*Slop 'developer'
post
I would rather have a Windows 11 copy without all the built in spyware and bloat hogging up performance, RAM and storage.
Y'all are missing the bigger picture here. This brings up the question of ai and intellectual property and throws the ball into adobe's court to set their army of lawyers on figuiring out a no-win solution for themselves.
- If they do nothing, a working clone normalizes copying their product's behavior and UI. Copyright covers that thinly anyway (the First Circuit in Lotus v. Borland treated a command menu hierarchy as an unprotectable method of operation). Trademarks need policing!! (And they suck)
- If they go after it, they have to argue in court that AI output built from public material is infringing or derivative. That's the precedent the AI industry wants to avoid, and it cuts against Adobe's own Firefly pitch. Imagine what that means for "art" made with an LLM
- If the clone's code is mostly AI-written, its copyrightability is shaky (Copyright Office, Thaler). But that cuts both ways: Adobe can't claim "you stole our code" without proving actual copied code. Which bites into OpenAI, Anthropic, Google's claims that they haven't stolen anything.
Edit: Sorry I should've used Super Intelligence, I don't want to be the enemy. /s
Thank you for the reimplementations of Adobe and Office, have not laughed this much in a long time.
Adobe VS AI slop, who would win ?
does it actually work though? Anyone tried it?
I have the whole suite of apps installed as a Flatpak. The Photoshop clone is the fastest one to implement features, but the whole suite is moving really fast. Looking at the commits and features shipped, I haven't seen anything like this. It looks very well organized.
I have been using Adobe Photoshop, Illustrator and InDesign professionally for ~30 years. These Craft apps are freaking amazing, they open in the blink of an eye and open multi-gb files just as fast. Working with these big files is also instantaneous.
I have done my own calculations, and what they say is true, the Suite will reach beta status by mid-late October depending on the app. But the suite in general will be ready with more features than what a basic-intermediate user would want.
I think what will make or break the Suite is if they go with local AI or not for Neural filters and generative fill. The community might turn on them if they don't.
But technically speaking It's truly an amazing feat.
These are my own projections, so it's not official:
| Repo | Ready today | Hours to beta | Beta (central) | Range |
|---|---|---|---|---|
| vectorcraft | 62% | 170–270 | Oct 13 | Oct 12–15 |
| lightcraft | 61% | 340–600 | Oct 15 | Oct 13–19 |
| photocraft | 45% | 1,100–1,800 | Oct 22 | Oct 17–31 |
| effectcraft | 45% | 350–620 | Oct 22 | Oct 17 – Nov 1 |
| filmcraft | 56% | 400–700 | Oct 22 | Oct 17 – Nov 1 |
| printcraft | 49% | 400–700 | Oct 22 | Oct 16–31 |
| designcraft | 47% | 435–775 | Nov 26 | Nov 3 – Jan 4 |
Does it matter? If it works we can laugh at Adobe, if it doesn't we can laugh at "AI", and however it goes we can probably laugh at Adobe vs "AI".
The fact that everyone despises Adobe lately (last few years?) for their extremely anti consumer practices makes this case so much more difficult to judge.
On one hand, it is a blatant steal of an intellectual property. Not the code, but the UI, the philosophy behind the programs, with zero creativity added. On the other hand, fuck Adobe.
It's still an early work in progress but it has very high velocity with PRs and issue activity:
https://github.com/storytold/photocraft/pulse
Something to keep an eye on and see how they progress against their roadmap:
https://github.com/storytold/photocraft/blob/main/ROADMAP.md
Lmao, why do people keep insisting it has 100% parity when it obviously doesnt. It might actually be usable and i wish a thousand deaths upon adobe, but claiming that its a drop in replacement is just a lie.
They say it's gonna be 100% parity in a month. Sand Altguy also claim we gonna have AGI in the next 6 months and Melon Husk claimed there will be full self driving today. We'll see by next month.
I'll be happy to check back in a month to see which toxic cesspit provides the best approximation of Photoshop-tier quality.
My bet isn't with the Redditor whose profile reads "To AI haters: AI is just a tool... Adapt." But their attitude is roughly as toxic as Adobe if you strip away the PR, so let them fight I guess
I'd guess you are right, but I don't know because I haven't tested it. I'd say: if this guy really devoted a solid month of focused LLM development time on the job, and knows what he's doing, there's a chance he's pretty close to what is claimed in TFA. But I haven't tested it for myself, so I just don't know. I'm guessing you are right because most of these projects aren't done by people who know what they're doing, and aren't real heavy effort even during the month they work on them.
