What makes us return to Second Life?

Hey, it’s not Microsoft Train Simulator, but it’s still fun!

You know how it is. You are in the middle of some activity — building something creative, enjoying a chat with friends, attending an event — and all of a sudden you get an IM from someone you haven’t seen in a long while. Probably someone who has a profile from 5 or 6 years ago. Someone with whom you enjoyed several weeks or months of in-world chats and common enjoyments (maybe even a virtual lover!). Then, after a while, they logged in less and less, and suddenly disappeared without a trace. After years they come back, see that you’re still on their friends list, and decide to drop you a message. In some cases, they go away again — for months or years at a stretch — and come back much, much later. On other cases, they promise to become “more active again” but, after a few days, you never hear of them again.

While you pick up the conversation interrupted for (possibly) years, the comments are usually of the kind: “oh, it still looks pretty much the same, but this viewer is so awkward” or “I see that Linden Lab hasn’t fixed the lag issue” or “what, we’ve still got the 100-avatar-limit?”. Sometimes you might get some feelings of surprise: “oh wow, meshed content looks so awesome!” And after a few days, you might get some final messages of the type “nobody I know except you seems to be around” or “all the places I’ve used to go to have disappeared”. In my case, I tend to point out that things like the Governor Mansion at Clementina, the Ivory Tower Library of Primitives or even the International Spaceflight Museum are still around, mostly because I don’t remember anything else (YadNi’s Junkyard is allegedly still around too).

On the other hand of the spectrum you have the 10,000 newbies that register themselves every day. Of the very few that actually log in, their in-world experience is usually measured in nanoseconds, or a few minutes at best. Only an exceptionally tiny amount of newly registered users actually remain for a few days. Even less than that stays for weeks or perhaps months. Like many other similar online things, getting bored after 2-3 years is common.

Or perhaps not. Considering that so few new users remain active for so long, one would expect that Second Life had long ago emptied. But this clearly isn’t the case. The population is declining, yes, but not so dramatically: it’s just that not all users who leave SL after years of being around are replaced by new residents with the same long-term stance.

Why is that so? What makes a “core group” of residents remain around, year after year, while the majority of new residents hardly complete the whole registration procedure? This, in fact, is Linden Lab’s most challenging question. And while they have been trying to answer it since at least the SL Viewer2.0 has been developed, there doesn’t seem to be a ‘right’ answer.

Hamlet Au recently spotted that LL’s CEO, Rod Humble, has been asking variants of this question on the ‘veteran’ forums SL Universe. Now some have criticised that this is the ‘wrong’ place to ask, since the issue is about newbies, not veterans — and veterans have quite different ways of looking at SL. Many (Hamlet included) think that ‘asking the FIC‘ will give LL a distorted view on what the majority of residents want, and that LL should simply listen to the majority instead of a strange ‘elite’ with their twisted views.

Well, I thought a bit about that. And then I saw many of the comments on Hamlet’s article, as well as some discussions popping up here and there. I cannot really make a summary of all that’s ‘wrong’ and the ideas that residents are constantly giving to LL, but I think that there is a very interesting pattern that is forming, which sort of reinforces my hypothesis — one that is little shared by others! — and, strangely enough, maybe Rod Humble is aiming for the same thing.

Five years ago, I tried to think a bit about why we are so special. I mean, it’s a good question to ask! If you look at any blog, e-zine, forum, and so forth, you will see a large number of outcries against Linden Lab, its policies, how they are constantly breaking things for their loyal customers, and how the technology never catches up with MMOGs and FPSs. This has gone on endlessly for years and years. Nevertheless, with some hiccups now and then, Second Life still thrives. There might be a small recent decline, but it’s not so huge and could be accounted for many reasons, most of which external to Second Life.

However, people are constantly bringing up the same reasons for SL’s lack of growth. Linden Lab even ‘believed’ some of them. Strangely enough, they don’t seem to correlate with the real reasons. Let’s look at some of them (I’m not going to be exhaustive).

The technology is too hard to master and has a high learning curve

Compare Second Life’s user interface to Cloud Party. It’s a nightmare of menus and options and more menus and more options, and each third-party viewer does them differently, in the hope of making it ‘simpler’. But somehow it never gets as simple as Cloud Party. Or Blue Mars before them. But is the absence of simplicity really the reason for lack of growth?

A lot of people really believe that. They claim that if the viewer was as easy to use as, say, Facebook (which is complex enough!), then millions would flock to it. So Linden Lab changed the viewer. They even launched the “Basic Mode”, which has a minimalistic interface. Did that make a difference?


But the interesting point is that it didn’t make a difference to the competition, either. Virtual worlds like Kaneva, Lively, Blue Mars, and recently Cloud Party, have had way simpler interfaces. Well, Google’s Lively was gone just 6 months after it had launched. Blue Mars hung on for a while longer. Kaneva apparently only resists because its owner stubbornly refuses to shut it down and pays all expenses out of his own pocket, even though Kaneva is not a commercial success. And Cloud Party, well… it’s too early to say. Sure, it attracted a few thousands of early adopters, who will pretty much go to anything new. But will Cloud Party attract interface-impaired newbies? We don’t know, but the point is, people are not flocking to VWs with ‘easier’ interfaces. They keep returning to SL, in spite of its terrible interface.

On the other hand, I have seen many so-called ‘interface-impaired’ residents in SL. Some, for example, are elderly people who are still not very familiar with computers in general. All had a huge learning curve to deal with, when they logged in to SL for the first time. All who I know have held on. Some took months or even years until they fully mastered the interface, but, once done so, they became ‘normal’ residents like everybody else — many even became mentors (more on that later). Of course this is just anedoctal evidence: I never did a survey to know how many people gave up on SL because it was ‘too hard’.

