By Ian Hollidae, 2025/06/05

There are moments when I wonder why 3D on the web remains a niche. Whether youre using X3D, WebGL or three.js, one would think that a technology that can provide so much isnt more accepted. The common answer as to why 3D isn't used more is that it simply lacks a use case. But I think it's too easy to say theres no use without offering some reason as to why. So I'll share a thought I've been wrestling with recently. Just some thinking out loud really.

If you look at how we use the web today, it's basically a content delivery system. It serves up text, images, audio and video for a variety of purposes. In order to fit in, 3D content needs to work along side the established content mediums. But if you give some thought to what 3D does, you could say it's also a content delivery system, serving up rendered meshes composed of shapes and images. If the delivery purpose of 3D is the same as the web, then 3D doesn't suffer from lack of use cases, it's simply redundant. and clearly, redundancy isn't needed.

I think this is where 3D on the web breaks down. The use of 3d content highlights an odd dynamic: a technology that can deliver much more than standard web media but can't find much of a place on the web itself.

Of course, a lot of 3D developers will point out the purpose of what 3D delivers is to display immersive worlds or some sort of interactive presentation. But the question of how 3D fits in still remains fuzzy.

I don't think there's a single answer that will solve the issue. I'm sure it'll come down to designers and developers as it always does in the case of web technologies. For now, the nicheness of 3D is still a step above being unused and ignored.

By Ian Hollidae, 2025/04/28

While a most of the tech press was covering YouTube anniversary, three.js was quietly celebrating 15 years of existence.

With all of my current projects running full speed, three.js is one topic I track but don't have time for. From my point of view, three.js a deeper level 3D package than X3D and projects requires far more involvement to complete. Since I refuse to short-change my effort, I'll have to remain on the outside looking in.

By Ian Hollidae, 2025/04/08

One issue with putting 3D worlds inside a 2D environment is integrating the app into the larger user experience. Given that 3D environments are rather expansive in terms of UX, it's easy to let 3D apps run out of control. My general approach in dealing with 3D is to keep things simple.

But in keeping things simple, I ran across an unexpected problem when it comes to animation. As it turns out, animation within a 3D world doesn't look all that different from animation in a 2D world if the viewer doesn't move or rotate the 3D app. It's the equivalent of a video that has no cuts or motion being no different from a static image. In fact, using a video would be rather pointless.

So what can be done to make 3D animation stand out?

I decided to experiment with one of my older X3D demos. It's a 3D maze where, in the original version, the viewer manually navigates through corridors like any standard maze. One idea was to rework the demo into a 3D non-gaming animation apps. Basically, an app that's more than standard SVG or CSS animation but something less than a full-blown game (I'm still not entirely sure what that looks like but OK).

At first, I automated the entire walk-thru experience. Just push a button and avatar would makes its own way through. This wound up having the feel of an extended animiated GIF.

Another idea was to leave the maze as it was and add a "Where Am I" button that lifted the avatar above the maze space so the viewer could see their location. But this really wasn't any better than a 2D birds-eye view map.

After a number of revisions, I broke the fully automated tour into several sections to try and make the UX more interactive. At the end of each guided section, the user is presented with a brief note about 3D Maze and an opportunity to continue the tour similar to how a setup wizard works on the web. Simple text signs are used for notes but anything, such as images, can be inserted. I left the original maze layout intact, dead-ends and all, but lowered the walls so you can see where everything is.

To move through the demo, press START to begin. Press CONTINUE, when presented, to move forward:

I tried to keep this demo as minimalist as possible. No texturing, no sounds, no distant visuals beyond the operating space. Just take a basic 3D mechanism and demonstrate a possible method of integration. I believe that motion, if done properly, can add the type of animation that standard HTML/SVG/CSS can't provide.

I'm sure there are other ways to maximize the 3D experience on the web which I hope to post in the future. Maybe I'll have time to make a more complex example. For now, I'll continue to focus on simplicity.

By Ian Hollidae, 2024/10/23

In the early 2000's, I ran across my first example of VRML (Virtual Reality Markup Language). It was a panaramic photo someone had stitched together. The image was used as an image texture on a circular wall and the user, sitting in the middle of the circle, could swirl around 360 degrees to view it. The swirling didnt work very well but the idea of VR web comtent was fascinating. I wondered in general about what else could be achieved.

