Aug 1, 2019

The Quest for Xanadu

Understanding Organic Behavior in Virtual Worlds





“Was it Laurie Anderson who said that VR would never look real until they learned how to put some dirt in it?”
― William Gibson





The Imperfect World

There is something to be said concerning the nature of imperfection and its role in any virtual world – whether that be an elaborate immersion with an HMD or some sort of XR Augmented Reality. It’s all really the same premise with different context.

Somewhere around the turn of the century, I was introduced to this epiphany and it has since shaped my comprehensive understanding of the intricate mechanizations by which virtual worlds operate. Not as compartmentalized notions independent of one another, but as a massively interconnected ecosystem.

It’s never been so simple as “Make a sound loop here” or “Increase the graphics fidelity”. There are always underlying considerations for each and every decision made in a live virtual environment which can affect a cascading change across the entire ecosystem itself.

In a way, we could say this is similar to introducing an invasive species into an area and watching it decimate everything.

Such is the case with today’s thought experiment, and why I’ve decided to pick up the digital pen to craft yet another post after so much time.

As per my previous assertions however, I’ll try not to retread old ground with this and instead introduce something new to the conversation.





Somewhere around 1995, a small company at the time had stumbled onto the “killer app” of virtual worlds. Rejected by the original parent company, Worlds Inc, the designers raised enough money to buy the rights to the new creation and embark on their own.

Worlds Inc was under the impression that they were selling off a failed experiment, but little did they know they were giving away what would be the biggest breakthrough in virtual worlds at the time. Alphaworld, and later ActiveWorlds would become the template for future virtual worlds afterward, whether we acknowledge this or not.

In a world dominated by Web Worlds discussion and VRML and X3D pages (at least in technology circles), ActiveWorlds went in the opposite direction. It created a custom browser (not a VRML plugin) that had servers and (most importantly) would allow the multi-user virtual world to edit that virtual world in real time.

For a short time, ActiveWorlds was on top of the tech world. A virtual gold standard to aspire to. But that changed very quickly, as the company behind it weren’t actually able to understand what made it work so well that it appealed to those end-users.

In a way, they had captured lightning in a bottle and just as quickly lost it.

This isn’t a new phenomenon today, as I’ve seen the same situation play out countless times since then. Each contender trying to figure out what component or configuration created that ecosystem so they too could capture that lightning.

Many thought it was raw graphical power that was the secret, and they were very wrong. Some thought maybe the publish model of Worlds Inc was a good bet (it wasn’t). Over time, many virtual worlds came and went due to their inability to understand the organic nature of these virtual worlds and what makes them appealing in the first place.

Of course, over these twenty years, I’ve known full well what makes them popular and what was likely to crash and burn. Even as they were putting out fancy press releases and marketing, being interviewed with technology blogs and magazines… You as my reader are free to look back as far as 2005 and see for yourself that I’ve landed about 98% of my forecasts with alarming accuracy.

If I wasn’t so confident in this understanding, I would never have agreed to work alongside some of the brightest minds from the largest multinational corporations. I would never have signed on to work with IEEE Virtual Worlds Standard group, and I surely wouldn’t have agreed to co-author a seminal paper on the subject of the past, present and future of virtual worlds.

There are two possible outcomes when you’re staring down a journal like ACM and submitting a paper, or working with folks from IBM and other multinational corporations to help define an entire industry. One does not simply predict the rise of Zero Barrier Retail and how it can manifest (Amazon Go).

Either I’m completely delusional, or I actually understand something most people don’t.

Over the past twenty years, it has overwhelmingly been proven the latter under countless instances.

After all, you are perfectly aware that the day Philip Rosedale announced High Fidelity, I raised concerns and outlined that it would not work out the way he was promising. It was no surprise to me that shortly after my interview with NWN talking about how High Fidelity was not the Metaverse nor would it likely ever be just by overall design, that Philip would make a rebuttal and then shortly after shut down the public servers to focus on enterprise solutions out of the scrutiny of the public.

The numbers just weren’t there to support his promises and assertions. They weren’t there from the beginning and they weren’t there when he claimed to be making The Metaverse later.

But this isn’t a post about kicking anyone in the teeth, though arguably Philip does a good enough job doing so to himself consistently.