What I can see is something different happening. Once the ‘spark’ of SL catches someone, no matter how hard the interface, they will remain around. If that ‘spark’ never arises, then the interface will make no difference whatsoever.

The super-hard-to-master interface even has a strange advantage. Once someone gets furious with SL and moves to someplace else, they will experience a VW with a much simpler interface. And then they get this odd feeling that there is little they can do in this VW, compared with SL, which has so many options — and they would be right. Blue Mars, for example, didn’t need a complex building interface, because building was not an option for ‘regular’ users. That meant that someone used to build things in SL and moving to Blue Mars would get disappointed: it meant registration as a developer, buy and learn a lot of new tools, and get used to a complex pipeline of content production (and approval) until they could place a plywood cube on their islands. That simply didn’t work.

I’m reminded of a phrase I heard once, a few years ago, when discussing this with some friends. They said that “a complex environment requires a complex interface”. Now, I know that Apple, for example, thinks otherwise: you can get a complex environment with a relatively simple interface. But how many people have utterly refused to move to a Mac because they felt that they had far less options than on Windows, and so it must be a simpler product, and not as flexible? (in fact, Mac OS X is built on top of Unix, and if you want to drool with its complexity, you only need to open up a Terminal window with a bash shell… but that’s another story!).

Photoshop is a very complex tool. But it’s because it allows graphic artists to do a lot of things. If you just need basic photo retouching, Photoshop is overkill. But a graphic artists used to Photoshop will just laugh at an application with a ‘minimalist’ interface — because it will mean it won’t do anything they need. So that’s the issue with SL residents: with enough experience, they become power users. And power users need power tools! Everything else seems to be hopelessly ‘simple’ and ‘primitive’ when looked at from the perspective of a SL resident.

It’s not just accessing the virtual world that is complex. The virtual world which we also know as “Second Life” (a problem with the use of “Second Life” is that it designates both the society and economy of a virtual world but also the technological aspect which enables the virtual world to exist. LL actually branded the technology “Second Life Grid” but the name is little used outside a very specialised group of people). It’s not just the interface and the application that has to be learned: you have to learn how to live in the environment as well.

And it’s not easy. Just think at some stupid things like the need to have an Animation Overrider to change your animations. But first you have to buy one. For that you need to figure out where the shops are, and, once there, how the vendors work. Once you buy something, you have to learn how to ‘unpack’ it (every vendor does it differently!) or at least how to search in inventory. AO HUDs are also complex to understand, because they are tied to the way the technology works… and so forth. This is just a simple example, of course. Another is looking for a regular event to attend, figuring out how to get announcements, join a group, and  so forth. Buying or renting a home is also incredibly complicated. Mastering permissions is a nightmare.  But much more complex things are in store for anyone who wants to become a regular resident, engaged in creating content. Building, fashion design, or programming are full of tricks and tips and workarounds and ‘oddities’ that one has not only to learn but to master in order to become a proficient content creator with the hope of selling something.

But even the ‘daily’ routine of a resident is complex. How do you know where to search for things? It’s not only the search engine that is hard to navigate. It’s the whole concept. Think of the things you had to learn until you became ‘familiar’ with SL. It’s not just the technical bits, it’s the social and economic bits as well. How long did you take to figure out how much things cost in general? Is L$1800 a good price for an AO? Are mocap anims that better? What about clothes — are meshed clothes better or worse than sculpties? What accessories are fashionable — and, mind you, a Radar HUD (out of fashion!) is an accessory as well in SL! What things are you supposed to have? What places are you supposed to have visited? What things do SL residents talk about?

Even the simplest, most user-friendly interface doesn’t help much with all the above. Today, just like in 2003, there is a whole experience of travelling to a new country, where there is a lot to learn until you start to get familiar with the environment. That takes a lot — really a lot! — of time. And even if someone is very familiar with the way other MMOGs work, that experience is hardly enough to master the basics: yes, you will have an advantage in understanding how to move around and have a detached camera (which, btw, is unusual for most VWs!), but not how to terraform land or make sure that only your friends are able to use your sex bed. All that requires a lot of ‘experience’ — often learned by oneself, or, with luck, with another resident.

And here is the issue: all this takes a lot of time to learn, until you really start enjoying yourself. And that takes commitment. You can learn to use Twitter in 5 minutes and quickly have fun with it (Facebook takes longer!). But Second Life takes hours and hours and hours, and even after a few years, you will learn something new, or a different way of doing something else. As I often said, it took me over two days to understand how to link two prims together. I learned with a fellow resident who had also just started SL at the same time. We spent hours on the Morris sandbox clicking on all sorts of things to see if any of them ‘grouped’ and ‘linked’ the prims together. The point is, most people would get quickly bored by that. For us, however, it was a lot of fun. We were stubborn and persistent! If everybody else learned to link two prims together, so would we! And eventually we found out how to select the two prims and click on “link”. Not obvious! Not easy! But… it was fun!

Obviously the most part of all residents you meet every day know how to link prims together. We sort of share a lot of knowledge of SL. And this is the amusing thing: SL is mostly about a large community that managed to learn a lot of ‘common’ (or should I say uncommon?) knowledge. Everybody who either refused to go through that long process, or didn’t care much about it, are not any longer around here. So this means that one characteristic of most SL residents is their long attention spans. Longer, at least, than the average. And we live in an era when short attention spans are the norm. No wonder that we’re unusual.

