Built to Last

Tony Parisi
Published in
5 min readMay 9, 2022


Standards for the Metaverse aren’t just about interoperability; they’re about creating content that stands the test of time.

Creators like Larry Rosenthal of CubeXR have been soldiering on with the comings and goings of 3D file formats for years now. But these 20+ year old VRML models still work!

Creators, how many times has this happened to you: your content stops working because the software company that makes the tool you used to create it went out of business? Or discontinued the product line?

If that content is a bitmap or video, probably not too often, because your tools work with widely supported media formats like JPEG, PNG and MP4. But if your content is a 3D world, it’s probably happened to you more than once in your career. The history of 3D media, especially on the Internet, is a history of one unsupported file format after another, because it’s been dominated by proprietary software. The few exceptions are industry standards like VRML and X3D. While their commercial success waxed and waned over the years, buffeted about by vicious competition and beleaguered by poor market timing, nevertheless there are still several software tools and runtime libraries that support loading and at least doing a basic rendering of content in those formats.

The same can’t be said for Viewpoint 3D, Shockwave 3D, Flash 3D — the list goes on. Not to mention avatars created for any of a number of defunct virtual world platforms. All of that content is unusable: dead bits on hard drives and web servers. Think of the thousands of hours of work invested into using those tools by dedicated creators — all just wasted. Why? Because those formats were controlled by a single commercial entity and subject to the whims of their business. Standardized file formats, while having plenty of issues of their own, are far less vulnerable to such fates.

The state of the art in 3D web games, circa 2009. Where is this Shockwave 3D content now?

Even the mighty Macromedia, ultimately acquired by Adobe, wasn’t able to create a 3D format for the ages. The image above is from a promo game for the 2009 film Paul Blart, Mall Cop. The game was built on Shockwave 3D, an extension to the popular but now defunct Shockwave web player. Shockwave’s usage was eventually eclipsed by the leaner Flash, which also incorporated some basic 3D. Flash 3D was never very good; but that doesn’t matter, because Flash is also approaching its end of life. So where is a 3D content creator supposed to go?

When most people think about standards, they think about interoperability: I make something that can be experienced in multiple apps on various devices, rendered by different software packages and so on. And while interoperability is a critical aspect of standardization — enabling a level playing field for vendors and more choice for creators and consumers — an equally important characteristic of an open standard is durability. Content in standard formats has a much longer shelf life. It doesn’t have a single point of failure in terms of enabling software, and it doesn’t have arcane knowledge about its inner workings embodied in a single implementation in code.

Why does this matter? Because while artists may use a certain piece of software to land a job today, they don’t expect their creative output to be lost tomorrow. No other media works this way. Films made a hundred years ago can still be watched today — factoring in the lifespan of the storage media they reside on. Music can be recorded at a kitchen table or in a high end professional studio; either way the tunes can be enjoyed on any music player for years to come. Why should 3D be any different? Answer: it shouldn’t.

Trailer for Harold Lloyd’s SAFETY LAST! The film was released in 1923. Can you still see and hear it?

Think about what kind of Internet we would have today if people thought their websites would break after a year or two. I would go so far as to suggest that we wouldn’t have a consumer Internet. Nobody would have invested the time and energy. Sure, sites require maintenance and care, and you’ll want to continue to upgrade to take advantage of the latest stuff; but you wouldn’t expect most of your content to just… stop working overnight. And even for those ancient web pages based on really old HTML, there’s a way to experience those using the Internet Archive aka the Wayback machine. That only exists because the formats the content was based on were largely open standards — otherwise nobody would have invested the time necessarily to create software that would load and view it. It would just be lost, forever.

My very first startup’s home page from 1996… courtesy of the Wayback Machine

The same is true for the Metaverse. Creators in the new economy want their content to endure: content that can be used by everyone; content that they can license to media companies or sell directly to consumers and collectors. And these buyers won’t abide by art that they can’t readily experienced in their own worlds or in their collections. There’s a reason that glTF, the prevailing standard for 3D asset delivery, is the preferred model format on NFT marketplaces (arguably key building blocks of the proto-Metaverse): it’s the only 3D format people trust — because it’s not controlled by a single software company. In the course of the last decade, glTF went from conception to collaborative development by a handful of individuals, to early adoption by leading software companies, to where it is today: about to be ratified internationally by ISO. And all of the work to get here was done in the open by people and teams who wanted to solve a shared problem and benefit the entire industry.

glTF is an asset format, and as such it won’t address the whole range of technical requirements for the emerging Metaverse. We’re also going to need APIs for multiplayer messaging and shared state, higher-level content formats for describing interactive behaviors and avatar characteristics, and so on. We should look at glTF as an exemplar of a process and an outcome that has multiple benefits, and with none of the downsides of getting locked into a vendor-controlled format.

As the stakes get higher in the fully-featured shared worlds of the Metaverse, expectations are rising, too, and the need to build fast will tempt us to lean on proprietary tools and formats all over again. But we should not get taken in. While they may confer capability in the short term, history gives every indication that such facile solutions will very likely become dead ends down the line. We may need to make heavy upfront investments in the unglamorous work of standards-making today to ensure that the 3D creations of tomorrow endure — because in the Metaverse, the only content that will have value is content that is built to last.



Tony Parisi

Metaverse OG. Entrepreneur. Investor. Co-Creator, VRML & glTF. Head of XR Ads/E-Commerce, Unity Technologies. Pre-apocalyptic author. Music. @auradeluxe