GIMP Releases Last Development Version For GIMP 3.0

[Myownfriend]

I was just using GIMP for the first time in awhile the other day and it's still so finicky to use. Layer grouping was also broken. If I made a group, I couldn't drag anything into it. If I selected layers first and then told it to group them, it would great a new empty group above each layer. Obviously that's a bug so hopefully that's fixed in this new build. If not then I'll have to open an issue if one isn't open already.

[hyperchaotic]

Just tell us when it's final. How many more dev releases.
2.999999999999999. Give me a break.

It's in the article, you must have overlooked it - it's that funny text someone put under the hadline: they're aiming for a release candidate next.

[hyperchaotic]

Funny, how they are finishing porting Gimp to GTK 3 which is already almost obsolete, because there is GTK 4 and the last major GTK 3 release was autumn 2018 (GTK 3.24) which is currently maintained as LTS release.

The work is not exactly the toolkit, it's because they had to replace all the custom controls/widgets with modern toolkit ones and because it works quite differently they then had to rewrite large chunks of the program around it. What matters is that it's now using contemporary patterns for the toolkit.

[SyXbiT]

It's in the article, you must have overlooked it - it's that funny text someone put under the hadline: they're aiming for a release candidate next.

I meant, don’t write posts about this until it is final.
there have been SO many phoronix posts about gimp 3.
planning, delays, roadmap, dev, beta, soon RC etc…

[hyperchaotic]

Even though I definitely acknowledge the fact that the have (very) limited resources (compared to most companies) I'm a bit surprised at the fact that they mix GTK3 porting (i.e. the Gimp 2.x -> *3.x* shift) with so much feature development. I understand that it takes a long time to port to GTK3 if they spend a large portion (a majority I guess?) on developing new features while also porting to a new GTK version.
A guess is that a reason for that is that many Gimp developers are more interested in image manipulation development over framework development?

And with "many gimp developers" we're talking maybe a couple of dudes with other dayjobs, doing 2.10 work, GEGL and Gimp 2.99 when there's time to spare.

The new architecture is much more pleasant to work on, hopefully it can attract people. It holds a lot of promise on a technical level.

[caligula]

What matters is that it's now using contemporary patterns for the toolkit.

GTK3 is pretty much obsolete. You need to look at Zed editor, web frameworks, JavaFX etc. to see what the state of the art looks like. I know even JavaFX is obsolete, but it has more modern features like scene graphs, layout constraints etc. Something GTK 5 is supposed to bring.

[Pranos]

I meant, don’t write posts about this until it is final.
there have been SO many phoronix posts about gimp 3.
planning, delays, roadmap, dev, beta, soon RC etc…

You dont need to read them. Why dont you ignore the news and click on the news when it says final? Why do you post then in this forum about a GIMP news?
Iam thankful meanwhile that phoronix inform us about the GIMP status.

[Azpegath]

GTK3 is pretty much obsolete. You need to look at Zed editor, web frameworks, JavaFX etc. to see what the state of the art looks like. I know even JavaFX is obsolete, but it has more modern features like scene graphs, layout constraints etc. Something GTK 5 is supposed to bring.

Well, if it works it's not obsolete. The "if it's not newer than 2 years it's old and broken"-mindset is plaguing the programming world and I believe it's counter productive. Its the thing that brings us web-based applications on the desktop using 500 frameworks and makes a modern computer grind to a halt instead of native apps with less resource footprint.

I definitely recognize the need for development, new ideas and frameworks and perhaps even shorten development time over performance, but if the only focus is the newest architecture and design paradigm you end up as Hurd. Constant refactoring and upgrades to newer versions or even porting to other frameworks.

Or re-write everything from scratch, a'la Netscape.

[Azpegath]

When they were doing the opposite, it almost killed the project.

GIMP development is imo only as alive as it is because they changed development focus to allow other featurework and even backports to 2.10.

I see, actually that wouldn't surprise me. The functionality is what people are after, not toolkit version dependencies.

I definitely have a tendency to focus a bit too much on toolkit upgrades myself. As a Gentoo user, it's really nice to see when that old library can finally be removed because no one longer depends on it.

I like though both those that would have posted "We need colour correctness" and "Non Destructive editing or NOTHING" in past posts by now suddenly have to find new criticisms.

I don't identify as either of those, but perhaps you were talking about someone else.

[Azpegath]

And with "many gimp developers" we're talking maybe a couple of dudes with other dayjobs, doing 2.10 work, GEGL and Gimp 2.99 when there's time to spare.

The new architecture is much more pleasant to work on, hopefully it can attract people. It holds a lot of promise on a technical level.

Thanks, I didn't realise they're actually that few. I would naively guess that many image processing nerds (PhD's and the likes) would be working on it as a fun playground with the ability to reach a wider audience.

I also forgot about the actual architecture change, and that it has been more about that then the toolkit. Thanks for the reminder.