Still, the point is that a ‘simpler interface’ like we have with ‘Basic Mode’ hasn’t really improved retention rates, and the argument that the competition has simpler interfaces and will thus be able to get more users has consistently been proven to be false. Cloud Party is the next candidate in line. Still, if you have tried out CP’s building interface, you’ll quickly see how much better and simpler SL’s is!…

No help from experienced users

When I logged in to SL, Lindens would routinely do in-world tech support: they would look for newcomers and help them out. You usually got your own ‘fairy godmother’ assigned to you — someone at Linden Lab who saw you for the first time and would help you out. Quickly this overwhelmed LL’s workforce, and so, for many years, we had the SL Mentors to fill in that role, with more or less success: at some point, there were 5,000 or more Mentors listed in the group.

Now, a common complaint by many residents is that LL has ruined the ‘first hour experience’ by getting rid of all those nice things that made a new resident comfortable and familiar. But the truth is that LL has not dropped everything — they have been experimenting. From a single ‘drop zone’ in old Ahern, we went to Orientation Islands, then those got access to Mentors and Greeters, then those groups were disbanded, then LL introduced several variations of Orientation Islands… when people complained that newbies needed some sort of ‘achievement system’, LL introduced some ‘games’ with HUDs, then, lately, the new ‘gamified new user experience’ with a redesigned not-quite-a-game-interface-but-almost. So LL didn’t really ‘give up’. Every year or so they try something new. But does it work?


Whatever they do, they are still unable to retain a significant amount of new users. Analysts and bloggers have complained about how LL always does everything wrong: the HUD-based system was unreliable, the whole of SL is always too unreliable, instructions are sparse, incomplete, or plainly wrong, and there is nobody around to help people out, anyway. Live chat was cancelled. The Mentor group was cancelled. So how do new users ask for help?

Here is a little secret. During the few years when we had Mentor-accessible Orientation Islands, retention rate wasn’t great, either. I used to spend a few hours every week on ‘Mentor duty’. Like many ‘veterans’, I was firmly convinced that this help was crucial for beginning users. But the truth is that not even one out of twenty new users would get in touch with the Mentors for help, even when they clearly knew that Mentors were there for help. And out of those 5%, very few remained around for long. And the few that did were exceptions. I still have many on my friends list. But… how much is ‘many’? I might have seen tens of thousands of ‘beginning users’, if not more. I might have talked to a few thousands. A dozen or so remains on my friends list. Do you see my point?

Of course you can argue that I was a terrible Mentor, and that all others were far better than me and still have thousands of friends on their lists. But, you know, somehow I’m not convinced. I do know a lot of ex-Mentors. Most of them are far better at explaining things than I am, and most of them just remain with a handful of ex-newbies on their lists. In fact, I shouldn’t be surprised: in my own case, I didn’t rely on Linden helpers or volunteer Mentors to help me out. I did it on my own because it was so much fun figuring things out. And, in fact, the majority of my old 2004 friends had a similar view. Obviously there are lots of exceptions: I do have a long list of users who have confessed that they would have given up on SL if they hadn’t be ‘lucky’ to find someone who helped them out on the first day, and this list is rather big.

But… correlation is not causation. I have always assumed that ‘good help to beginners leads to long-term residents’. Now I think — another of my working hypothesis! — that there is a special ‘spark’ on those people who made them stay, in spite of whatever help they got: the help only made the ‘spark’ be ‘lighted’, but it had always been there in the first place.

This obviously doesn’t mean that I’m against Orientation Islands or for bringing back the old Mentor group. Rather the contrary! In my obscure ‘theory’, there are a lot of variables that have to be ‘triggered’ in the ‘right’ way to turn a newbie into a regular resident. One of them is definitely having someone to help you right from the start: it’s the simplest questions that one cannot figure out that, when answered, will keep people interested. Also, Mentors/Greeters/Helpers fulfill another very important role: they’re the introduction to the social environment of Second Life.

‘Where is everybody?’

Linden Lab sometimes does surveys to residents that never return, asking them why they left. A few of those actually bother to answer that they didn’t find anyone around — either SL was ‘always empty’, or, more likely, they couldn’t find any ‘interesting’ people (for example, their RL friends) to meet.

Now this is a problem that is not to be neglected. The problem is that Second Life is so old that it still adheres to the ancient Netiquette: things like ‘what happens in SL, stays in SL’, anonymity, not asking people for their a/s/l, and so forth are part of the days where massively harvesting RL user data was seen as a Big No-No and the cause for shutting down people attempting to do that kind of thing.

But Facebook changed all that. The ‘new’ Netiquette is about revealing everything about your life, to your most intimate details; it is about connecting to all your friends; it’s about dating, dating, dating (and some porn, which is always present on the Internet). This makes everybody expect that when they log in to something new, the first thing they wish to find is a way to connect to existing friends who might be on the same service. All systems work like that these days.

Second Life doesn’t. Even if you figure out how to search for residents, there is no way to know if your friends are on this virtual world or not. They have anonymised profiles. A few lucky ones might have been able to register with their real names (until, of course, we have tens of thousands of ‘JohnSmithXXXXX’ accounts — which one is your friend?). Some might put their RL data on their profiles, but there is no way to search for that. Some have linked their profiles to Twitter, Facebook, or LinkedIn, but you cannot search for that, either. Putting it bluntly, there is no way to figure out if any of your friends are registered users of Second Life or not. And this will very likely not change, even though it runs contrary to all other online communities. So, obviously, the working assumption is that ‘nobody you know’ is on SL, and newbies give up.

