Haha, I wasn't ready to post this to HN. It's not anywhere close to ready. But thank you aghuang. (I want to post this again when we clear Beta phase.)
These are still super early alpha. All MIT/Apache licensed. No telemetry.
I'm going to try to drive these to 100% parity but it's going to take a minute. And I'll need lots of help. (Anyone want to join me? I'll pay stipends and bug bounties.)
I'm going to do Microsoft Office, AutoCAD, and Solidworks too. Probably a lot of stuff to collapse.
It's not just SaaSocalypse. Everything can be open source multiplatform Rust now.
Party like it's 1999 and the Internet / tech scene belongs to us again. We can literally deenshittify the world without breaking a sweat.
nayuki•49m ago
This project looks extremely ambitious. Are these applications coded by AI by any chance? I can't find any easy documentation about the development process.
slopinthebag•27m ago
are you kidding? its all vibe coding
praet•14m ago
They are clearly all vibe-coded
4b11b4•48m ago
I like this mindset but damn that's a lot of tokens? At what cost? And then to continue to maintain and .. idk.. I suppose it'll work? If the motivation is strong enough and momentum is sustained?
echelon•44m ago
Only have to pay the price once.
I'm sick of Gimp, Krita, and Inkscape being the best Linux and open source have to offer.
dofm•34m ago
I m not really a fan of GIMP but Krita and Inkscape are both serious and competent apps with their own principles and features, made by people who care and who seriously understand their knowledge domains, and I would rather have either than AI-slop clones of Adobe apps that enshrine Adobe’s dated workflows in open source that no human wrote.
I think you can do this project of yours without being dismissive of the work (and the merits of the work) done by real people over years.
Be careful of what you might damage by doing this.
fschuett•37m ago
Just wanted to say at least for the non-creative apps I am already doing that at https://azul.rs
But its a lot harder if you want to maintain at least some level of quality. For example, your PDF export exports texts as paths and correct PDF encoding and font subsetting took months of trial and error (I maintain the printpdf crate).
So you have to actually use your apps, test extensively, etc. - don‘t underestimate the effort required for QA and testing.
diffstrokes•35m ago
Amazing work!
qchris•32m ago
I'd love to see something like this for Solidworks, but think you're in for a boatload of trouble there. OpenCASCADE is less capable than Parasolid, and a true Rust-based parametric geometry kernel capable of boolean joins, fillets, etc. needed for anything resembling a modern feature set is wickedly difficult (that's one of the reasons why everyone licenses existing ones and don't try to roll their own).
Don't let that discourage you, though! A fast, local, fully-featured 3D parametric CAD program that takes advantage of modern hardware parallelism that can be extended/scripted is just basically one of the holy grails of software and even exploratory efforts might uncover some great paths forward to making that a reality!
dofm•17m ago
Indeed. If you are going to burn tokens to clone something, then assembling a team of people who could help you burn those tokens in service of a bRep kernel (let alone a Parasolid kernel) would be an interesting undertaking.
With photoshop, illustrator, etc., there’s a fair amount of code out there. No LLM has however ever seen the source code —- or I suspect even much detailed discussion —- of a bRep kernel better than Opencascade, but with the right people to decide what maths to work on, maybe there is progress to be made. This knowledge is held by few people, many of whom will now have retired!
brookst•16m ago
It’s not what you’re after but onshape is pretty respectable, and the free offering is decent.
rancar2•20m ago
The videos of the features are very well done, and it looks like there is a unified vision, which is not what any of the other products have. Did Google copy some of the UI/UX? I was curious where their inspirations were, and I think this may have been it.
I look forward to the beta and beyond!
cjonas•16m ago
Despite whatever else happens with AI, I am excited about the near future where we have an open source version of basically every app. I hope this is the beginning of the end of the subscription model. Keep it up!
…which was bought by Canva and has already been enshittified.
