Hacker News Reader

Comment by JohnBooty | Hacker News Reader

Surely you must understand that the intersection of people that want to run graphical manipulation software and have the time and expertise to work on the GIMP codebase is rather small?

Sure, the Steve Wozniaks of the world have always had the ability to write their own GIMP plugins or whatever, and you're right: that hasn't changed.

But there were a few decades there where the rest of the world could actually own commercial software, and that had a lot of upsides. Those days are drawing to a close.

(Yes, I know that EULAs attempted to "license" software to you instead of truly letting you own the software. But for all practical purposes, you did at least own it to the extent that it would continue to work even if the software vendor's servers went down, or they decided they didn't like you using their software, or your government decided they didn't like you using the software vendor's software, or whatever.)

View on HN

Even so, in this scenario lots of people will discover that GIMP is good enough and so its popularity will rise, interoperability may become better and more people will take an interest in working on it.

Photoshop piracy was the worst thing that ever happened to the GIMP.

If your vision of how people can continue to own their data in the future relies on GIMP being improved to the point that's not unbearably painful to use, you're failing to reassure anyone.

Daily GIMP user here, pretty much the only thing you can't do is print CMYK.

The problem with GIMP isn't its featureset, but its interface. It's so foreign to anyone who works with graphics that whatever UI metaphors it uses fail to resonate with people, so they just end up confused.

What GIMP probably needs is an overhaul that copies what Photoshop does. People understand Photoshop, and it's the industry standard.

It's so foreign to anyone

It's foreign to photoshop users, GIMP is just different. I really don't like those open source software that want to copy some proprietary one. They tend to copy the mistakes of the other software too. Maybe GIMP doesn't have all the features that photoshop has but copying photoshop is not the way to go.

I've never understood the problem with the GIMP UI, what is it about it that is so foreign?

From what I can tell the UI's look fairly similar too: https://gtchan1.files.wordpress.com/2012/06/5-gui-photoshop.... http://www.gimpchat.com/files/196_2012-05-12_214936.jpg There's various widgets that contain e.g. tools, layers, history, fonts, tool options; and of course each tool might have its own options.

(Also, the team itself have said in various ways that they'll never have as their goal to be a "Photoshop clone", as I understand it that's an unreachable and thus depressing goal. It's a lot more motivating to simply try to make something that works well than trying to cater to every whim of those who don't want to pay for Photoshop.)

From what I can tell the UI's look fairly similar

They may look superficially similar, and at a high level have the same tools, but from a fundamental workflow point of view there are too many things that are just very different (non-destructive editing, smart objects etc).

the team itself have said in various ways that they'll never have as their goal to be a "Photoshop clone"

And that would be fine and if they'd done something unique or innovative with the UI and really made it their own. Unfortunately they didn't and they're stuck looking kind of like a half-baked Photoshop clone. And as long as that is kind of what they look like, then that is how they'll be judged.

> they're stuck looking kind of like a half-baked Photoshop clone

So does the GIMP look like Photoshop, or does it not?

It looks like Photoshop at a glance, but doesn't really behave like Photoshop once you get close.

Agreed, GIMP is a horrendous piece of software. I know it is extremely powerful (for example combined with G'MIC) but the fact you can't do the most simple transformations (rescale, move, strech and rotate) on the freehand tool speaks a lot about its usability.

Inkscape on the other hand really is a viable alternative to Illustrator.

Every time I try Inkscape it has been a big disappointment. It doesn't come close to old versions of Freehand, not even speaking of current Illustrator. Even indie devs like http://bohemiancoding.com/sketch/ do a lot better job at illustrating.

Photoshop's interface is awful, and the reason you like it is because you're used to it. GIMP should spend absolutely no time in imitating its mistakes.

Of course, we're offering zero evidence that either position is true, just bald assed assertions. That I vastly prefer multi-window is my only evidence.

>People understand Photoshop, and it's the industry standard.

They've been trained to. It's as intuitive to you as the way home from work. It's as intuitive to me as the way to your house from your job.

> That I vastly prefer multi-window is my only evidence.

That's ironic, given that the latest Gimp (2.8) defaults to a single-window interface. I'm having a hard time adjusting, but this doesn't mean it was a bad idea.

Not sure if you know or not, but GIMPshop (GIMP with the Photoshop interface) has been around for a long, long time:

http://www.gimpshop.com/

Still not sure why Adobe hasn't gone after the Linux market. They could completely dominate.

...or use most classes of non-destructive adjustment layers. Or use many of the more desirable and time-saving (read: cost-effective, even when compared to "free as in beer" plugins. Gimp isn't useless by any means, but it's not a Photoshop replacement yet, any more than Corel's Paint Shop Pro is. (But it will be soon, right? It's been "going to be" for a very long time.)

"Surely you must understand that the intersection of people that want to run graphical manipulation software and have the time and expertise to work on the GIMP codebase is rather small?"

Of course I do, and those people will be faced with an uncomfortable choice, suffer the 'free' tool or pay for access to the 'non free' tool. And the people who run the 'non free' tool will use that money to pay engineers to improve it. And the engineers working on the 'free' tool will do so without remuneration or reward other than knowing they built something they like.

Reply on news.ycombinator.com