The second thing that most online communities offer is a way to search for interests and join groups. Of course, if you are able to list your friends, and your friend is in a group, it’s likely that a good system (like Facebook…) will prompt you to join that very same group. In Second Life, obviously there is no way to figure out what groups your friends have joined. You can add some interests to your profile, but it’s not fully integrated. Even if you figure out how to click on a link to get a list of residents interested in the same thing you are, there is no way you can recognise any of these people, and groups are not searchable that way (or, if they are, I haven’t figured out yet how it’s done!).

Now you all know I’m all for the pseudonymity of Second Life. Things like IRC and discussion forums are still launched every day where everybody is anonymous or pseudonymous, and that never prevented these from continuing to thrive. It’s a cultural thing! I’m more comfortable with the way Second Life is run — and so are the million regular users. Nevertheless, I’m not for a choice of all-or-nothing. We should have a way to allow friends to find us if we allow that explicitly.

But is this really so important? Well, again, the evidence tends to show otherwise. While some people evidently joined because they were invited by a friend, the majority of the regular users enjoy their privacy, and that never prevented them to grow their friends list — most of which are perfect strangers they never met in ‘meatspace’. This is absolutely normal — for Second Life. And the ones that ‘get’ Second Life — the ones with the ‘spark’ — have no problem in dealing with that. So, in general, I don’t really believe that making people more searchable and having less privacy will lead to more user retention. It will, for sure, attract more people who are only interested in dating — and RL dating, at that — but they will quickly figure out that it’s much easier to do so on Facebook than on SL (imagine how easy it is to get undressed and do a show on a webcam; now think how many attachments and clothes you need to do the same in SL! Casual sex in SL requires a lot of preparation…)

But finding friends and adding groups is just half the story. For Facebook, that’s all it takes. But Second Life has so much more to offer! Groups about music don’t merely discuss music in IM chat, they attend concerts. And that requires figuring out which venues are worth visiting; which have the ‘right’ kind of music, according to our tastes; which have actually someone playing right now. But it’s not just about ‘music’: in Facebook (and most social environments) the kind of interactions you can do with your ‘friends’ is limited. Mostly it means ‘liking’ them, sharing links, doing private messaging, posting on public timelines, that sort of thing. In Facebook and other places you can also play a few games with them.

Now compare that with the immensely rich variety of things you can do in Second Life! The list is endless. And the problem is, most of the things you’re not even aware that they exist. I have to admit that there are actually very few things I found by myself; in most cases, I read an article about it or chatted to some friends who gave me a tip. There are obviously exceptions, but… that’s not the way people find out about things in Second Life.

For example, did you know that the Caledon islands actually have a railroad? I didn’t; I I stumbled across it by mistake. But I already knew about Caledon, of course. Why? Because I met someone, years ago, who knew Desmond. And then I was invited to host an event on Caledon. While I’m not a regular visitor of Caledon — not even an irregular one! — I’m aware of Caledon, I know what it is, and I have a vague idea what kind of events and things are happening over there. And even if I don’t know exactly what is going on right now, I know how to search for that information — or know whom to ask. That’s the kind of thing that is expected of a resident.

But a newbie has no way of knowing — based on a few minutes on the Orientation Island, or the current equivalent — that Caledon exists, or what it is about. They have no idea how many communities there are in SL, and where they are, and what they do. And they have nobody to ask. So unless you have at least some exploration interests, most of SL, if teleported to randomly, can quickly become boring, uninteresting, and, most of all, completely empty.

Again, I still think that there will be a huge improvement in the retention rate if you’re able to add your (RL) friends right from the start. The Destination Guide is supposed to give you a taste of what it feels to be a resident in SL. But most places on the Destination Guide might be empty — unless, by chance, you’re thrown in the middle of an event. Even so, it’s likely that the lag will be so great that a new resident will not understand what’s going on. We all have seen hopeless newbies dropping in the middle of an event and having no idea what they’re doing or what’s happening. Depending on one’s luck, this can be an awesome experience or a completely putting-off one. There is little that LL can do to make it better.

The technology sucks!

Let’s admit it: when SL was first launched, it was already dated.

And the more we progress through time, the more it seems to be stuck in the state-of-the-art of 5 years ago. There seems to be no way that LL manages to catch up. And everybody complains about lag.

Now I’m not going to be very thorough about LL’s technology improvements, I’ve blogged enough about it before, and half the residents in SL have already said far more than what I could have. The point here is the “wow factor”. When someone with a less-than-top-of-the-line computer logs in to SL with the default settings,  things look rather bleak, specially if the person logging in is a gamer and has unusually high expectations. Well, non-gamers have high expectations, too: after all, a webpage loads in a few seconds, so they ‘expect’ complex virtual worlds to be ‘instantaneous’ as well…

The clumsiness of SL doesn’t only affect visitors with high expectations. They drive content creators in despair — specially content creators who are used to work for different platforms. We finally have meshes, and these can even have physical bounding boxes/meshes for the physics engine, but LL didn’t implement the full range of options in the COLLADA files: no bump maps, no lighting maps, no materials, no special shaders…

Ironically, many of those things are ‘partially done’ in Second Life. People point out to the amazing things that you can do with water. You have all the tools that ‘modern’ rendering engines are supposed to provide. Unfortunately, they only work with water. And you cannot even ‘set’ an object to water — rivers flowing downhill have to be ’emulated’ with clumsy prims, sculpties or meshes and ‘fake’ texture rotation, like a bad 3D game of the 1990s. SL always had a selection of horrible bump maps to apply to prim faces, but for some stupid reason, LL never allowed people to add their own! (Why!?! If all the work has been done…) You actually can apply materials to prims, but it only affects the physics engine (and for someone like me, who is clueless about the complexities of Havok, I don’t see a lot of difference between setting a prim to ‘glass’ or ‘flesh’. I know there is a difference because the Wiki explains it!). For years we have been plagued with ‘Linden Trees”; even though Linden Lab is supposed to have bought a license of SpeedTree® years ago. Instead, we have to get used to multi-prim trees (or even meshed trees!) which might take a long time to render properly to give us a semblance of ‘immersion’.

