I just tried pdf craft on fdroid and really liked it, but its on beta and only 2 versions released.
Is it going to be well maintained and become an alternative or just stay unstable ai dlop? It had more stars on github than some other pdf viewers.
Whats your take on that. I am excited to see some decent apps on linux. Especially for .appimage and launching instantly unlike photoshop.
Believing these apps will succeed is based on an assumption along the lines of LLMs being able to produce quality code at 100x the speed of a competent human engineer. Maybe they can produce slop that fast, but using LLMs in a disciplined fashion to produce a quality product usually means LLMs are only providing a modest productivity boost in my experience.
It’s not realistic that these apps are anything but slop given the timeframe and scope spanning multiple complex apps that each took a decade or more to develop.
In my 20+ years of experience as a software engineer, I’ve never seen a successful project that hastily pumped out a massive pile of buggy code and then went back and fixed all of the problems later without doing a complete rewrite. You have to build incrementally and bake in quality every step of the way. LLMs may make the process a little faster, but the same rules still apply.
A poorly built codebase will always eventually collapse under its own weight. What I anticipate is that they’ll play whack-a-mole for a while and then give up. If they were targeting a small subset of one app’s features over a timeframe of months, I might believe it, but several massive apps all at once and it will all be done weeks from now? Stop bullshitting yourselves.
Things are different now. AI can do a rewrite / refactor all night and fix whatever issues were there. Also these are offline apps, so it’s not like they have to do migrations or care about uptime. I’ll wait and see how things go.
I’d sure like to see Adobe and their ilk get their comeuppance, so I won’t mind being wrong about this. We shall see.
I mean the majority of functions in the adobe suite are either not present or not working properly. It’s buggy nonfunctional slop and there’s frankly no world where it improves in a major way and stays supported long term
I mean I love that they are sticking it to Adobe but they are slop by nature. If it’s at least useful slop or if you should just use the actual open-source alternatives is debatable, I don’t have any talent with these programs anyway. I’d say learn the open-source alternatives but maybe try the artcraft stuff if you truly can’t do something in the open-source program, but that’s just my uneducated opinion.
100% slop
I wouldn’t trust them to be long for the world, that and IMO they’re legally dubious.
Just use one of the established alternatives like GIMP/Krita, RawTherapee/Darktable, Inkscape, etc.
My tinfoil conspiracy theory is that artcraft is actually a 5d chess move from the “anti-ai” camp to draw bigtech attention and get them to turn against LLMs on principle, or at least get them to implement some sort of anti-app-cloning thing like anthropic recently did for games.
Not really but that’d be funny i suppose.
I’ve checked its girhub before and photocraft has a lot of contributors, over 400. But the apps are made mostly by claude code. Personally id never use these apps but so far theyvebeen receiving updates quickly with humans adding new bugfixes (humans or maybe more vibe coders…).
I installed the flatpak out of curiosity and these artxraft apps receive updates fast in my opinion, a d they add more things like a reverse engineer app or solidcraft thing.
in my opinion, even with human bugfixes and that, its still slop, I trust other foss alternatives.
This raises another question I’ve been kicking around in my head for a while - at what point does a slop project stop being slop? Say I one shot a slop app and then continue to iterate on it with human-written commits. Or I find a vibe coded app I think has potential, fork it and then iterate on it. At what point does this become non-slop, or does it ever?
In my opinion is possible to stop being slop, if an app is very well made thanks to human checks (like fix garbage code made by llms). Llms can’t be trusted to competely write code imo, but if a good programmer knows how to use llms to improve their workflow I think that’s a good llm usage.
artcraft apps are too new and still have a lot of bugs, but maybe someday it will be more useful in my opinion.
If the code is checked by humans, like on the linux kernel for example, id say I trust it. And I dont think its slop.
and llms have great uses for programming in my opinion, like the opensuses OpenQA, etc. ai has found cves on linux too which is a great usage for it. what I disloke its that some people think that by using tons of tokens they can be programmers by just prompting something…
Artcraft, has a problem that they release quantity over quality, there are like 15 apps now and they aren’t well tested… but who knows, maybe in the future it gets better. Maybe id change my mind of it being slop
I keep hearing people talk about trust, and it’s really doesn’t make sense to me because we’ve never done software development based on trust. Humans make mistakes all the time, and I’ve seen plenty far worse human code than anything an LLM has crapped out over my career. The way we ensure quality is through having a process. It’s done by having good specifications, testing, code reviews, planning, and so on. All these things equally apply to code written by hand and that’s generated by an LLM, and trust has nothing to do with it.
I tried photocraft and it’s certainly usable. The video editor however wouldn’t play anything I added to the timeline at all. I also tried the MS word clone and it’s just janky-looking (like the text formatting seems to change if I click on the page? Hard to explain). I’d take pirated PS over it anyday though. One thing I can say is that photocraft is very fast to launch compared to Photoshop.
I’ve been using them happily, and they do what I need. So, to me, this is real useful software that’s replacing proprietary solutions.