Software defies economics. It can be replicated almost infinitely at almost no cost. When software companies sold software one package at a time in the 90s anti-piracy was a big deal because it was very simple for any given person to "copy a floppy" and receive the full package of software from another individual without the company knowing or being involved. It was like borrowing a rake from a neighbor, but the neighbor got to keep his rake as well.
Because of this, software companies decided instead to treat software as a service rather than as a product. In the 2010s many software companies shifted to the subscription model where instead of selling software one package at a time, the service of the software would be available to rent while the company guarded the propriety aspects of their software. This was like the company owning the rake and both you and your neighbor needed to pay the company to borrow the rake.
Now software can be reverse-engineered without the source-code. Now if you rent the rake from the company, you can keep a copy of it and freely distribute it without the company being involved or aware. I'm not sure how software companies will reckon with this advancement.
just as a curiosity what happened to the Adobe stock price ?
what's he's suggesting was the entire reason for the "SAApocalypse"
this guy could perhaps have done a Trump and made a bucket shorting Abobe BEFORE his "announcement"?
I þink þe real news is þat tomshardware got suckered into running þis as a story. Þere's so much abandoned slopware out þere just as complete.
Most who've tried to do professional-level stuff report that it doesn't really work. When first released, the "developer" said he "one-shotted Photoshop, the whole thing." Then he back-peddled all the way to, "It's just an alpha release give us a year." Now I guess it's a month again? Whatever. It's a UI framework with some basic functionality. There are free alternatives created by real developers, which have greater feature parity. Why does everyone want to think this knock-off junk is the nail in Adobe's coffin when Affinity and Photopea weren't? (Yes I realize there are other applications in the suite.)
The guy who makes this is also not really great or trustworthy. He enjoys seeing people lose their jobs to AI, and exuberantly telling everyone that AI made him rich, if it didn't make you rich than you didn't "dream enough" and are doomed.
For years he's made bizarre claims like by the end of 2025 AI will kill the entertainment industry, and AI will enable us to upload our brains and colonize the moon. He regularly takes weird, aggressive positions about every day things, like the economic impacts of owning a bicycle and whether or not a woman can ride one while menstruating.
He clearly suffers from delusions of grandeur and AI psychosis, but worse is that a lot of this is just marketing. This Adobe thing is not a benevolent endeavor to democratize software. Its purpose appears to be to bring attention to his token-reselling AI image and video generation service. He also owns FakeYou, a commercial service used to create audio deepfakes and do voice cloning for scams. To me it doesn't look like he wants to erase Adobe for the good of mankind; rather he wants to usurp and replace them with his own nickel-and-dimed-for-tokens scheme.
Yes we've all heard about slop-o-shop by now
Would still prefer it over fucking adobe cloud-shit though. Hate aside.
See how compatible it actually is, when there's no way any meaningful regression testing could have happened in that timeframe.
One of the TomsHW commentors made a good point: if this slop actually gets to a state where it functions properly for a lot of workflows, then it can help people chained to Windows because of Adobe products be liberated and move to Linux.
That developer will get bored once they realise how much of an effort this actually is. This is an admitted pre-alpha yet everyone acts like Adobe is crushed by this and not by competition from Affinity and Canva.
Alternative title: “Solo dev delusional enough to think they can create an enterprise software suite without years of in-house testing or user feedback, because typing code is the bottleneck in software development, apparently”.
You tell them what’s broken and if they’re still around they ask Claude to fix. Next thing you know they’re begging for money for tokens. Amazing waste of electricity all around.
I agree that it's delusional to replace a company with more than 30000 employees with a bit of LLM and a single motivated dev. Yet, with the rapid progress over the last two years, I do see a realistic risk for standard software in general.
Adobe software can cover thousands of use cases with all their complixities and in all combinations. A single customer usually doesn't need that full set, but only a fraction of these features. So far, it was the rational decision to just buy a capable software off the shelf because development, maintenance, migration etc. for a custom solution was too expensive compared to the license fees. And there I'm quite sure that it will change in the future. Building something tailor-made is getting cheaper and cheaper and might result in more efficient workflows.
Writing code is such a small part of software development that even if you forget about designing cohesive whole, refining features and testing, you still have to consider that software has users. You might create bespoke software but where are you going to get employees that are skilled at using it? Do you want your graphics designers to moonlight as your primary QA? You can’t escape putting in the effort.
all 2 comments