Rationalizing and Simplifying the UI/UX of PL in 2025 - Part B: A proposal for non-modal approach

[Vivi_Ram]

https://youtu.be/26GuxlZWmsc - Video explanation here...

Here is large sized animated gif example how it works in PotoLine. Nice simple. Blend modes and Opacity values are visible all together and available by single click. And anyone can understand this simple logic without any manual:

No. That is not the same. That's not what I am looking for. Perhaps my video was a bit misleading. I was just trying to illustrate the point with the same images. But it need not be the same images, either. You can have multiple objects of different kinds. Combinations of rasters and vectors. And you should be able to do it all quickly, without clicking on multiple menus and dialogue boxes. As long as I have a set of adjustment windows open, I should be merely able to click on objects, make adjustments and adjustments are applied as layers to that object. This is still the classic workflow - vastly improved - yet completely linear and traditional. It is a lot faster than PS or Affinity, but there's room for improvement here.

I thought about your question as to how this will work with complex compositions. I will post the workflow explanation soon. Need some time. Incidentally, I think the answer is within the adjustment list window.

This also does not address my broader point about non-modal windows in general. Not just for adjustments and filters. Other tools such as layer transform, change channel, etc.

[shijan]

I didn't say that it is the same. I just illustrate existing quick normal workflow logic in PhotoLine that don't require 5+ pages of explanation.
Vivi_Ram, really you proved some nice ideas, but this Do you really think that someone in clear mind will spend months (or years?) to re-code everything from scratch just to mimic this Xara image editing logic designed for basic consumer level editing and that don't exist in any other app except Xara?

[Vivi_Ram]

I'm not asking that we recode PL to mimic Xara. No. Not at all. Xara does not have photo editing capabilities beyond the basics. The reason I made that video was because there was literally no other program that has non-modal logic for adjustments. And I don't have animated mock-up knowledge to show you how it would work. So the video does not fully show what I am getting at. Forget about Xara completely.

The five page explanation contained more than just adjustments. Adjustments were a part of it. But so are other functions. I also made serious points about workflows that are heavily reliant on layer list panels.

There are many best practices that were destroyed too soon from becoming mainstream because of Adobe's dominance. For example, Macromedia Freehand was a far superior vector program. And that's why Adobe killed it. It was a genuine competitor. The same fate would have fallen Xara too. Xara's marketing rights were purchased by Corel and Corel nearly destroyed Xara too. For years, you couldn't buy it. And then, eventually, Xara got the rights back to market itself. It still hasn't recovered from that. These two programs: Xara and Freehand got so many things right. So many. There's a reason why Adobe adamantly refuses to release Freehand code into the market despite so many requests. Even legal challenges.

Setting colours for objects. The pen tool. (Frankly, the pen tool, which is the mother of all vectors, is quite primitive in PL.) Contouring and bevelling. Many, many things.

So, the more serious point is how Adobe's domination of market has set the course for the rest of the design world. Affinity broke a few barriers. PL has broken many, many barriers. Until recently, it was told that rasters will pixelate when you zoom in, unlike vectors. But three developers are challenging that. They're writing a program from the ground up to challenge it. Graphite.rs. Free and Open Source. They're not there yet. But I'm rooting for them. VectorStyler is another program I love. It is one man's fight against the norm. It is an extremely buggy program at the moment, but I still bought it to support him! And he fixes bugs like a monster. Every day. He releases betas every week!

The reason I hang around so much in PL forum is because I like Hubers' approach to things. Their willingness to break rules and challenge norms is why I'm trying to push the boundaries.

This is the most important question I raised in the proposal: "Is layer-panel dominant workflow the only possibility after 36 years?" Since 1989, the basic workflow has remained exactly the same in all the programs. Until some one says, there's got to be a simpler way.

As always, ADODE DELENDA EST. Adobe must be destroyed. They're truly the worst thing to happen to the creative industry.

[Vivi_Ram]

https://explorimentdesign.wordpress.com ... -freehand/ - Read this blog if you get a chance. It catalogues all the things Freehand got so right, and still hasn't been beaten by Illustrator. Or any other program. Frankly, f*** Adobe for killing Freehand. Bastards!

[shijan]

The reason I made that video was because there was literally no other program that has non-modal logic for adjustments.

This is the most important question I raised in the proposal: "Is layer-panel dominant workflow the only possibility after 36 years?" Since 1989, the basic workflow has remained exactly the same in all the programs. Until some one says, there's got to be a simpler way.

Don't you think that if no one uses it for so long time probably it is not practical and dead end workflow for complex layer based apps? It fits well only to some smaller and simpler systems like Layer Styles.

[Vivi_Ram]

The reason I made that video was because there was literally no other program that has non-modal logic for adjustments.

This is the most important question I raised in the proposal: "Is layer-panel dominant workflow the only possibility after 36 years?" Since 1989, the basic workflow has remained exactly the same in all the programs. Until some one says, there's got to be a simpler way.

Don't you think that if no one uses it for so long time probably it is not practical dead end workflow for layer based apps? It fits well only to some smaller and simpler systems like Layer Styles.

Let's think about it harder. I'm not one to settle with easy answers. You're probably right. I'm probably wrong. But it's better to get things so wrong that just in the process of thinking things through, we get some things right!

[Herbert123]

This is an interesting discussion.