Personally I’m more annoyed about Linden Lab’s ‘half-done’ projects than the ones they never started. It’s just because it seemed to be a waste of time — so many developer hours invested into making something outstanding, and then everything gets aborted, and nobody will pick up the old projects any longer.

So, well, builders get frustrated because they don’t have all the tools that a ‘serious’ platform should have, while users are frustrated because SL either takes too long to show anything, or because it’s way below the state-of-the-art of modern computer games. But does this really make a difference?

Blue Mars was supposed to be way better. The full range of tools that content creators ‘demand’ was available. Blue Mars loaded relatively fast, and had a fantastic gaming engine. Did it survive? No. So does technology matter that much?

I think that this question has two answers. At an individual level, yes, it does. Some people will see, for the first time, such a ‘poor’ environment that they will think that SL is a bad joke and not a ‘modern’ virtual world at all. Others, after years of waiting for LL to implement something, will give up. The important thing to notice is that none of these kinds of users will go to a different social virtual world — because, except possibly for Cloud Party, none have the look that SL has, or the kind of building tools that SL provides. So they will go elsewhere. That is, they will spend their time doing something completely unrelated to virtual worlds.

On the other hand, a million people are still around. Of those, perhaps hundred thousand are still creating content, which gets better and better every year. In spite of all of the limitations, content creators circumvent them and amaze us with new, clever ideas on how to make content that looks great. This hasn’t stopped; rather, it seems to grow more and more — and the growing economy tends to show that. Content not only gets better, it gets sold more. This is surprising — if ‘technology’ were a limiting factor. But apparently it isn’t such a limiting factor. Creativity seems to be more important — and often the desire to sell more and better content! — and, in fact, if you have seen other virtual worlds, the sheer amount of creativity in SL will make everything else — yes, even Blue Mars! — look like a poor copy of SL.

There is an old argument that to make SL grow, LL has to keep content creators happy. And, to a degree, that’s what they have been doing. Very slowly, things get a few improvements, here and there. But the truth is, if one takes a look at the kind of comments that content creators make, one would think that everybody who knows how to model a 3D mesh would have been gone away from SL, ages ago. But that doesn’t seem to be the case. Content creators complain but continue to create.

Why? This ties into the next issue.

Money is important

Catch a random veteran from 2003-2005, and it’s likely that you get someone still complaining that LL’s biggest mistake was to create an ‘economy’ and that everybody should be sharing their content, and our virtual world would be much ‘better’. I’ve met with these people since mid-2004. They’re still around, though, and not always on OpenSim, as one might have expected.

The point is, artists want to get paid. In Second Life, a good content creator — an artist — can get paid. And although some business savviness is important — to make sure one’s shop is well known — ultimately it all boils down to creativity and constant quality. If you have those two talents, you can make a living out of SL.

This is interesting by itself. One of the main reasons why SL content creators do not flock to OpenSim — when they hate LL so much as they say they do!) — is not because OpenSim has no economy. It does. You can exchange US$ or Euros for the currency on an OpenSim grid, and some currencies also work across some grids. Permissions are still not perfect, but getting better with each release. No, the problem is really having a market.

That market doesn’t exist yet in the many disconnected OpenSim grids. In terms of landmass, they might even be larger than SL, but they have less than 1/30th of the population. And, worse than that, a prospective content creator offering their services on OpenSim will have to register on a lot of grids if they want to attract all the market. So there is actually a lot more investment than on SL! And does it pay off? Not really. OpenSim pioneers are used to free content. Even the ‘second wave’ of OpenSim residents will not trust the existing OpenSim grids — here one day, gone the next — to exchange money on them.

Launch a new virtual world, using an ‘incompatible’ technology, and the trouble is… it will be empty. Or, well, have company-designed content. Until things like Cloud Party start to have a critical mass of content to make them worthwhile… they usually exhaust all their funding money. And if there is no content for sale, people — specially the ones coming from SL with high expectations — will give up. Who cares if the latest and greatest virtual world has the best technology and the nicest tech support, if, at the end of the day, there is nothing to buy to personalise your avatar and/or land?

Now Linden Lab was lucky. When they opened in 2003, hundreds of residents had already populated the virtual world with some content, and it was already for sale back then. A year later, when I joined in, the economy was in full force, and everybody (including yours truly!) wanted to become a content creator, because it seemed to be so easy to do. After a few years of this, well, Second Life achieved critical mass. They were lucky because there wasn’t ‘real’ competition, except perhaps for There.com, which required content pre-approval. Aye, that meant they didn’t have flying penises around like we do, but it meant a slower pace of content growth.

So ‘building cool things’ is important, yes. But just for the ones who ‘build for fun’. Being able to sell content is far more important, and people are selling content right now, and even selling more than usually. New content with higher quality attracts eager consumers, but also newbies who have far more choice and a wider range of prices — from freebies to highly expensive items — to pick from. It’s mostly thanks to the thriving economy that SL is the success it is.

Alas, there is just a shadow blotting this ‘content paradise’, and it’s called tier. With a contracting RL economy — or whatever the real reason is — some people are seeing sales dropping. Others are downgrading from Premium to Basic and dropping their land. After so many years of saying ‘you don’t need to own land to have fun in SL’ people are actually doing what they have been preaching. And why? Because the cost of tier seems to be too high.

