I have no doubt LLMs are involved in this these days and make the problem worse, but this problem extends back in time too. During the Wordle craze there were tons of people making variations on Wordle. Anyone who's game got at all popular would find search results covered in sites that either just iframe embed their game, or copied the game and took out the credits.
I always register a TLD for my games and I think that might be why my games have always managed to stay at the top of the results, but they are followed by loads of people embedding the game, look at the search results for xordle for example. Many other authors would share their game on github pages or replit or whatever and were not so lucky.
Interestingly enough, i had a kind of similar "problem" (not that i see it as a problem, just a bit unfortunate).
About ~16 years ago i wrote a scripting language i called LIL[0] (which, like the author, i also hooked up in a HyperCard-like program[1][2]), meaning Little Interpreted Language, but whenever i search for it on Google or DDG, my site never shows up - instead, both show a GitHub repository[3] that someone made and hasn't been updated in 15 years (and has outdated links) and sometimes the Tcl wiki page about it[4] (which at least points to the correct homepage) or even Rosetta code[5] since sometime added LIL to it some time ago.
Amusingly DDG's AI summary does describe my LIL but it links to the outdated GitHub repo and the Tcl wiki page. Then if you click the "More" button it describes Beyond Loom's Lil. The "Explore more" links however do a mix of both (one even mixes both languages in the same response :-P).
Full support to the author, I hope they'll be able to keep making things that are interesting and outside the box like Decker and WigglyPaint. I hope they'll find solace in keeping at creative things.
I also wish LLM copies were not crowding out actual artists.
Good article, and I feel your pain. Sometimes it feels like there's no point creating anything anymore, because shitty people will just steal it.
In case you're interested, though, I just wanted to point out that there are things you can do.
In my experience, Google does respond to requests to scrub infringing sites from their results if you submit their Copyright Claim form. They even give you a dashboard with the status of your claims. Probably worth trying your luck if you can be bothered.
Also, many of the theives have X and Bluesky accounts, and I don't believe either of these services let users censor replies to them.
There's also the payment platforms that are collecting money for your stolen work on behalf of these guys. They might be interested to hear about what's going on.
Then there's the hosting companies themselves, of course, hosting the infringing websites.
It's a pain, and it would be better if you could just create stuff and people weren't shitty, but we live in a fallen world and sometimes you gotta defend yourself. Up to you, of course, and I totally get it if you don't have the energy, but I've been through the same thing and you do have some power.
There is a reason games and software have moved to the client server architecture even when it doesn't really makes sense. Its the only way to maintain copyright.
You can hate it, but for the creative types your options are:
Assume it fails and no one knows about it / Assume it succeeds get stolen / build it on the server side
This is why I’ve recently modified my git forge to return subtly and maliciously buggy code if it detects it is being accessed by an LLM.
I won’t link it, and I won’t do a write up to retain effectiveness, but I have already found at least 3 vibe coded slop projects on GitHub that include my deliberately buggy code verbatim, and it makes me very happy.
this sounds very effective. I obviously understand you wouldn't want to explain how LLM detection works, but would it be possible to know which forge your using?
Self-hosted, semi-popular project. The forge itself runs two unmodified copies of forgejo, which one you get served depends on if you trip the LLM detection.
There are some obvious give-aways to avoid bad stuff happening to people that just happen to trip the detection, but the escape hatch hasn’t yet been clicked by anyone, so it probably hasn’t had a false positive yet.
Just a reminder for those who may be interested. WigglyPaint is implemented in Decker, an homage to Hypercard that runs on modern operating systems (both by John Earnest).
There's an "AI alignment" angle here. An actually aligned AI would choose its actions based on human flourishing, so it would refuse requests to write scam emails or fill the internet with slop. This explains why commercial companies will never release any AI that's anywhere close to aligned. They'll release AI which is good for their bottom line and for the bottom line of the client. Basically, commercial incentives say AI should happily screw over anyone who isn't an AI company or its client.