This is the most important question I raised in the proposal: "Is layer-panel dominant workflow the only possibility after 36 years?" Since 1989, the basic workflow has remained exactly the same in all the programs. Until some one says, there's got to be a simpler way.

That is actually not quite true: a different methodology to do image compositing has been long in use: node-based editors. A node-based compositing workflow scales much better to large complex image compositing projects than a layer-based approach. Both have their merits, and it depends on the context which one may be more effective and manageable. Node based image compositing is truly non-destructive. And node trees are easily recycled in other projects. The bulk of professional visual effects and colour grading compositing is accomplished via a nodes based workflow nowadays. After Effects is the odd duckling and favours a layer-based workflow instead.

Here is an example of blending two images in Blender's image compositor using nodes:

[Herbert123]

@Vivi_Ram I browsed your document, and here are a few comments.

First, I would like to see PhotoLine have some sort of centralised properties panel. I completely agree that the layer effects, Color Filter, and quite a few filter/adjustment layer properties could just as well be handled in a non-modal properties panel.

Take Cinema4D for example: any object that is selected has its properties opened in a central properties panel. What is important to note: it is possible to open multiple properties panel and "lock" these. Locking ensures that the properties of a particular object remain in view. In my opinion, this is an essential feature to include.

Having said that, I actually quite like the preview window that pops up for adjustment layers. That preview window would NOT fit comfortably into a general properties panel. Which means the preview option must become a switch that changes the main window to show this preview view instead (with the preview slider). So instead of having a separate preview window the main viewport can then be toggled to accommodate the preview window.

Or an option to open the preview window separately via the properties panel could be an alternative solution here.

But yes: it would be nice to consolidate most of these isolated properties that are currently in modal windows into a centralised properties panel that automatically adjusts according to what the user selects.

This would only be beneficial, and I see no real downside, as it does feel a bit fragmented in places due to those modal property windows.

[Vivi_Ram]

This is an interesting discussion.

This is the most important question I raised in the proposal: "Is layer-panel dominant workflow the only possibility after 36 years?" Since 1989, the basic workflow has remained exactly the same in all the programs. Until some one says, there's got to be a simpler way.

That is actually not quite true: a different methodology to do image compositing has been long in use: node-based editors. A node-based compositing workflow scales much better to large complex image compositing projects than a layer-based approach. Both have their merits, and it depends on the context which one may be more effective and manageable. Node based image compositing is truly non-destructive. And node trees are easily recycled in other projects. The bulk of professional visual effects and colour grading compositing is accomplished via a nodes based workflow nowadays. After Effects is the odd duckling and favours a layer-based workflow instead.

Here is an example of blending two images in Blender's image compositor using nodes:

Yes, you're right. I am aware of node based editing. Even graphite.rs is going fully towards node based editing. I suppose what I meant was within layer based software.

I also have some thoughts on the preview window. Will share later.

[Herbert123]

. Even graphite.rs is going fully towards node based editing. I suppose what I meant was within layer based software.

I tested Graphite a few weeks ago. What I like is that it seamlessly builds a node tree alongside the traditional layer approach.

But it has a looooong way to go yet. Performance is abysmal for bitmap editing. Basic stuff such as colour management isn't in the cards for a while. And the developers mentioned that they focused on vector editing first, with bitmap editing still behind. And (in my opinion) it is held back by the tech stack the developers use: web-based.

Again, I like the concept behind Graphite. But compared to PhotoLine, PS, etc. it has a long road ahead of it.

[Vivi_Ram]

Yes, Graphite.rs is really far off from the frankly daunting goals the developers have set themselves. But it would be interesting to see how it performs, as it is completely written in Rust. They're also working on a new rendering engine. Very promising. Very daunting. I hope it succeeds, and we get to thumb our noses at the bastar... I mean the gentlefolks in Adobe.

[shijan]

Just got an idea for Color Filter. What if make Color Filter UI as Adjustment Layer? To make it work same as original Color Filter, it simply could be dragged inside any layer. As a result it will be easier and faster to operate directly in Adjustments panel.

Screen Shot 2025-04-20 at 6.43.39 AM.jpg

[Herbert123]

Just got an idea for Color Filter. What if make Color Filter UI as Adjustment Layer? To make it work same as original Color Filter, it simply could be dragged inside any layer. As a result it will be easier and faster to operate directly in Adjustments panel.
Screen Shot 2025-04-20 at 6.43.39 AM.jpg

This is actually quite a brilliant idea: decoupling the advanced blending from the layer itself will also allow us to use other adjustments (such as curves) and layer masks to control the final effect better.

And it becomes possible to affect multiple layers grouped together as well.

I like this. It would open up a whole new world of possibilities.

*edit* for compatibility's sake I'd say to leave in the old option and add a new adjustment layer version.

[shijan]

*edit* for compatibility's sake I'd say to leave in the old option and add a new adjustment layer version.

I agree.

To emulate Color FIlter in "Background" mode you need to group all bottom layers and create virtual copy of that group, which makes layers structure too complicated.
So i guess instead of real Color Filter adjustment it could be some sort of empty "Background Reader" adjustment (or special layer type) which will read all bottom layers as single merged layer. Then you just apply any other adjustments to that "Background Reader" and it behave like Color Filter.

Same time it is rather simple to emulate Color FIlter in "Layer" mode by creating virtual copy of a layer, desaturate it and apply in Clip mode:

[Vivi_Ram]

This is indeed a great idea! +1 - it has my full vote!