I say ‘seems to be’ because there is no plausible, logical, rational reason for thinking that tier is ‘too high’ now. Please don’t misunderstand me: I wouldn’t mind a huge tier reduction. For top content creators, this would mean a larger profit margin; for smaller ones, it would mark the difference between being successful and bankrupcy. Communities could grow and afford more ’empty’ space, which they could use to display nice things that might attract more people — while right now they drop islands to cut their costs. In fact, the downwards spiral seems to have really set. It’s still a slow decline, but it’s a steady decline. And you know how that is: people start seeing their favourite places disappearing, Premium users being converted to Basic ones, and wonder if they shouldn’t do the same.

Well, the strange thing is that there is no ‘reason’ for tier being ‘too high’ — except perceived value. Somehow, after the Golden Era of SL, plus the RL economy contraction, although LL is providing the same service (or even a better one, at least in my experience), people value land less than before. This is not ‘explainable’ although we can give a lot of psychological reasons. The problem here is that Linden Lab is not really willing to drop prices, since all they sell is land — that’s their core business. Cut into their margins, and it means a worse service — which, in turn, would mean fewer people willing to pay. This is not something easily solved. I imagine that Linden Lab will not reduce tier, but they might calculate tier fees differently — again, I won’t go into this again. The point here is that all good companies have to adapt to changing market conditions, but not all are flexible enough to do so. Look at Microsoft: for over two decades, they had over-inflated the price of their operating system. But they had no choice, since it was their main source of revenue. Changing market conditions — Apple sells their own operating system cheaper and cheaper with each release; on the mobile & tablet market, Android is for free — forced Micosoft to admit that Windows 8 will be priced far below what is usual. But Microsoft can do that because they have another source of income: Microsoft Office.

Linden Lab has no such luck. All they make a profit of is from 3D content hosting. Sure, they get a slice of the LindeX, and another slice of the SL Marketplace, but these two don’t compare to the amount of tier they receive.

So they have three options.

The first is to massively cut their costs. Then they could lower tier. However, this means a complete redesign of the whole grid. I don’t think that’s feasible in less than 3 years. The changes they’re doing right now — pushing more and more content downloads from Amazon, keeping less and less things on the simulators, so they can increase sim/server density and provide adequate service — are a start, but not really enough for dramatic cost savings. They might cut into costs enough to compensate for the lack of tier income that they will get for 2012, but not much more than that; if the shrinking landmass continues in 2013, even these ‘tricks’ will not be enough. Of course, LL might be working on “Second Life Grid 2.0” as we speak and release it in 2015 — with new tier prices — and smoothly transition to the new architecture once it’s fully developed. We have no way of knowing.

The second is to ‘obfuscate’ their pricing. As I suggested elsewhere, one way of doing it is to move from a ‘prim-based’ economy to a ‘Land Impact’ economy, but using other abstract metrics — say, pay for Ktris. Prims are easy to understand, but more abstract metrics are not, and this would mean that in some cases people would pay less tier and be very happy about it, while on others, they would pay the same as before. Obviously residents are clever and will quickly figure out the trick, but it might help during a ‘transition phase’.

Transition to… what? A few articles back, I suggested a third option: imitating Kitely. Kitely’s ultra-cheap pricing are not ‘speculative’ in the usual sense of the word — many grid operators start with ultra-low prices to attract customers, and, as their running costs start to skyrocket, they increase pricing until it becomes almost as high as LL’s (which is not surprising!). Kitely, by contrast, has a rather clever business model — sims-on-demand-on-the-cloud. You just pay for the amount of time your sim is online and being used by someone. This is actually very clever, because it means next-to-zero infrastructure (they just use Amazon’s cloud services) and they can afford to add just a little overhead on top of Amazon’s pricing, which is enough to make them profitable. If everybody gives up on Kitely, since they don’t need to ‘own’ any infrastructure, they can survive with little risk of going bankrupt. It’s a solid business model!

Linden Lab could most certainly emulate that model and do pretty much the same. And that would really mean ultra-low sim prices without endangering LL’s business model. Well, to a degree. The problem is that they have a staff of 200 which they have to pay, beyond all the infrastructure costs. A sim-on-demand model would be good as an extra option, but not as a replacement, since that would mean a collapse of their income 🙂

It’s not easy. My guess is that LL is trying to see if they can at least contain the landmass reduction while it is still manageable. But changing the business model at this stage is very, very risky, specially because they have such high running costs. This will definitely require a lot of thought.

The truth is, it’s even harder to draw a correlation between new/old users and land owners. I’m a rather ‘old’ user these days, but I don’t own a sim. I’m a Premium user since September 2004 but haven’t bought any land beyond the original 512 m2. I do rent a few parcels, but nothing really worth talking about. There is nothing that LL can do to make me spend more money in SL: I’ve reached my limits a long time ago 🙂 All the money I make from the few content sales enable me to pay for tier and the Premium account — and little else besides. On the other hand I know a lot of people who have jumped into SL and were buying sims just a few weeks afterwards, and are still around doing business. So it has nothing to do with ‘age’ — the next newbie that registers might become SL’s biggest landbaron, while oldtimers from the Closed Beta might just be happy with their ‘free’ 4096 m2 and never buy any more land. While reducing tier would be great for existing users, it is not a way to attract new users, nor a way to keep them returning. The truth is, before you buy your first land, and evaluate what it’s worth, you really have no idea if tier is ‘expensive’ or ‘cheap’. It also depends on one’s spending habits; for some, spending US$300 per month on a sim is ludicrously expensive, for others it means giving up eating for a month 🙂 So the argument that ‘lower tier will bring in more residents’ is not exactly accurate. It might be true, but only very indirectly. It might prevent some people from leaving, depending on the price cut. As I’ve explained elsewhere, for some, a 30% price cut is still too expensive. Reducing tier to, say, US$30 for a full 15,000-prim-sim is another story! That would certainly get the landmass growing again, and fast. But it won’t happen, since it will blow Linden Lab’s business model apart.

