I imagine it's really hard to spot a comma in the wrong place, or a missing sentence in a 10 page contract unless you wrote it yourself, or you assembled it from some battle tested templates.
I was writing a similar comment to yours, thinking about "value to society", and value to loved ones, even negative value of enemies. I agree that talking about "value of humans" is not very useful.
Then I realized that on an individual level, everybody is infinitely valuable to themselves. You are your whole universe. From that perspective, I agree with the author that "Humans are valuable."
We have laws keeping humans alive and safe because we are valuable in that sense.
I don't agree that we need to go out of our way to preserve human art though, or their thoughts on the value of "creative artifacts". People will make art if they enjoy making it. Whether or not other people appreciate that art is irrelevant.
Context is everything. If the wider discussion was about how men are better than women, and in that context it was shown that "Most good engineers are male", it would be natural to draw the wrong conclusion.
> Artists work in whatever canvas size makes sense for the motion they’re drawing.
made me think "what?" Artists making concept art or drawing for fun, sure. Whatever resolution is comfortable is fine. But for animation or in-game art, a resolution is always defined. Even the most amateur artists learn about sprite sheets the first time they try 2D animation, and rule one of sprite sheets is using a uniform resolution for every frame.
The problem here wasn't that frames in single animation had different sizes, but separate animations. The example images are a bit misleading since they show only one frame from each animation.
Every animation tool will provide consistency across frames of single animation, but functionality for aligning and managing transitions between pairs of different animations is less common and more in the territory of game development tools instead of animation drawing.
That's still something the game programmer linking animations together should be familiar with. You could still have artist use single frame size for all animations, but that means having gigantic square many times the width/height of character since you can have some animations which are stretched horizontally and some which are very tall. And if your workflow doesn't include good sprite packing, such gigantic frames can easily become impractical in terms of texture size.
Having the game programmer define single anchor point per animation (assuming aligned frames within each animation) doesn't take much time at all. You wouldn't want to realign individual frames within each animation (that's part of artistic intent and artist should do at least that much), but even that would take order of magnitude less time than the process of drawing those frames.
There are still times the programmer might have to go through each frame defining anchor point/hitbox, when artist picked a wrong config during export producing packed sprite sheet and there is no time to reexport it, or when dealing with animations that mix animated and software driven movement like large jumps.
In every game I've made, I've had a resolution set for characters. Resolution is consistent across animations. It's week one gamedev stuff. Feels like Claude should've picked up on that, but clearly the 1 trillion they've put into it already hasn't made it as competent as the typical 14 year old noob dev.
Don’t like to cast stones, but this feels like Claude trying whatever it can to make things work, without fixing the underlying process and problem.
“It looks like the user wants to run curl on windows machine, I need to bootstrap Linux under docker, and channel bash commands into shell inside docker, so the user would be able to run curl natively.”
It's hard to care much about how software is written when superpowers are invading their neighbors, democracy itself is under attack, the food supply chain is failing, and the world is hotter than it's ever been.
software plays a big role in enabling 2 (arguably 3) out of those. Least we could do is not make it worse. But it seems most of us are still just children. They give us something shiny that makes pretty lights and we happily burn the world to keep it.
I've been thinking the same: the smaller the codebase, the better AI performs. So a way to scale AI is to modularize your architecture to maximize the number of leaf nodes in the dependency tree, and split out separate libraries where it makes sense.
It is huge for token usage also, Claude grepping the codebase for context it doesn't have is the main consumer of input tokens from what I can see.
The whole AI powered utopia is a bet on two possible outcomes:
- humans and companies somehow stop being greedy, selfish and cutting corners to optimize for revenue and time-to-market
- LLMs are the path to artificial superintelligences that will be able to deal with the exponential increase in tech debt from throwing AI slop at the wall (vibecoding) because no one has time to do things “the proper way”
The former is impossible. The latter is extremely unlikely and an existential threat to humanity.
The so called Luddites are the only ones to have even engaged at all with these concerns. Everybody else is just focused on the selfish game (see bet #1) of staying afloat in a rapidly changing ecosystem.
I don't know why you are getting downvoted. Perhaps because people don't like the sentiment. But its true, people are hired as tools to write programs.
Both people and AI make mistakes. Perhaps the AI makes more, a lot more, but its so fast, and works around the clock, and has no ego, there is a chance that the benefits outweigh the costs.
Unfortunately a lot of fiction (sci-fi/fantasy) ebooks are effectively kindle exclusive these days (amazon publisher deals exclusivity), due to the near monopoly amazon has… and since they have locked things down even harder lately, it is much more difficult to export purchases to other readers.
Huh. OK, fair point. I have my portrait buttons swapped because my thumb naturally rests next to the top button. I don't use in landscape enough to be an issue.
BTW, slapping a pop socket on the back so I can comfortably read with one hand was a game changer.
I did consider a Kobo, but the terrible website did turn me away. When trying to browse it was wait for button on tracking consent to be active so I could click reject. Then change language to what I'm fluent in instead of getting language based on IP, reject tracking, reject changing region back to IP based and the same two reject for every link clicked. Half the times pages was still shown in language based on my location instead of what I have set or what my user agent tell I want.
When a site ask about allowing tracking for every page shown I assume the company care more about selling PII than getting customers, so not a company I want to use
How did you get siding with Amazon from giving up on the Kobo website? There are more than two options, re-reading some of the dead-tree variants I have would keep me occupied for years. Hopefully a new alternative would come out before I run out of space for books
reply