checked an item list to see what I missed, so I could backtrack for what I could use. And then I destroyed my fun in act 2 by checking guides before starting an area.
When the answer is to grab the fork seventeen levels back, and to not use it on the dog 3 screens before so that you had it to look at after answering a riddle written backwards in Spanish that is actually an in-joke from the devs childhood you’re damn fucking right I’m not wasting my time to “figure it out”.
Video games are not reality, I can’t look at an easily surmountable barrier and just walk over it like I could in real life to solve the issue, I have to take some deranged imagined route by a dev. I can’t logically work my way out of a situation that is some guys bullshit idea of a solution.
This is a extremely apt take
I prefer apt-get, but whatever floats your boat.
I feel like this is some programmer humour I’m too not-programmer to take
on Ubuntu, there is apt and apt-get for package managing. They do the same thing, mostly.
This comment section is full of boomers
it was so hard for me to play grim fandago without looking up the answers but i did it! 10 hours later and lots of critical thinking and i finally solved the first puzzle!
We played Leisure Suit Larry with my brother at somewhere under 10 years old without knowing one full sentence worth of English, and it took hours to even get the game to start. There was a quiz about US history and politics or something for age verification, and it took a lot of tries to guess our way through and memorize the answers. Didnt get that far in the game either.
That was vintage copy protection. They would print the answers and stuff in the back of the manual, so you could only start the game (or get past a certain point), if you have a legitimate copy of the game (or just a copy of the manual lol).
There were all sorts of creative copy protection schemes prior to DRM.
Yeah, I’m aware of all the manual and code wheel based copy protections, but I’m pretty sure that the quiz in Larry was just a rudimentary age check. There’s even a button combination to bypass it, which would have been nice to know at the time.
Police Quest 2 had mugshots.
You had to look in the manual and type the correct name to start the game. That was their DRM. I remember praying it’d be Jessie Bains, because he was the only one I memorized.
except walkthroughs are much more accurate…
Except game walkthroughs provide correct information, whereas LLMs can just make things up. So it’s more like looking at a walkthrough where each step is from an entirely different game.
But the process of “get the answer from another source instead of figuring it out” is the same
We’re entering an era where we need to decide where some lines are drawn.
How much prior understanding is acceptable to incorporate into our reasoning? If the answer has already been figured out, is it reasonable to use that, or should you do the work a second time?
If you consider figuring out how to play a game to be “work,” what are you even doing playing that game?
Y’all - For nearly a quarter of a century Nintendo published Nintendo Power, a magazine that was a combination of self-hype and how to beat their own games. In the 90s, it was indispensable for any game worth its salt.
Nintendo used to run a 1-900 number for tips on games. You’d call a real human who would walk you through where you were.
Looking it up online is only “cheating” in the sense that it’s immediate and free. This stuff used to cost money.
Yeah, LLMs are like if you called the Nintendo hint line, and the person on the other end just made shit up.
Plus with games never explaining how some of their mechanics work and not giving you any realistic way to experimentally determine it, why wouldn’t I look it up online?
A big one that comes to mind is stuff like attacks, armor, and HP. Games handle them differently and very rarely tell you exactly how they work.
