I think they should focus on incorporating LLMs and stable diffusion and make them more easily accessible.
Is this radical redesign of GIMP possible now?
141–150 of 306 posts
Re: Is this radical redesign of GIMP possible now?
#142Earlier quoted context omitted.
> At that point you realize that what you really want is MS Paint from Windows 3.1 Or a drawing program like Krita.
Or Paint.NET, which does maintain aspect ratio on shift.
I switched to macOS several years ago and the only thing I still miss from Windows is something like Paint.NET.
Re: Is this radical redesign of GIMP possible now?
#143Just to contrast the negativity here I'd like to say that GIMP is great software, and I really appreciate all the hard work that has been put into it over the years. It's powerful, easily scriptable, and I'm eternally grateful for everyone's contributions to the project. https://www.gimp.org/docs/
Have you actually been using it?
Re: Is this radical redesign of GIMP possible now?
#144All these programs like GIMP, Blender, Dune, should split into two. There should be a library that has the functionality and data structures to manipulate data. The UI should be a separate project with different people. There could be multiple GUI's for the same project or same GUI framework for multiple projects.
Sounds like you are sort of describing GEGL in the context of GIMP, though it would be difficult to draw that abstraction in a way that makes implementing competing GUIs straightforward or DRY.
Re: Is this radical redesign of GIMP possible now?
#145What about an integrated LLM assistant for GIMP? Seems like that's the future for any professional application (e.g., PhotoShop, CAD, Reason)
Sure. Then we could have a Beowulf Cluster of them.
LLMs on consumer grade hardware are a thing and will become more so. Something that only needs to understand the space of the application is certainly attainable.
Re: Is this radical redesign of GIMP possible now?
#146Earlier quoted context omitted.
It would not shock me to learn that GIMP has been a multi decade study to find out how bad a UX has to be before the FOSS community starts to care.
GIMP has one of the best user interfaces of all community driven open source software. I frankly can't think of any that are better, now that Firefox followed Chrome in hiding all its most important features.
> GIMP has one of the best user interfaces of all community driven open source software.
/s
Re: Is this radical redesign of GIMP possible now?
#147Earlier quoted context omitted.
Confirming what the GP said? This is all misdirection. Bottom line the GIMP UI/UX is garbage. There are extremely few people that have used a variety of image editing tools and claim GIMP is actually better. That’s all that matters really. This isn’t some rhetorical exercise.
The previous poster was talking about Krita (and the one before that Krita and Gimp) and while I agree with some of the criticism of gimps UI, I stand by my comment a lot of that criticism is based around "it is not like program X, that I'm used to". I mentioned it in a different post, blender is the prime exhibit for this, its UI received even more criticism than gimps. Now that it has become the (or one of the) mos…
Wat? Blenders UI was revamped. Even so, prior to the rewrite it wasn’t such a sorry mess as Gimp, it already had some professional adoption.
And my point is, regardless of those claiming gimp sucks because it’s not what they’re used to: it absolutely does not matter an iota because there is near universal acceptance that it’s terrible for anything but toy work. UI aside, internally it isn’t fantastic either, a problem that blender didn’t have.
Re: Is this radical redesign of GIMP possible now?
#148The one thing I can say is that fuzzy search like IntelliJ is a really good interface for experienced workflow. Maybe an AI fuzzy search would help with discoverability.
I was thinking this would be a good use case for AI. Perhaps you could have a Gimp co-pilot that makes suggestions about what you could do with the image after uploading it to a server somewhere. Then we would know if the AI is more or less intelligent based on whether it can actually figure out how to find that specific blur function within all of the various dropdowns and sub-menus.
Requiring a copilot is a proof for bad UI design.
Re: Is this radical redesign of GIMP possible now?
#149The search tool for GIMP is Google. How do I draw a rectangle? Google brings you to https://docs.gimp.org/2.10/en/gimp-using-rectangular.html which starts with "GIMP is not designed to be used for drawing." and then goes on which "However, you may create shapes by either painting them using the technique described in Section 14.1, “Drawing a Straight Line” or by using the selection tools" At that point you realize th…
Or DeluxePaint II from the early 1990s.
I still haven't seen a truly-adequate replacement for that. Every time someone tries to create one, they begin with good intentions and end up with a bad Photoshop clone.
Re: Is this radical redesign of GIMP possible now?
#150It's better than what currently there is but why does open source software have to have such bad UI? what are those lock and checkboxes? a 5 year old could come up with better icons... What's up with the thick lines around UI elements? it all looks so 2008. It sometimes looks to me like absolutely 0 designers are actually involved in open source, at all
GIMP is a power user tool, like Photoshop, or Excel, or a Bloomberg terminal. If you don't need all that, Canva or the photo editor on your phone will offer a simpler UI.
No "power user" is touching GIMP with a ten foot pole.