Given how close the project matches the original, I have a hard time imagining this could've been done without examining how the original functions. And whether it's copyright infringement or not is of little interest to me since I view copyrights as being inherently harmful. However, since the implementation is using a different language, that does make the implementation itself a clean room one.

Ultimately, it's a moot point whether they decompiled or wrote a spec from scratch though. What matters is that the apps work and it mimics what Adobe ones do sufficiently well to be useful.

And it still takes time to add features using LLMs. It's not magic, and I'm more amazed how much they managed to add already rather than it not having parity yet.

Reimplementing Photoshop is obviously a much bigger effort, but I'm just using that as an illustration that LLMs are pretty good at reverse engineering work. Which makes it plausible to me that they could get a lot of the features figured out quickly.

I'm not a designer, so my needs are meagre. I want layers, to be able to select objects, add text, do color correction, and so on. Both GIMP and Photoshop are huge, so I haven't had a chance to explore the depths. All I can say is that I was able to get the things I normally do without any real friction. It does appear to work with PSDs fairly well, but again I don't have really complex ones because I use the tool in an amateur way to begin with.

What I can definitively say is that the UX is strictly superior to GIMP. Doing stuff like text manipulation like changing size or selecting fonts has always been a pain in GIMP, and as much as I want to like it, I find it painful to use. Photocraft apes Photoshop UX which I find to be superior. So, the feature parity isn't really the stumbling point for me, it's the UX.

But that's my whole point here, most people are in the same boat as me. They do a few specific things with a tool, and if those things work then the tool solves the problem.

How long it’s been developed is… two weeks, as of today. How many users? It’s far too soon to say.

But this is something that'll sort itself naturally over the coming months. If the developer stays active and continues working on it, then it'll keep getting better. If they abandon it and move on then it'll either die or somebody else picks it up. But it is in a fairly usable state as is from my experience, and I'd argue a lot of the hard work has already been done. So, we'll have to just wait and see what happens with it.

Another point to mention though, is that it is becoming viable to maintain your own tools now. For example, if Photocraft got abandoned. It works well enough for me, and I could have an LLM fix any gaps in the way I'm using it. I don't have to maintain a general tool that everyone can use, I can just have one that does what I need.

The barrier before was that you had to spend a lot of time to figure out how any one app works to make meaningful changes to it. And you only have so much time in a day, so you can't really take ownership of a bunch of tools like that. But with LLMs it is becoming much more practical. Grab an open source app, adopt it to work the way you want, and just keep using it. If you ever need new features, you just have the LLM add them in for you.

And if we get to the point where all this stuff can be done using local models, then we really could see Stallman's vision for GNU become a reality. Regular people, who aren't developers, could use fully open source stacks, and they could modify software to work exactly the way they want. If we're lucky this might kill closed source software entirely in the long run.