Unfortunately, I never got the time or had the need to dive in further. Eventually, VRML faded away. It didn't seem to really make that much of an impact to begin with. Flash won the plug-in wars and that was about the limit users were willing to put up with in slowing down the browsing experience.

A Second Beginning

Fast forward to 2020, and just by chance, I stumbled upon VRML's successor, X3D. It was a complete surprise the effort was still around. Out of curiosity, I began browsing through the documentation. Just as before, I began asking who would be interested in using VR on the web in the age of walled-off social media. However, as I kept reading, the one thing that stood out was that a plug-in was no longer a needed. It was all pure XML and Javascript.

My own personal interest was rekindled.

Fortunately, I was able to find the time to truly get into X3D. I spent several weeks making small demos in order to get a handle on things.

During my exploratory process, I discovered existing X3D examples and while most were good for visual demonstrations, they weren't very useful. After some time, it became clear my main goal for using X3D was for interactive webapps. Some of the possibilities of interest include:

1. 3D photo presentation (Example)
2. 3D photo galleries
3. VR product showcase picture (Example)
4. Simple games
5. Virtual space demonstrations

I'm sure there are plenty more uses than the ones I listed. It seems the only limit is your own imagination. Of course, items such as performance comes into play. The larger your application is, the more you need to download, constraints have to be applied. You get the picture.

So now that I've been using X3D for a year, I thought it would be a good post to share a brief assessment. This is in no way a comprehensive review. Just some thoughts about what has stood out the most.

X3D As A Markup Language

The XML-based script integrates well with HTML along the lines of SVG. In fact, there are some rough parallels between the two.

The grouping of objects, transforms and object reuse are a large part of both markups. The X3D 2D component, the ability to draw 2D objects in a 3D world, is basically an SVG equivalent.

The ability to create custom shapes in X3D using prototyping goes well beyond the SVG <PATH> tag. But even if you don't want to go that far, there are several path-like constructs that get the job done on the customization front.

For simple embedded displays, no javascript or complex animation, getting up and running is no more difficult than standard web development.

X3D Libraries

What you get out of X3D is largely dependent on the library you use. The two libraries I know of are X3DOM and X_ITE. Both have somewhat different approaches to providing X3D functionality.

X_ITE is a full blown player and framework. It's programmable and configurable. It also seems to be a more complete implementation of the two.

X3DOM, on the other hand, is closer to a straightforward library for web use. Fewer moving parts and not much different from other standard javascript libraries in usage.

Each library has plenty of benefits and I use them both.

X3D Tools

A generic text editor will do the job for basic and moderate tasks. For now, this is what I do. But for more advanced content, dedicated software is needed. It seems there are two choices: Sunrize and Blender.

I currently don't know much about Sunrize (I wanted to focus my learning on direct coding efforts) but Blender has long been considered a beast. And Blender isn't specifically an X3D tool. It just exports content to the X3D format.

Will tools be a problem for X3D down the line? Given the size of X3D, compared to HTML and SVG, it's possible. And the more complex the construct, the more cumbersome X3D can become without dedicated tools. Where those tools will come from is anyones guess.

The Past Is The Present?

As I said before, my main use for X3D is embedded web apps. In a lot of ways, this sorts of mirrors the old days of Flash and Java applets. However, there doesn't seem to be the huge performance issues Flash and applets suffered from. At least for small things.

One other difference from the past is that X3D implementations are all javascript based and X3D is an open standard. Anyone can provide an implementation and that means there is fair chance that X3D development won't get bogged down by corporate concerns.

Conclusions To This Point

I think X3D has a lot of potential to offer content creators and developers. Will X3D get broader adoption? Like any other technology, it's hard to say. I think it really is a matter of finding appropriate uses.

Obviously, one person may or may not make a difference but for my part, trying to output regular content is one way to help spur adoption. Besides my basic tutorial, I created a portfolio page to show what can be done. Who knows how things will unfold. Maybe a year from now, I'll write a two year assessment of X3D where everything has changed.