Mars Attacks!… Mobile platforms

A bit over a year ago, when Blue Mars entered the open beta, a lot of residents speculated if this would be, at last, the much-awaited “Second Life® Killer”. It came with some seriously interesting features: user-generated content was possible for developers (not everybody was allowed to be one, though); it had a land-renting model similar to Second Life’s; and it featured the astonishing CryEngine 3D rendering engine, allegedly way more advanced than SL’s. All that put together seemed to have enough to convince residents disgruntled with LL and SL in general to move over.

With Desmond Shang’s announcement of his investment in Blue Mars, by opening Caledonia — replicating the Caledon community in BM — there seemed to be some good reasons to switch. Caledon is one of the most active communities in SL, and by opening a path for residents to drop right into a community in BM where they would feel immediately familiar (same people, same environment, same fun 🙂 ), it would make a lot of sense for people to move over.

Then some universities and a few companies added their virtual presences in Blue Mars. It seemed to have a great head start!

Finally… due to the incredibly high demands on the required hardware (BM would only run on Windows-based “gamer” hardware; that was alright since the majority of the earlier adopters came from that market segment anyway, and these were the largest group really interested in using BM’s platform), they announced a streaming solution. Now users would be able to use low-end hardware and still enjoy Blue Mars to the fullest; and, well, they hired Hamlet Au to do some promotional writing for them. It sounded everything was just peachy 🙂

Nevertheless, the number of registered users on their forums was little over 3200. It’s really not much, compared to, say, OSGrid, which has about 2000, and from a purely technological point of view, OpenSim doesn’t look as good or as interesting as Blue Mars. Estimating the overall number of users based on the forums is never a reliable metric, but it still gives a very rough idea on the amount of actively engaged users on a platform.

Linden Lab apparently still thought that Blue Mars was their closest competitor: first, they made sure that meshes would be released as soon as possible, and launched a preview quite quickly; many of us had given up on their endless promises. Meshes are not yet on the main grid, but the technology is promising; it’s clear that they could not postpone it for much longer. Not when one of the biggest argument for organisations to leave SL and join BM (or any other virtual world) was the lack of mesh support, thus disallowing designers to reuse their content on SL as well.

And shortly after Avatar Reality announced their streaming solution, Linden Lab experimented with their own; it’s still very crude, but at least they showed the world that they were slowly catching up. For high-end graphics cards, the SL viewer started to feature a lot of intriguing new optical effects like the ability to focus on an object and have the rest blurred. This was LL’s response to the criticism that their rendering engine is “outdated”; they’re showing that it’s just the 1.X series that hit a road block with their monolithic design, but SL 2.X allows LL developers to push the limits again and again. Even though all the new features (well, except for meshes…) will only be available for high-end computers; but that’s were BM’s market is.

And then I was reading Inara Pey’s blog and saw her post on Avatar Reality’s sudden change of strategy. All of a sudden, the company was downsized (even their CEO, Jim Sink, was sacked), further development on the PC desktop version was stopped, and they announced that they would only develop for the mobile market instead. Hamlet Au and Iris Ophelia stopped being their “consultants”.

Why the sudden change in strategy?

Well, let’s see the good side of it. There are few — if any — 3D social virtual worlds for mobile platforms. Apple-haters — e.g. 88% of everybody who owns a computer — refuse to acknowledge that Apple’s really poised to become the world’s first trillion-dollar company, perhaps as soon as 2012, according to the most optimistic analysts. It matter little if that will be really the case or not; what counts is that the mobile market will continue to grow, thanks mostly due to incredible hype — but hype that translates into sales, and most of those sales come from each innovative design that Apple releases (for the rest of the world to hopelessly copy) — and these will be mostly mobile devices in the future. Developing for Apple’s iOS is a good strategic move, specially because it’s highly likely that Apple might simply drop the “old-fashioned”, traditional way of running desktop applications on Mac OS X, and simply adopt iOS for everything — yes, even laptops and desktops. The reasoning behind it is obvious: it’s far, far easier to use.

So Avatar Reality’s funders bought this idea. Develop the first 3D social virtual world for the mobile platform that has a bright future ahead; corner the market (IMVU’s own offering for the iPhone is not 3D); wait for Apple to start launching iOS-based laptops and desktops; get rid of the future competition by having a very strong market share and become the reference product on iOS. The irony is that 18 months ago they despised Apple users by adopting the Windows-only CryEngine and having no plans to port it to anything else besides a PC (or eventually a Xbox). Their funders, however, seemed to have talked sense into them!

It remains to be seen if a dumbed-down version of Blue Mars running on an iPhone or iPod (or even iPad!) will be able to gather more revenue than the current business model for BM on Windows high-end desktops. Still, they will be able to ride on Apple’s hype, and that will make funders happy for a few years.

Because it’s clear by now that it is impossible to replicate Linden Lab’s business model. You can throw money and technology at an idea, but it’s impossible to make money out of it; what that means is that sooner or later, funding will be exhausted without a valid business model. This is something that nobody is truly grasping. For over six years I’ve read people wishing for “competitors” to appear out of thin air and force Linden Lab to deal with them — and by that, they mean that LL would have to improve SL in tons of different ways until they become as good as the “competition”.