No, the solution is not tier. It would help, but won’t ‘solve’ anything.

It’s not about new users…

And now we come back to Rod Humble and why he’s addressing veteran residents, and to my working hypothesis.

I think that the ‘user retention dilemma’ is just a smoke curtain. LL gathers a lot of statistics, even though these days they don’t share them with us any more. But we still have historical data. Until 2006 or so, it was quite easy to see a correlation between ‘news on the media’ and ‘new signups’. I forget where there was a nice link showing that trend: every time some major news venue said something about SL, signups would rise. At some point, however, it got stuck at the 10,000 signups per day. Then, for a while, with the complete redesign of the website and the launch of the SL 2.0 Viewer, signups peaked at almost 20,000 a day, then dropped back to the 10,000 or so we still have today. Nothing seems to change that. All announcements that LL does are never picked by the media these days, and so there are no more signups.

But signups are worthless. It just interests people like Facebook — 800 million users, yes, but how many are active? And how many are paying customers? You see what I mean. LL doesn’t care to claim that they have almost 4 million signups per year (!). Even if they had 40 or 400 million, why would they care? None become paying users. And no matter what LL does, the signup number doesn’t rise, and, more important than that, the number of active users doesn’t rise, either — it’s actually on a slight decline. The important point here is no matter what LL does.

This should be a hint.

Of course, what residents claim is that LL is always doing the wrong things and that’s why nothing seems to work. Granted, there is an unlimited amount of different things that LL can do. But the more ‘obvious’ ones are somehow mentioned on this article. The rest are pretty much variants — ‘improve scripting, more users will come’; ‘get us new avatars, more users will come’; ‘give us a better inventory, more users will come’. Possibly. But unlikely.

This is my working hypothesis: due to the very extraordinary circumstances of what Second Life actually is, it has hit market stagnation. There are no more users for social, visually contiguous virtual worlds with user-generated content. We have all there are in the world!

Now consider this for a moment. People — and I include myself in the list! — always claimed this would be the next generation of social platforms. There were so many good reasons for it. And even the entertainment media — movies, TV series, etc. – helped along, showing how cool virtual 3D environments could be. But somehow this never becomes reality. It’s like the domestic robot or artificial intelligence — always a few years away 🙂

Why? I think that 2D interfaces are getting easier, and that’s the appeal they have to mainstream users. We also have one generation of people who have already been born with 2D interfaces. That’s what they learn to use in school, and will use all their lives. So anything 2D is accepted immediately. 2D interfaces are also far easier to learn and use, and visually appealing, compared to the previous generation — console-based text interfaces. And the generation before that — punching cards! — was a nightmare of user-unfriendliness.

By contrast, 3D interfaces are harder to learn and use. They require people to have at least some spacial orientation skills, and that is actually much less widespread than people think. The way SL works requires a lot of learning. For instance, instead of a ‘cursor’ to represent where you are on a 2D screen, you have an ‘avatar’. This requires a lot of explaining (granted, on a first-person-view you might avoid having an avatar, but you will still interact with other people’s avatars).

But some things are utterly different. A few years ago, sharing a video with a friend would mean sending them a link to YouTube. These days you can go on Google Hangouts and chat with them while you two watch the video together. How many of you have actually done that? Probably not many. Why? Because the current communication paradigm is ‘quasi-near-real-time’ — we send messages which others will read and reply to ‘after a few minutes’ (or hours, or days; the point is that you will do it a bit after receiving the message).

While real-time video chatrooms are great for dating — because we already use the phone for dating and are used to real-time communication, at least using voice — there seems to be little else to do on a video chatroom than to do dating, or keep in touch with family members. I mean, I personally never have seen anyone using videoconferences except for dating, meetings, and pretty much replacing the phone. Novel ideas like sharing videos and watching them together simply are ‘too new’ and somehow slow to catch on.

But Second Life is all about doing things together online, in real-time. It’s harder to set up a video session where people can watch a YouTube video together, but it’s possible. Things like Metanomics show how a large group of people can watch the same thing and chat all the time. But it’s not only that. When you shop online on a 2D shop, it’s a lonely experience; in SL it can be a shared one, and it often is. You can listen to music on your computer and even share with others. It’s not unusual for friends to get together, find some music on YouTube, and enjoy themselves for a bit. But how many do the very same thing regularly online? Except for the occasional “share video with friends”, things like spending hours listening to music or video is usually done in private. Not so in SL. While you can set a streamer just for your own use, usually you hang out with others on clubs or live music venues. It’s a completely different experience.

So different, in fact, that very few people are used to it. Now I can imagine, since so many academics are researching SL, that the next generation of students will get used to the whole array of concepts that virtual worlds bring. They will be harder to learn, yes, but that process will happen. I mean, students get taught Photoshop in school too, right? And SL is not that much harder to use than Photoshop. While 20 years ago nobody would even know what Photoshop was good for, much less use it. These days, however, it’s become commonplace. Why? Because we have a generation of people used to it.

I think that’s the ‘problem’ with Second Life. It is too complex, too different, bringing completely new experiences, which are hard to learn. The vast majority of mainstream users simply don’t have the appropriate mindset to deal with it. For them, it’s an incredibly complex tool with very little entertainment value. But for those few for which learning complex tools is part of the fun, SL is the right place to be. The point is that there are very very few people willing to do that. And I claim that all of them (or pretty much all…) are already in SL, or, worse, they have already left.