Gigachad•18m ago
It really hasn't been enshittified. They made everything that already existed in Affinity free without any catches. And then added some new AI powered editing tools which require a subscription. Tools which you don't need and can ignore easily.
slopinthebag•27m ago
i like this project, the scope is enough to test most of the claims being made about vibe coding
Gigachad•16m ago
I'm interested to keep an eye on this. Doesn't seem like it's in a usable state right now but if it ends up workable this will really shake up the entire software industry.
colesantiago•25m ago
As soon as I saw this software suite about 2 days ago, I shorted Adobe Inc immediately.
Adobe to $120 price target at least, could be even lower.
brookst•18m ago
Same for Microsoft and others. There is still a business in selling and supporting software, but not in creating it, and barriers to entry are now purely about sales and service.
somacore•11m ago
These companies are going to start tagging files not made by their software as not being "genuine" and freak out a bunch of IT people, who will then ban these apps.
Svoka•11m ago
Are we looking at the same stock? I am confused where this data is coming from. What I see is stock slightly higher than it was week ago, fell compared to day ago, and the change is 1-2%, way less than you 50% fall claim.
praet•15m ago
Adobe's moat is so small, so it's a matter of time until someone frustrated with their pricing would rather just rebuild the entire Creative Suite. I have also rebuilt the features from Photoshop I want using Rust recently (primarily because the company I work for started blocking Photopea, but I digress...) and it's been a great to work with. I can imagine the same would hold true for Lightroom, Illustrator, and Indesign.
Overall, looks like a cool project. Looking forward to the *Craft repos on github gaining traction
nonethewiser•3m ago
Pretty crazy. I think we’re already very numb to AI. Can you imagine seeing something like this pop up 3 years ago? Youd have been dumbfounded and it would have been major news.
echelon•56m ago
These are still super early alpha. All MIT/Apache licensed. No telemetry.
I'm going to try to drive these to 100% parity but it's going to take a minute. And I'll need lots of help. (Anyone want to join me? I'll pay stipends and bug bounties.)
I'm going to do Microsoft Office, AutoCAD, and Solidworks too. Probably a lot of stuff to collapse.
It's not just SaaSocalypse. Everything can be open source multiplatform Rust now.
Party like it's 1999 and the Internet / tech scene belongs to us again. We can literally deenshittify the world without breaking a sweat.
nayuki•49m ago
slopinthebag•27m ago
praet•14m ago
4b11b4•48m ago
echelon•44m ago
I'm sick of Gimp, Krita, and Inkscape being the best Linux and open source have to offer.
dofm•34m ago
I think you can do this project of yours without being dismissive of the work (and the merits of the work) done by real people over years.
Be careful of what you might damage by doing this.
fschuett•37m ago
But its a lot harder if you want to maintain at least some level of quality. For example, your PDF export exports texts as paths and correct PDF encoding and font subsetting took months of trial and error (I maintain the printpdf crate).
So you have to actually use your apps, test extensively, etc. - don‘t underestimate the effort required for QA and testing.
diffstrokes•35m ago
qchris•32m ago
Don't let that discourage you, though! A fast, local, fully-featured 3D parametric CAD program that takes advantage of modern hardware parallelism that can be extended/scripted is just basically one of the holy grails of software and even exploratory efforts might uncover some great paths forward to making that a reality!
dofm•17m ago
With photoshop, illustrator, etc., there’s a fair amount of code out there. No LLM has however ever seen the source code —- or I suspect even much detailed discussion —- of a bRep kernel better than Opencascade, but with the right people to decide what maths to work on, maybe there is progress to be made. This knowledge is held by few people, many of whom will now have retired!
brookst•16m ago
rancar2•20m ago
I look forward to the beta and beyond!
cjonas•16m ago
timschmidt•2m ago
You may want those in new high-performance infinite precision: https://github.com/timschmidt/hyperreal
Exact curves: https://github.com/timschmidt/hypercurve
Meshes: https://github.com/timschmidt/hypermesh
and more in the ~22 crate stack.