Photoshop alternative

[koraks]

Won't AI allow newcomers to start their own editing sites pretty easily and compete with Adobe and others?

You could prompt it by saying: "Reduce the brightness of the sky and increase contrast in the foliage on the right side. Eliminate all dust spots. etc. Slightly blur the background behind the person. Add a small black border around the picture."

There's two ways AI can and will affect this. The first is by what you said; i.e. integration of AI/LLM technology into the editing suite as such. This is already happening; the kind of prompt you just made can already be given to Copilot, Claude etc. and it will give a re-made image back to you with your specifications. Like @_T_ points out, it will technically not be the original photo anymore, but we have plenty of other threads to address that issue...(so please, let's stick to the question of Photoshop alternatives here and by extension, the ecosystem of photo editing suites, not photographic philosophy/ontology; thanks).

The second is also happening: AI enables smaller teams with less competence than before to build apps, including image editing apps. In the foreseeable future (12-24 months or so) I estimate that it is within reach of an amateur user with a modest budget to have e.g. Claude build an alternative to something like Affinity that maybe not yet covers all of its features, but enough of them to constitute a viable alternative for many users. It's very hard to tell how this may/will affect the landscape. Presently, the codebase is a bit of a bastion; it's complex, takes experience to navigate & modify and thereby represents an encoding of the collective application wisdom of the developers into an implementation. If this codebase becomes more of a flexible sideshow that anyone with a Claude sub can re-generate to their own desire, instead of a crucial asset, the net result may be something like a huge proliferation of 'editing suites' that are tailored to the individual needs of the user. In other words - we may no longer be using Photoshop, Affinity etc. as such, but instead use a self-generated derivative that suits our own needs, where these needs may evolve on a daily and on a job-by-job basis. That horizon seems further away than the 1-2 years I mentioned; in that timeframe I expect the landscape to be dominated mostly by 'indie' alternatives popping up and making a mark on competition in the industry.

To illustrate the second point, please see the several apps and solutions that have been popping up for specific purposes here on Photrio already. We have people coming in who have presented their own negative scan inversion software, web-apps for film and paper calibration, someone posted an interface utility to use old ProPalette film recorders the other day...and surely, there's no end this in sight. Quite the opposite. I think it's a matter of time before we see the first real alternative to a photo editing suite emerge on this forum or Reddit made by a private individual who up to a year ago would not have been capable of producing even a simple desktop app.

 

[gbroadbridge]

made by a private individual who up to a year ago would not have been capable of producing even a simple desktop app.

Of course, that individual is still not capable of producing even a simple desktop application, because the use of AI in that way imparts no knowledge and is a dead end.

Unless AI is used to create new knowledge it is of entertainment value only.

 

[koraks]

Of course, that individual is still not capable of producing even a simple desktop application, because the use of AI in that way imparts no knowledge and is a dead end.

No, you're stuck in backwards thinking. If someone has no coding skills, but they can produce working & functional software, then that's an observable and inescapable fact.

There's also a parallel within software engineering as it used to be: someone could be capable of writing in a higher-level language (say, C++) without knowing the first thing about register-level access, machine code and no ability whatsoever to write assembly language, for instance. This generally does not stand in the way of successfully producing functional code.

It's the same with AI, which isn't all too different from an entrepreneur hiring a software engineering agency to solve a problem. If the solution works, then there's no dead end.

 

[brotz]

Among whom, based on what data, what does 'unbearably bad' constitute" and what's "a couple"?

Welcome to Photrio btw!

You are right to challenge my statement, it was a bit brute.

I was looking at free software for minimal photo editing in an office that I work in sometimes. A friend of mine recommended Affinity, because it's free since January. He raved about it actually.

Me personally, before I install anything that's free and potentially transmits my data around the world I look into the motives of the people providing it.

So I checked on Reddit and over there a lot of people were of the opinion that it's almost playbook for a word that, as a new user, I don't think I should say. Ensh..cation. Getting a large userbase and over time allow certain functions to be only used via paid models.

Getting comfortable in a piece of software takes some time and I don't want to risk heading into a direction that could feel uncomfortable down the line.

For my use case, I am sticking with GIMP, even if it can feel a tad cumbersome coming from Photoshop.

Thanks for welcoming me here, @koraks.

 

[Romanko]

Exactly what @brotz said. "Enshitification" is an established term in the software development world. I see no problem in using it in this forum.

Serif, the company that developed Affinity, stated that they will never sell the company. They did. When Canva purchased Affinity, they announced that they will always keep Affinity as a subscription-free product. They lied.

"Free" Affinity Photo version 3 saves the files in a format incompatible with v2 that I purchased a few years ago. I can easily imagine a situation where version 3 is discontinued and in order to get access to your files you need to buy a subscription-only version 4. A couple of years is an average period between major versions updates for stable commercial software.

While I am happy with my version 2, I no longer recommend Affinity Photo and I am looking for open-source alternatives.

 

[koraks]

Okay, I see; thanks for clarifying. Personally I think it's always good to be cautious, and at the same time I like to ask myself whether I could be benefited from something even if it has potential (or future) drawbacks. In my case, I've had Affinity installed since the free release they did a couple of months ago. I occasionally use it and the learning curve has been quite minimal, based on prior experience on Adobe and GIMP. In the end, all these tools do more or less the same thing although the user interface and workflow are subtly different.