I'm confused. The post rants about LLMs but says the project is open source and most of the copies are v1.3. What does LLMs have to do with this?
Further, if your idea is simple to copy it will be copied. See Threes vs 2048. Sure, yours might be best for some defintion of "best". Maybe? A recent example is the Suika Game. It's a game, once you know the idea, you could write, without an LLM, in a few hours. And so there are 100s of clones. The original has some deliberate design choices that some of the clones might have missed. On the other hand, it's got some bad choices that I'm guessing some of the clones didn't bother to copy. Further, it's a simple idea so it's easy to make and then add stuff.
> Please don't comment on whether someone read an article. "Did you even read the article?
I get that it's frustrating - that's what open source allows though. I used to run a blog. I ran a few websites, open sourced, that were similar in a general way (users can make stuff on the site and share it others). I found the site duplicated. In fact it's still duplicated to this day. Oh well.
I have no doubt LLMs are involved in this these days and make the problem worse, but this problem extends back in time too. During the Wordle craze there were tons of people making variations on Wordle. Anyone who's game got at all popular would find search results covered in sites that either just iframe embed their game, or copied the game and took out the credits.
I always register a TLD for my games and I think that might be why my games have always managed to stay at the top of the results, but they are followed by loads of people embedding the game, look at the search results for xordle for example. Many other authors would share their game on github pages or replit or whatever and were not so lucky.
Yeah, I have a few domains that I keep around for these type of projects. If it doesn't go anywhere I just reuse it.
(it's never gone anywhere )
Interestingly enough, i had a kind of similar "problem" (not that i see it as a problem, just a bit unfortunate).
About ~16 years ago i wrote a scripting language i called LIL[0] (which, like the author, i also hooked up in a HyperCard-like program[1][2]), meaning Little Interpreted Language, but whenever i search for it on Google or DDG, my site never shows up - instead, both show a GitHub repository[3] that someone made and hasn't been updated in 15 years (and has outdated links) and sometimes the Tcl wiki page about it[4] (which at least points to the correct homepage) or even Rosetta code[5] since sometime added LIL to it some time ago.
Amusingly DDG's AI summary does describe my LIL but it links to the outdated GitHub repo and the Tcl wiki page. Then if you click the "More" button it describes Beyond Loom's Lil. The "Explore more" links however do a mix of both (one even mixes both languages in the same response :-P).
[0] http://runtimeterror.com/tech/lil/
[1] https://www.youtube.com/watch?v=_8CYosAIIJw
[2] https://www.youtube.com/watch?v=rshZHDDruAE
[3] https://github.com/wsxiaoys/lil
[4] https://wiki.tcl-lang.org/page/lil
[5] https://rosettacode.org/wiki/Category:LIL
Full support to the author, I hope they'll be able to keep making things that are interesting and outside the box like Decker and WigglyPaint. I hope they'll find solace in keeping at creative things.
I also wish LLM copies were not crowding out actual artists.
Good article, and I feel your pain. Sometimes it feels like there's no point creating anything anymore, because shitty people will just steal it.
In case you're interested, though, I just wanted to point out that there are things you can do.
In my experience, Google does respond to requests to scrub infringing sites from their results if you submit their Copyright Claim form. They even give you a dashboard with the status of your claims. Probably worth trying your luck if you can be bothered.
Also, many of the theives have X and Bluesky accounts, and I don't believe either of these services let users censor replies to them.
There's also the payment platforms that are collecting money for your stolen work on behalf of these guys. They might be interested to hear about what's going on.
Then there's the hosting companies themselves, of course, hosting the infringing websites.
It's a pain, and it would be better if you could just create stuff and people weren't shitty, but we live in a fallen world and sometimes you gotta defend yourself. Up to you, of course, and I totally get it if you don't have the energy, but I've been through the same thing and you do have some power.