I’m not saying that among those 4 million new registrations per year a few will not stay. Since LL hardly promotes SL, there is always the chance that someone who has the required mindset to become a proficient SL user will finally hear about it, log in, and remain ‘hooked’. This has been harder and harder after 2007, when the media lost interest in SL. That means that there might be some untapped potential over there. But it’s a question of the order of magnitude. How many more potential residents are out there? Certainly not ‘a billion’. I would claim that very likely they’re not even a million, but more likely close to hundred thousand. So, assuming the perfect marketing campaign, what this means is that, at best, LL can hope to duplicate their user base. But that’s all. Until a new generation is adequately taught how to use 3D virtual worlds, and understand its value — in socialisation, in entertainment, for work, and so forth — this user base will not grow. Not substantially, anyway — of course, over the decades (yes, decades!) more and more people will join virtual worlds, as they become familiarized with its purpose, but that will take a long time.

Now at this point I should make two things clear.

The first is that this is not a bad thing. I’ve often written about how LL doesn’t need to be ‘ashamed’ of creating a successful niche market — one that has a digital economy worth 500 or 700 million US$ annually. That’s not exactly bad. There are a lot of companies specialising in niche markets, which will never ‘go mainstream’. Autodesk is a typical example: not everybody in the world has the need to use AutoCAD, Maya, or 3DS. But the millions that do require those tools are Autodesk’s market, and they’re enough to make the company profitable. Adobe is a ‘mainstream’ brand, but that’s because PDF and Flash are mainstream products. But they’re used for free. Adobe’s line of products, from which they do derive an income, is for a professional audience — which is very limited in size. Adobe does not offer Photoshop to 2 billion humans with a computer, but just to a small subset of them — enough to be a very profitable company. You can fill in examples for whatever you wish. Just think of a niche market that is not mainstream and you’ll see successful businesses in all of them. It’s true that companies addressing the mainstream market tend to be bigger and more valuable, but even the ‘comparatively smaller’ companies targeting niche markets are successful (Autodesk, for instance, used to have as many employees as Microsoft, even though they address a much smaller market).

So it’s not bad that SL has a small market. LL has a huge market share in it! in fact, if we consider that IMVU is in the same market as LL, this means that LL has abut 30% of the consumers in this market, but if we look at the amount of transactions happening inside SL, it’s probably closer to 90% or so. Again, this is not bad. It’s just ‘disappointing’ for the ones among us (including myself) that would like to see SL grow to become a 3D Facebook or a 3D desktop operating system — universally used by pretty much everybody with a computer.

The second thing is that probably Linden Lab is aware that they cannot grow their market. They might be a bit puzzled why 4 million users still bother to register every year (specially with next-to-zero promotion), and maybe if they do something about it, they might get a few thousand new active residents. But they might be aware that the ‘salvation’ of LL’s budget and profitability lies not on the new users.

Instead, it relies on the existing users, and making sure they consume more — namely, that they grow their tier, or whatever LL introduces next. So Rod Humble, unlike his predecessor, is aiming to upselling SL to its existing residents. And in this case, it might not be so surprising that he’s asking veterans what to do. After all, the veterans are the most outstanding among all outstanding residents: no matter what, they remain around, year after year after year. Complaining all the time, that’s true, but they’re still faithful. They might have reduced their tier, but they haven’t abandoned SL. Among all power users, in the middle of this tiny market, the veterans are the ones that have this special ‘spark’ which makes them stay around. Now I’m not saying that the next newbie that logs in doesn’t have that ‘spark’ and that only the existing veterans are worth talking to. Today’s newbie becomes tomorrow’s veteran. But by studying how veterans think, Rod Humble might be able to figure out what makes them ‘tick’, and understand how to — eventually — turn more residents in similar ‘veterans’. The truth is that most active users in SL are veterans — or, if they technically are not (i.e. registered recently!), they very likely have this ‘spark’ that will make them stay for years and years.

So, for the short term, what Rod Humble needs to figure out is what to do to make sure that veterans don’t leave, and, instead, start consuming more. Also on the short term, he will need to figure out, among the current active population, how many have this ‘spark’ that will turn them into long-term customers of LL, and how many will probably leave after a few months. Dear old Rodvik, himself a gamer, and used to deal with the strange ways that gamers relate to their games, is trying to figure out what makes a SL resident become an active user and later remain for years and years. He’s trying to understand why people complain so much but still pay their tier; or, conversely, what will make them complain less and pay more tier 🙂 Failing all that, losing a veteran is a catastrophe, since they’re hard to replace — what can LL do to prevent veterans from leaving? Specially, of course, veterans owning a lot of land.

It’s not about new users really. It’s about making sure that LL keeps the current market share, understanding that it will not grow substantially — not in this generation — and making sure that they can adjust the company to provide for the existing user base, keep them happy, keep them paying tier. Companies like Adobe and Autodesk — which have insanely expensive software! — have figured all that out, and, as a result, even if their own customer base doesn’t grow much, they know how much they can sell them, and how often they will come back to buy more. Based on LL’s current business model, Rod Humble has to figure out what LL has to do to do the same.

For wannabe content creators, this is not so good. Amateurs became rich during the Golden Age of 2006/7, when it looked like Second Life could become a mainstream product, and, as such, even bad content would sell. In a saturated market, only the best and more creative content creators will survive, like they have done so — admirably well, in my opinion — for the past five years. But the years of ‘mainstream’ SL will not come back. Not for a couple of decades at least.

Print Friendly, PDF & Email
%d bloggers like this: