2 Blender Features that may be the End of (some) add-ons?!

When NOT to script something

To fellow add-on developers: maybe, just maybe … your add-on idea doesn't need to be an add-on.

Now, hear me out. I am not saying you abandon it. I'm talking about expanding your horizons. Let's get into these 2 features that may end (some) add-ons.

The first is already available:

Geometry Node Tools

Instead of using geometry nodes as a modifier, they can now be applied immediately. As a node tool, they are the same as a Blender operation. Complete with a redo panel. It can even take things like the mouse cursor position into account.

Many add-ons I know use geometry nodes, or modify or generate meshes on their own. Some generate or deform meshes based on parameters. These add-ons are often just wrappers for geometry nodes. Why wrap around geometry nodes, when you can just share the nodes themselves?

Not only are node tools more intuitive; they are easier to design and hardware accelerated. Using compiled geometry nodes can be far faster than Python scripts.

Did I mention these can show up in menus too? And if they are an asset cataloged under an existing menu name, it will be added to that menu.

Also, the current understanding is that, while Blender add-ons must be GPL compatible, node assets are not (this could change, I am not a lawyer and do not consider this legal advice). This means you could have more licensing freedom (and less hassle) with node-based assets than an add-on.

And Geometry nodes can now have gizmos, so you don't have to tweak numbers the entire time? So intuitive.

And finally, geometry nodes and node tools can be marked as assets. So you can literally drag and drop them into your file.

Asset Browser

Speaking of assets, there are a lot of asset manager add-ons out there: Blenderkit, Poliigon, IKEA, Poly Haven, engon, TrueAssets, Urbaniac, CG Trader Asset Browser.

But these are add-ons. Surely Blender isn't planning to allow access to asset libraries remotely, and downloaded on-demand: oh wait, they are?

"To integrate the asset system with the internet, a new type of Asset Library will be supported: Remote Asset Library. This will work similarly to the existing (Local) Asset Library system ... The library content can be downloaded on-demand and cached locally for reusability.... Extensions and asset libraries will be configured independently. But they ... will also be available from the Blender Extensions Platform."

As Blender contributors add more features to the asset browser, asset add-ons have less need to add new features.

Obviously this may take a while to arrive. This is not the first time Blender devs have announced something that hasn’t happened yet ... layered textures, I’m looking at you...

But Blender accessing online assets will be a game changer (and I don't say that lightly). Asset manager add-ons can be broken down to just the assets they are sharing. And what better way of accessing all your assets than the asset browser?

Imagine searching your Poliigon, Blenderkit, and Poly Haven assets in one place: the asset browser. Most of my time spent finding assets is having to check every website, every asset add-on, every bookmark I’ve ever saved, remembering where to check for what kind of assets.

Having it all accessible in one place within Blender is absolutely brilliant.

Why I'm not worried

So yes, these features are nice and all. But why will these replace Blender add-ons?

Because the hardest part of developing solutions is how to implement it for the user. Blender offers many interfaces: assets, node tools, add-ons with panels, menus, pie menus, workspace tools, or even custom drawn interfaces if you’re feeling fancy.

And yet, many just end up as add-on panels pinned to the 3D view shelf. Some are even just add-ons wrapped around a .blend file. Why?

I know the answer from experience. As an add-on developer, I get so excited with being able to solve a problem, I may forget to consider how to solve that problem.

Certainly some add-ons do deserve to be in a certain right-hand shelf. But since we see it so often, it is usually the first solution that comes to mind for users and developers.

Light Painter had this problem. It was a side shelf add-on initially.

Blender screenshot: two panels on the shelf, with buttons for drawing and adding lights

But then I realized it'd be more intuitive as a pie menu.

Blender screenshot: a pie menu of different buttons

One or two promo videos later, I realized Light Painter should be workspace tools instead. And that's where they are now. And it feels far more intuitive there now than ever before.

Blender screenshot: a list of tools for the workspace

So, my plea to add-on developers: take time to learn about these interfaces. Remote asset libraries may take a while, but node tools are already here. Consider if your add-ons can utilize them. Maybe your add-on can just be an library of node tools instead.

Regardless of the end result, be willing to pivot. Be willing to adapt. You are not selling an add-on; you are selling a Blender solution.

A solution can be implemented in many different ways. You need to determine which solution feels not only intuitive for the user but also for Blender.

The most intuitive add-ons feel so natural, you forget they are not part of Blender. And as geometry nodes and assets become more popular, developers and product owners need to shift to that new norm.

Let me know your thoughts on these features! And as always, thank you for reading.