Post

gimp is kinda jank yeah but when people have Serious Takes about how it needs to improve all i can read is "here is how these four guys working intermittently for free have FAILED to live up to the international company making twenty billion dollars a year"

Chris Person@papapishu.bsky.social

GIMP is objectively unintuitive and also can’t parse text layers of PSDs properly, it’s designed by FOSS people that have never created tools for professionals. Telling someone to use it for their job is like saying you can make a feature on iMovie. Technically true but not serious.

"personally i would just make it be good" great, stop fucking posting about it and get to work then

this is frustrating in a way that's hard to fit in a bleat but it boils down to people not appreciating the outright miracle that open source software is in the first place and then wanting to be treated like a customer when it's not an exact replica of something that costs $700

the thing krita and blender have in common is that PEOPLE GAVE THEM A LOT OF MONEY

Lufty@lufty-luft.bsky.social

But still... blender actually got some traction once they decided to make things more user friendly. Krita's newer than the GIMP, and much more popular. UX is something that does matter, even making allowances for 'free and volunteer made' open source software.

the developers of other foss software aren't idiots waiting for someone to tweet "have you tried making good ux?" at them — they just don't have the resources to hire ux designers and do user studies and whatever else because they are giving their fucking work to you for free

7 reposts101 likes

It would be cool if we could do the free open source software thing without both sides having some sort of pretentious anger associated with it. I definitely understand that a lot of these developers and Linux enthusiasts are tired of people asking for help or complaining.

However, I would argue that putting some effort into making the software easier for people to use goes a long way in solving that problem. If multiple people are having a similar issue with my software, that's a signal to me that I need to do something better.

Consumers are always going to be dumb and silly, so the best you can do is partially tame that by implementing at least some measure to solve their problem or to at least defer the problem somewhere else.

In the case of Photoshop or even substance painter, I have yet to find a legitimate alternative that actually covers all of the bases. For example, I paid for clip studio paint because it's a pretty good software. However I literally can't paste into a mask which I think is incredibly basic.

ok i think there are a couple particular factors here ① sometimes people treat FOSS as though the thing it's /supposed to be/ is "software i already use, but free, and without any of the stuff i dislike". but that rarely happens — it's creative work made by different people with different ideas

② "why not just copy photoshop" is really the same genre of Idea Guy post as "why not just make a pokémon game with every region in it", but software isn't often recognized as a creative endeavor so a lot of serious professionals nod along (and then i look like a lunatic for finding it grating)

③ when google is free and photoshop is """free""" (on the pirate bay) and programming is basically regarded as some sort of wizard activity anyway, it's very easy to go "well why isn't this free thing as powerful as this other free thing"

④ similarly when all software is interchangeable the blame gets reversed. case in point, OP complaining that an open source project doesn't read some detail of psds correctly, rather than "oops we all standardized on adobe's file format and that's a form of lock-in"

I mean, if they dont have the resources to hire a UX designer and to run studies, then it would make sense to use the design off of some other company that does right? I mean, its basically getting UX design and studies for free.

at best you become a shoddy knockoff with no identity of your own, and i cannot imagine anything more joyless to work on and then what do you do with all your existing features that don't map to the ui you're just copying

I think it also is another thing: some people actually LIKE the 'bad' UX, so it puts them in a weird position: if they did honestly redo everything, it'd make it lose the identity that a group wanted.

Remember as well that UX is subjective and has many dimensions. Some people prefer to have direct access to a lot of complex features for efficiency, others prefer a clean and simple layout to avoid being overwhelmed. Both are valid. We're biased towards learnability for software we don't know well.

There's also the issue that much of the FOSS userbase is deeply hostile to telemetry as a concept. And making it opt-in rather than opt-out also means that, if you do have telemetry, the sample is the people sophisticated enough to know how useful telemetry is, the opposite of your targets.

Another thing people ALWAYS gloss over is that Photoshop has bad UX as well. More people are just familiar with it. I learned on GIMP and got used to it, and liked it. I decided to "get serious" and tried learning Photoshop and it was MISERABLE, the workflow was backwards and everything was hidden

yes the few times i've been exposed to photoshop i haven't been able to find a goddamn thing. people do regularly confuse "thing i've been using for twenty years" with "good UX"

I mean in a certain sense it very much *is* good ux to leverage existing familiarity, especially if existing users of previous versions of your thing are the vast majority of your addressable userbase

I learned the GIMP in like 1999 (RTFM & experiment), with no prior pro image software experience. When I've attempted to touch pro software, I've been extremely lost & found it unintuitive. Familiarity provides a huge bias about what feels natural & sensible. I think people can forget that

Also, if you pick up a tool expecting to accomplish a certain goal, without knowing how to use that tool to reach that goal, discovering the process will feel difficult bc of learning detours. Picking up a tool just to experiment & learn how it works, then using the tool for a goal, is much easier

Yeah, for me I used it mid 2000s, but I had just learned a bit of Photoshop, so my muscle memory between ALT and CTRL was pretty conflicted. Granted, I was able to figure everything out (I'm not an artist), and could accomplish everything requested in the classes, but I also get that was different.

OTOH sometimes it’s kind of both. Was one of the core developers of a FOSS project that got investment and hired actual UX professionals to tackle areas users has persistently complained about. There was a shocking amount of pushback, even anger and outrage, by devs who insisted it was a waste…

… and that the effort was “just time wasted by pixel pushers who can’t write their own patches,” etc. Things have improved over the (decade+!) since then but the UX team was p badly burnt by it, and it taught me a lot about the fact that simply resourcing something doesn’t change a community.

that sucks. can't imagine what the hell was going on for a project to hire UX people and then be like, surprised, when they have UX suggestions

hmm i bet the friction here is that people often work on foss projects to make software that Does something, but a lot of UX programming is just rearranging existing stuff, and after hours/days/weeks of work your software doesn't really Do anything that it didn't do before

The dynamic you mention (tool made by/for devs also has a crowd of non-decs wishing it were easier for them) is real, but also when FOSS projects and advocates get Big Ideas about project evangelism and decide a project should be used by everybody… some of them morph into “ux is just fluff”

this makes me realize — adobe surely has separate core vs UX teams. but a foss project with only a few major contributors doesn't! it's the same people. so do they prioritize "mere" UX or e.g. better colorspace support (which i've also seen cited as a reason gimp isn't "serious" software)?

My issues with GIMP are less about the UX and more about how the parts never fit together quite right and how some of the tools’ quality is lacking (in terms of how it handles blending etc.) and when I’ve made suggestions for improvement the immediate response is “you have the source, you fix it”

and like sure it’s valid that they’re overworked and their priorities are elsewhere but it’s always been an attitude that shuts down my willingness to collaborate and actually work on the code because having the source is different than being able to work on an existing complex codebase

I do respect a lot of what GIMP has done for F/OSS in general and GTK in particular, but to me GIMP is very much, to use a hamfisted analogy that will probably annoy you very much, the PHP of art software.

you cannot bind right-click to colordrop (or anything else actually) so i am less annoyed than you might think

also as someone with mobility impairments, the name does bother me. I know they didn’t intentionally name it after an ableist slur, but the fact it happens to be one is still a problem.