I have to say as regards the ensh... or the potential thereof, is that so far I find Affinity relatively well-layed out and easy to use. I do stand by what I said earlier about a potential for lock in at some point, but then again, if I ask myself whether I'd like to be locked in to a product that I already have to pay a monthly fee to use (e.g. Photoshop) or one that I might have to pay for in the future (e.g. Affinity), I generally side towards the latter due to lower barriers of entry. And yes, that does mean there's a potential that "I am the product" as is sometimes said about 'free' products/services. However, so far, I don't see much potential for Canva (maker of Affinity) to do something with my usage or content data that I wouldn't feel comfortable with. Very bluntly said, if they 'see' or collect data on how I invert a color negative, well, be my guest. I hope they use that data (assuming the collect it in the first place at that level of granularity, which I doubt) to make something that other people find beneficial.

As to security, there's always the possibility to disallow traffic to certain servers or from certain apps so that you remain in control over what data is being transmitted. I think that could take the edge off for some people.

Evidently, we all decide differently; that's OK. I'm always interested to hear why others make the choices they do as it helps me figure out what I find important myself. Your response helps in that regard, so thanks!

"Free" Affinity Photo version 3 saves the files in a format incompatible with v2 that I purchased a few years ago.

IDK, most of the time when I save anything it's really an export into TIFF or JPG.
I do occasionally use GIMP's native file format to save something, but that's very, very rare. And frankly, in that scenario I'm basically locked into GIMP's ecosystem and I just have to hope that later versions will be able to correctly interpret earlier saved files (although it's exceedingly rare for me to run into that scenario to begin with). Mind you, there's no doubt in my mind that later version of GIMP will in principle/theory support older file formats, but will they be interpreted correctly? I have very severe doubts about this. GIMP presently still struggles interpreting correctly what you do in terms of dynamic filters etc in some scenarios within the same working session - let alone between different versions etc.

In one case you're locked into the commercial model of one party, in another you find yourself locked into an ecosystem that's uncertain to remain existing and that you also have no control over. There's never an ideal solution; we always compromise. I don't think there's ultimately going to be a final solution to any of this.

Much of the time I think the question is how big of a problem we want to create for ourselves. While we have little or no control over the ecosystem as such, what we do control is our own requirements/criteria, and to a large extent our knowledge and competence. I personally find it more productive to work on that end (where my work actually results in anything) than to fuss over the part of the equation that's beyond my control to begin with.

 

[koraks]

"Free" Affinity Photo version 3 saves the files in a format incompatible with v2 that I purchased a few years ago.

Is it a realistic expectation that a manufacturer will ensure forward compatibility ad infinitum? I don't think so. Have you ever built a technical system involving architectural choices and interface definitions? How possible has it proved to maintain development within the same architecture over a longer period of time while adding significant innovations to that architecture?

Why do you think architectural and radical innovation exist in the first place? Do you perceive them as fun toys for manufacturers to torture their users with?

 

[Alan Edward Klein]

There's two ways AI can and will affect this. The first is by what you said; i.e. integration of AI/LLM technology into the editing suite as such. This is already happening; the kind of prompt you just made can already be given to Copilot, Claude etc. and it will give a re-made image back to you with your specifications. Like @_T_ points out, it will technically not be the original photo anymore, but we have plenty of other threads to address that issue...(so please, let's stick to the question of Photoshop alternatives here and by extension, the ecosystem of photo editing suites, not photographic philosophy/ontology; thanks).

The second is also happening: AI enables smaller teams with less competence than before to build apps, including image editing apps. In the foreseeable future (12-24 months or so) I estimate that it is within reach of an amateur user with a modest budget to have e.g. Claude build an alternative to something like Affinity that maybe not yet covers all of its features, but enough of them to constitute a viable alternative for many users. It's very hard to tell how this may/will affect the landscape. Presently, the codebase is a bit of a bastion; it's complex, takes experience to navigate & modify and thereby represents an encoding of the collective application wisdom of the developers into an implementation. If this codebase becomes more of a flexible sideshow that anyone with a Claude sub can re-generate to their own desire, instead of a crucial asset, the net result may be something like a huge proliferation of 'editing suites' that are tailored to the individual needs of the user. In other words - we may no longer be using Photoshop, Affinity etc. as such, but instead use a self-generated derivative that suits our own needs, where these needs may evolve on a daily and on a job-by-job basis. That horizon seems further away than the 1-2 years I mentioned; in that timeframe I expect the landscape to be dominated mostly by 'indie' alternatives popping up and making a mark on competition in the industry.

To illustrate the second point, please see the several apps and solutions that have been popping up for specific purposes here on Photrio already. We have people coming in who have presented their own negative scan inversion software, web-apps for film and paper calibration, someone posted an interface utility to use old ProPalette film recorders the other day...and surely, there's no end this in sight. Quite the opposite. I think it's a matter of time before we see the first real alternative to a photo editing suite emerge on this forum or Reddit made by a private individual who up to a year ago would not have been capable of producing even a simple desktop app.

I think you agreed with me. Thanks.

 

[koraks]

I think you agreed with me. Thanks.

Absolutely, I think there's validity to what you proposed and I see something like what you suggested certainly coming our way!

 

[Alan Edward Klein]

You could do that but what you would get back would no longer be a photograph. It would be an image generated by the ai as prompted by the original photograph but no part of the original image information would remain

No change in the image other than the same edits provided by let's say Lightroom. It's just that AI prompts replace the Lightroom sliders. I imagine Adobe is working on this for their own programs.