But that’s just wishful thinking. Just waving economy books in the air and yelling that the free market allows competition to emerge doesn’t actually make them appear 🙂 Tateru Nino shows how investors actually think about this niche market. It’s not that the market for virtual worlds is small; according to Kzero, this market has almost twice the size of registered users than Facebook. Yes, of course many people register several accounts across virtual worlds and are counted repeatedly; yes, of course registered users are not the same thing as active ones (Facebook doesn’t have 600 million active users, either, and it all depends on what you mean by “active” anyway — logging in once per month counts as “active”?). Still, the virtual world market is huge, not a statistical anomaly.

What is small is the market for social virtual worlds with user-created content. When “it’s not a game” but simply a place where users produce their own entertainment and create their own content for others to use — for a tiny fee — then the market shrinks. Perhaps to a million active users. Perhaps there simply are not more than that. And all are Second Life residents.

This is what people are really forgetting: there is simply not a bigger market than that. Oh, sure, by making SL “fast, easy and fun”, or even by pushing games designed in SL ahead (like Rod Humble seems to be encouraging), might give them a few more residents. Say, perhaps even 2 million. But it won’t go beyond that, until something radical changes — and no, it won’t be “technology”, although technology could make a difference.

This market is indeed an oddity. It seems that if you don’t set any rules at all, people are completely lost, since most of us are conditioned since our tender ages to be passive consumers of entertainment, not entertainment providers — not even self-entertainers. When I was a kid, I was encouraged by my parents to create my own toys, building them from scratch — with Lego pieces, stitching bits of cloth together to dress up the plushies, or just assembling cardboard boxes into “cities”. The current generation is just taught to turn on the TV, or, more recently, to turn on the desktop computer at home — or the iPod. This has a serious influence in the way we think later on. If all you got from your parents was how to quickly turn on passive entertainment, that’s all you will be able to experience in your adult life. We few SL residents are today a statistical anomaly — either we were still lucky to have been taught how to design our own toys and entertain ourselves on our own, or we discovered that by chance, even though we were just conditioned to get cheap entertainment from the TV. The Lindens never realised how unique their resident population actually is: those “billion” users of other virtual worlds are passive entertainers, they simply don’t see a point in anything like SL.

Avatar Reality, like a few others (like, well, Metaplace a year ago), were founded by people with similar profiles to Philip and Cory. They liked to have fun building things, and thought it was a great idea to give users a platform for them to have fun building together. And there are certainly a lot of people thinking that way — just not hundreds of millions, but just a couple of millions at most, world-wide. The perversity of the system is that, even though most people in Second Life are not content creators, they wish they could be. And in SL, they can be, without limits of any sort, except for taking some time learning how to glue prims together and stick some textures to it. BM was targeted for professional designers who would sell content to consumers. The problem is that although this is what makes the economy work in SL — only a few are professional content creators, most of us are basically content consumers — it’s not what the residents perceive it to be. We have all the illusion we could become extraordinary content creators on out own if we only could get enough time to do so. Or, if we’re not interested in the economy at all, we’re fine in having fun with plywood boxes, even though we won’t be able to sell them. It’s the equivalent of assembling old cardboxes together and call it “a city” when we were kids — we knew very well that it didn’t look remotely like a real city (or even a city we could buy from Mattel or similar toy manufacturer), but we nevertheless had a lot of fun doing it on our own. That’s what SL offers us all (even if we don’t use it that way), and that’s what makes it unique.

One could replicate this model easily enough using other platforms and technologies. However, LL was very lucky. By chance, they created a currency after the residents requested it, before the virtual world officially launched. And also by chance, in 2004 a crazy resident thought it would be a nice idea to offer a currency exchange between US$ and L$. These two “chances” — not planned from the beginning — were what made SL’s economy worthwhile, and you cannot “replicate” coincidences and expect the same results. A third coincidence was the huge exponential growth in 2006/7 — because the demand for content was so high that pretty much everybody could glue two plywood cubes together, label it “a desk”, and sell it for a few L$. This encouraged a very dynamic in-world marketplace; residents were so desperate for content that they would pretty much accept anything. As the growth went from exponential to linear and now to pretty much zero, hobbyists simply failed to sell as much as before, and had to close their shops and sell their sims, leaving the content creation for high-quality professionals.

Now here comes another twist: those residents were very frustrated, because the time to make easy money with little effort thanks to the huge influx of utterly clueless newbies was over (and obviously they blamed LL for it). But the point is that there is nowhere else to go. There is no market outside SL for low-quality content — because there is practically no other virtual world with user-created content. The few exceptions (IMVU, Blue Mars) demanded special requirements to become an “official designer”, and that meant excluding the hobbyist, who didn’t know how to use high-end professional 3D modelling tools. So they had no choice but to… remain in SL. Even without selling content, even though they complained all the time, they still remained around. In fact, what’s the point of changing to another virtual world — one that promotes high-end designers and couldn’t care less about hobbyists — if there is no way they could have fun building their cardbox cities there?

The ugliness of the mainland is a tribute to SL’s unique model. It means that each and every one of us in SL can have fun by themselves. We don’t need to be professionals, we can still enjoy ourselves on our own building and creating things. But there are not many of us; we’re outsiders, mavericks, in this society of cheap passive entertainment. That’s what Avatar Reality (and Metaplace and so many others) failed to grasp; and that’s why ultimately all future attempts at doing a “better than SL” virtual world will ultimately fail.

Oh, I can imagine that here and there some crazy but talented techies will be able to squeeze a few millions out of some business angel or VC company. Someone might persuade one of those companies to give them a few million dollars to do a SL clone for Facebook or even the iPhone. And for a few years, while the funding lasts, they will make huge media splashes, get incredibly hyped, and poison the SLogosphere with announcements of the imminent end of Second Life. We’ve seen this happening since the end of 2004 (at least!). Just look back at all those news of companies that would “take over” and ultimately bring doom to SL.

Each one of them utterly failed. (Some might be still making money, but that’s because they were clever enough not to develop their product as a competitor to SL.)

Hopefully this might be a lesson for business angels and VC companies: if you get a bunch of hackers claiming they have a wonderful idea to create a “SL Killer” and take over the market of social virtual worlds with user-generated content, chase them out of the office. Or, if you happen to like them, put them developing apps for Facebook, iOS, or Android. That’s the only way to make sure the money isn’t thrown out of the window 🙂

I believe that we’ll still hear about Avatar Reality and their iOS platform for a long time, because they hype surrounding iOS is insanely huge, and they might be able to secure some funding for a few years, and perhaps even get lucky and find a sustainable business model for that. If they’re clever, they would think about Android and Facebook too (although I predict that both are condemned to disappear well before Apple goes bankrupt; give it a decade or two; still, a decade is enough for VC to get their money back). If not, someone will. Maybe even Linden Lab, who knows.

As for the push to “more games in Second Life”, well… Avatar Reality had a far better game-enabling platform, and that was not enough. Multiverse is still around with their platform, but it’s not a connected virtual world — just a platform that allows games and virtual worlds to be designed. SL used to be a “game platform” very early in their beginning, but Philip had long since dropped the idea when it was clear that:

  1. The “platform” would never be fast and robust enough to compete with console First-Person Shooters;
  2. Nobody (that is, no game company) will wish to develop a huge game with 30-60+ hours of engaging content for SL’s tiny market.

While role-playing sims are the rage in SL, they’re independently done, mostly by volunteers, who simply have lots of fun doing the games. If they didn’t — if they were all about the money — they would be doing 3D RPGs for Facebook instead. But no, they’re in SL because it’s fun for them.

Still, I don’t dislike the strategy to “focus on games”. Rod Humble very likely has a pretty good idea of what makes a good game interface — and perhaps SL Viewer 3.0 will be acceptable for all. We will all benefit from that. Good social games also require good avatar-to-avatar communication — and FJ Linden is already rolling out some incredible enhancements. Games also need low latency and better performance; and if Rod gets the developers working on that, too, we will definitely appreciate the changes (have you noticed how smooth teleports and inter-sim travel now works? If not, to truly appreciate the difference, walk into any OpenSim grid and notice the difference…). And finally, designing and programming games in SL requires a more powerful programming language, and one that has less arbitrary delays and bottlenecks. Fixing that will also fix most script-related lag (and issues). And, who knows, Rod might get the developers to finish the few interface-based game-y tools (when was the last time you were in a zone where “harm” was turned on? Do you still own a 2003 revolver that shoots bullets that reduce an avatar’s health to zero and forces a teleport back home? All this was part of the interface and was never completed). And, who knows, this might even get LL to start their work to support a thousand avatars in a single region (soon available on an OpenSim grid near you). So… whatever enables game designers — even hobbyist game designers! — to develop more interesting and compelling games in SL will benefit us all. The “push to games” will very likely fail, but I really hope that at least LL makes a serious effort to improve their product to enable games actually to be playable — because by doing so, it will dramatically enhance SL for us all, not just for the gamers and role-players among us.

As always, there remains just one valid competitor to Second Life: Second Life itself, or, rather, its open source reverse-engineered version, OpenSimulator. While it trails behind SL’s own simulator software, the beauty is that almost all viewer improvements immediately enhance OpenSim automatically (some require minor tweaks to work; but it took only 24 hours for OpenSim to support SL-compatible meshes, too). And although I’m not too hopeful to see an OpenSim grid with the same performance as SL in the next 5 years, it’s still true that for some residents performance is not everything, and trading off performance for cost of ownership is definitely a choice for hundreds of thousands of users. So if LL is “fearing” any hypothetical competitor, it’s the competitor that actually has the same technology, the same business model, the same user base, and addresses exactly the same niche market with precisely the same tools as them. And they grow way faster than LL right now — which is just another statistical anomaly, since it’s easy to grow by leaps and bounds when there are comparatively few users. Things will become interesting if Intel/IBM really start offering alternate grids to SL for residential users and hit the “million active users” threshold. If that will happen in the next 5 years or not is anyone’s guess.

About Gwyneth Llewelyn

I'm just a virtual girl in a virtual world...