What this is about is something that has been persistent in the virtual worlds industry for twenty years. Even sitting in the conference room with Ebbe Altberg at Linden Lab, and trying to convey talking points on a white board, even then something important came up from him that I’ve heard before many times in the industry.

Most importantly that places like Second Life seem to have a userbase limit of around 100,000 and then some sort of glass ceiling. Of course, they’ve all tried the A|B testing and focus groups to find out what users prefer, but that doesn’t seem to be helping.

Nor would it ever likely be of much insight to begin with.

In this point alone, we’re living in the “now” when we should be figuring out the future. In the immortal words of Steve Jobs:

“Some people say, "Give the customers what they want." But that's not my approach. Our job is to figure out what they're going to want before they do. I think Henry Ford once said, "If I'd asked customers what they wanted, they would have told me, 'A faster horse!'" People don't know what they want until you show it to them. That's why I never rely on market research. Our task is to read things that are not yet on the page.”


In this statement, most companies are horrible at predicting the future. They pour over market research and analytical forecasts to see what the supposed trends are going to be.

This is how SANSAR came to be, originally betting the farm on VR Headsets being the future.

It was a bad call then, and you as the reader already are well aware, I had said so when the hype originally began. The difference between market research and forecasts, the A|B testing, and the attempt to preempt the future with some offering is that it doesn’t actually take into consideration the overall understanding of that ecosystem but instead tries to dictate it based not on anticipating what people will want but what self-interested companies and market research firms are paid to tell everyone in order to create that hype and therefore demand for their reports and services.

That’s what has always set me apart from the mainstream, and also why I have such an uncanny ability to forecast accurately.

I’m not telling anyone the real trends and outcomes because I want to be a Debbie Downer. I’m just stating the likely outcomes ahead of time and also stating why.

So what can we learn from this?

Well, we can say that the end-user doesn’t actually know what they want (mostly) but that is because the companies don’t even know what those end users want or why. So we’re asking all the wrong questions, focusing on the wrong parts of the equation.

While we do get answers for doing this, we more often than not get inconsequential answers that do little or nothing to move the needle of progress forward.

We’re no closer today in understanding that lightning in a bottle than we were in 1995, despite widespread efforts in attempting to recreate it.





Lightning in a Bottle

That being said, I’m overgeneralizing. Of course I know why that lightning in a bottle happened multiple times and then immediately was lost. However, it’s not enough that I understand the situation because what ultimately matters is if the people who are in charge of these virtual worlds systems understand it as well.

Those are the people who make these decisions, (not myself), and so those are the sort of people who plot the course for better or worse (usually worse).

As of this moment, it is safe to assume that they do not understand the organics of the ecosystem. Otherwise, why else are these systems still (for lack of better words) floundering?

I’m not implying that a place like Second Life is a failure, or even SANSAR for that matter. But they are failing. There’s a difference, and more importantly it’s more about the big picture and perception. It’s about the overarching nature of perspective. When taken alone without comparison, then yes, these are two massively successful endeavors with a rich and glorious offering.

But taken within the total ecosystem, they aren’t faring too well. Minecraft is nearly as old as Second Life and still consistently has 91 million players each month with 15% of that over the age of 30.

“Pff… Minecraft. That low resolution kids game?” I hear you say.









Yes, Minecraft. That game which allows for Path Traced Lighting, weather and more. The “more” part we’ll get to in a bit because it is vitally important in understanding why Minecraft has tens of millions of “players” over the age of 30 and Linden Lab is wrestling with a glass ceiling of 100,000 and less.

The image above is actually Minecraft, and while this is an extreme case where you’d pretty much need an nVIDIA RTX card to handle it, the point is that the graphics in Minecraft can and are evolving well beyond anything I’ve ever seen in Second Life and maybe SANSAR.


But that’s not the real point here.

What we’re getting at are a couple of points, really. SANSAR on Steam wasn’t going to help Linden Lab out. I could have told them that from day one just based on common sense. I probably did say something to the effect when the other blogs reported it.

Adding the ability to create quests in SANSAR is pretty neat, but only in its own bubble. Which is to say, the moment you put SANSAR in the same market as AAA games, you’re coming up short immediately in direct comparison. For all the advancement SANSAR is by comparison to Second Life, when compared to the typical AAA game market and player, you’re in the territory of Lowered Expectations out of the gate.


Loki Eliot in SANSAR… presumably doing Loki things.