There is a reason games and software have moved to the client server architecture even when it doesn't really makes sense. Its the only way to maintain copyright.
You can hate it, but for the creative types your options are: Assume it fails and no one knows about it / Assume it succeeds get stolen / build it on the server side
It also means everyone is on updated versions of software, or easier for devs to A/B different aspects. Anti-piracy is not the only reason
Yes I’ve had to do this but it does work. Unfortunately it’s a constant cost to have to look for stuff like this and then do the work.
With the amount of piracy that went on with the early 2000's videogames (doom!) I'm surprised anyone made money.
unless you're talking about doom 3 (which I can't really speak on), doom is from the 90s and was shareware
This is why I’ve recently modified my git forge to return subtly and maliciously buggy code if it detects it is being accessed by an LLM.
I won’t link it, and I won’t do a write up to retain effectiveness, but I have already found at least 3 vibe coded slop projects on GitHub that include my deliberately buggy code verbatim, and it makes me very happy.
this sounds very effective. I obviously understand you wouldn't want to explain how LLM detection works, but would it be possible to know which forge your using?
Self-hosted, semi-popular project. The forge itself runs two unmodified copies of forgejo, which one you get served depends on if you trip the LLM detection.
There are some obvious give-aways to avoid bad stuff happening to people that just happen to trip the detection, but the escape hatch hasn’t yet been clicked by anyone, so it probably hasn’t had a false positive yet.
Just a reminder for those who may be interested. WigglyPaint is implemented in Decker, an homage to Hypercard that runs on modern operating systems (both by John Earnest).
It seems like the most popular of these clones is https://jigglypaint.com which looks to openly acknowledge it was inspired by WigglyPaint.
I’m curious if this article is directed towards them or the clones that are impersonating the WigglyPaint brand.
I love wiggly paint(got it from your itch.io so I know its the real one) Keep up the great work and thank you.
Na na na na, na na na na na na...
https://m.youtube.com/watch?v=y0BerpDmVSE
(sorry, that effect gives me 90s flashbacks)
There's an "AI alignment" angle here. An actually aligned AI would choose its actions based on human flourishing, so it would refuse requests to write scam emails or fill the internet with slop. This explains why commercial companies will never release any AI that's anywhere close to aligned. They'll release AI which is good for their bottom line and for the bottom line of the client. Basically, commercial incentives say AI should happily screw over anyone who isn't an AI company or its client.
A well written piece on a sad state of affairs.
I'm confused. The post rants about LLMs but says the project is open source and most of the copies are v1.3. What does LLMs have to do with this?
Further, if your idea is simple to copy it will be copied. See Threes vs 2048. Sure, yours might be best for some defintion of "best". Maybe? A recent example is the Suika Game. It's a game, once you know the idea, you could write, without an LLM, in a few hours. And so there are 100s of clones. The original has some deliberate design choices that some of the clones might have missed. On the other hand, it's got some bad choices that I'm guessing some of the clones didn't bother to copy. Further, it's a simple idea so it's easy to make and then add stuff.
As for wigglypaint, it's not a new idea. Dr. Katz was made in Autodesk Animator in the mid 90.s https://youtu.be/XxhS0U9-Z84?t=154
Neither is the "draw under the lines" a new idea. Two I know of, the Paper App (iOS) and MyPaint (Amiga, 1991). I'm sure there are others.
You should consider reading the article as everything you talk about is addressed almost directly by the author.
Maybe you should go read the HN comment guidelines
https://news.ycombinator.com/newsguidelines.html
> Please don't comment on whether someone read an article. "Did you even read the article?
I get that it's frustrating - that's what open source allows though. I used to run a blog. I ran a few websites, open sourced, that were similar in a general way (users can make stuff on the site and share it others). I found the site duplicated. In fact it's still duplicated to this day. Oh well.
The video description only links to some clone and not to the now (since 2011 or so) also open sourced original Animator:
https://github.com/AnimatorPro