But SANSAR looks incredible!” I already heard you say.

I never said it didn’t. There are a number of advancements besides graphics that also are impressive for a virtual world. While Minecraft  can handle path traced lighting, it’s not the potential for high end graphics that is the secret sauce. Nor was using the Crytek Engine the secret sauce for Avatar BlueMars.

See where I’m going here? SANSAR looks suspiciously to me like a new version of BlueMars. That alone is enough for anybody in the industry to throw red flags.

In a manner of “Whoa there, Nelly! Take a step back and rethink where you’re going with this. You’re flirting with a perfectly avoidable failure.”

It’s not to say SANSAR is doomed. It’s still operating, people are still working on it internally to improve it. So it really just needs a better understanding of the organic ecosystem applied for it to really flourish.

Second Life only has the glass ceiling because the people who engage with it regularly mostly “get” the point of it being a social virtual world that is user generated – a platform.

Outside that community, people more or less just don’t “get” it. How do you play this video game? What’s the point?

It’s a mismatch of target demographics, and an internal misunderstanding of what the nature of the product is which appeals to the masses. I don’t really think anyone actually knew what the magic was to begin with and were just as caught off guard by the sudden success as anyone else. But the minute they ran with it, they inevitably ran it into a wall.

One mustn’t mess with forces they do not comprehend



The Oracle Has Spoken!

Ok, not really.

The killer app of virtual worlds are the people themselves and the enabling of their ability to create and exist in a dynamic and organic ecosystem.

In a perfect world of understanding, the company (Linden Lab) offers nothing more than the tools, platform and services to enable while not interfering with that ecosystem. In a perfect virtual world, a company like Linden Lab doesn’t duplicate services or offerings that already exist via the user generated ecosystem, nor do they call it Premium. That’s not a symbiotic nature, and instead is acting the part of the invasive species.

In a perfect virtual world, a company like Linden Lab offers only added-value proposition as premium that only they can offer, and which augments the ecosystem without duplicating services.

That being said, this is far beyond any blog post or single whitepaper I could ever hope to write. There is no real way to reduce twenty years of experience and implicit understanding of such a vastly complex ecosystem into a single tweet, a one hour meeting, or just a quote or interview.

There really is no way to convey all of this in a keynote speech or presentation.

All I can say to the entirety is that the understanding is like understanding chaos theory. It’s not necessarily the knowledge that matters but the ability to see the patterns in the chaos, then articulate how best to navigate that chaos.

More or less, this is why I’ve become infrequent in my writing here. It’s why I stopped doing interviews, and it’s also why I have no desire to do any keynotes or presentations.

We’ve reached a point where the understanding and insight is beyond that which can be articulated in concise form, and also tends to retread old ground (at least to me). Overall, the industry is going to do whatever it is they intend to do regardless of forewarning or sound forecasting. Jumping on trends, subscribing to the hype, propping up the corporate talking points – and as a matter of fact, I wouldn’t expect otherwise. It has been the only constant I’ve come to expect in twenty years.

So to all those companies: You do you :)

This is now at a point where I’d just have to be hands-on in the process for anything further I could explain to make sense. It’s now a contextual situation.





Current Events

Linden Lab Lawsuit: Not really something I care about, really.

Arguably they probably could have just had her meet with the legal department to sort out the perceived issues and gotten confirmation that such concerns were taken care of, or had she raised to them issues which they were unaware of, some sort of action plan to sort it out. Of course, nobody really knows the situation entirely and it could just be bad actors. From the outside looking in, that’s about all I have to say about the situation.

It’s not for me to decide what is what. That’s what the courts are for.



Strawberry Linden: Congratulations :)

Tilia: 

Inactivity fee – This is new, but it does make sense. Much as we probably all hate it, if you have money in Tilia and you’ve abandoned the account for whatever reason, they’re going to charge the upkeep fee and take the money.

The Verification process thing. Let’s talk a minute about that.


I understand it’s a legal requirement. The process was fine for me here in the United States, but I’ve heard a few mentions that people in other countries it didn’t work for them. Hopefully it gets worked out, because a lot of content creators and stores are run by international users…

Second Life Mobile: Recently wrote up a short ten page document outlining some concepts for Active Design and arranging a mobile version of SL and UX. No idea if Ebbe actually read it or if anything outlined in the write-up would be used.

Your guess is as good as mine.