In the MUDs, MOOs, and MUCKs the ghost town was a phenomenon of over-creativity but this wasn’t much of a problem as this clutter was largely invisible and consumed little resources. It was just some disk space and few systems got beyond a single server. The ghost town phenomenon in VR habitats is troublesome because it is an inescapable visual blight and obstruction of free movement squandering computing resources and ultimately limiting user creativity. It’s a consequence of fundamentally bad design. the expression of the essential flaw of mixing private and public activity in the same space which is often exacerbated by server architectures that don’t dynamically allocate resources but instead simplistically divide them across a grid. It’s potentially harmful in that it can be an expression of a system that perpetuates the neurotic pathology of property; the creation of false scarcity on which to build potentially exploitative economic models. SL was created to invent a new kind of real estate with which to conduct speculation as a basis of monetization. It re-created the basis of the contemporary real world’s economics with predictable results–like gentrification, discrimination, and even prostitution. There are Youtube videos of people in SL being verbally abused for walking through open doors into the wrong place in a public space. I’m sorry but, If that’s routinely happening in a ‘game’ you’ve developed, you screwed up. This is why it inspired a whole Hobo Island as a protest against this and the invention of ‘skybox’ scripts as a hack around it. Originally, SL wasn’t even supposed to have real estate. It was originally intended to be a VR theme park whose entire content was proprietary and where even avatars were mostly limited to a few standard variations. But the developers quickly realized they lacked the manpower to create that content. And they weren’t all that sure the theme park monetization model could work because the Internet culture abhors toll booths. So they created this real estate model to harness free end-user labor to cultivate the environment and build value for virtual property they ultimately own entirely. And people tend to get creative with whatever they are allowed to get creative with, which was mostly the landscape and junk you could put in it until people discovered ways to hack around the inherent limitations to expand avatar options.
We have no control over the structure of the physical world. It is what it is. Its existence is independent and purposeless. The built habitat is all adaptive reuse. But, unless you are creating an environment to be a very literal simulation of the physical world, there is no point to including its limitations. In the physical environment we often don’t have a choice about the aesthetic continuity in the built habitat–though we often create institutions, like Home Owners Associations, to try and address it with sometimes oppressive results. There is no reason to perpetuate that conflict in VR because there is no reason for aesthetically disparate, clashing, spaces to directly coexist. VR space has no independent existence and is always purposeful. We have the ready option of a multiverse and the ability to teleport and portal between purpose-built spaces regardless of the virtual distances between them. VR has no inherent topological or geophysical constraints. There is no real distance, no limit of areas or volumes. There is no logical reason to have universally consistent rules of physics. There is no hard need to impose a grid on it, as we compulsively tend to do to physical space in our culture. The only real purpose to an all-inclusive contiguous VR space mixing private and public space is, quite simply, to create a commodity and exploit its artificial scarcity.
VR, in a general sense, is not a simulation. If you are using it to create a simulation of something like a living, evolving, forest, you are engaging in an active computational experiment filled with interactive agents that have a reason to persist independent of a viewer. There’s nothing wrong with that if you have the computational resources to spend. That doesn’t strictly need VR to do and if you are applying that to VR, you’re likely using it as a way to create a particular kind of dynamic backdrop to the VR experience that may, or may not, need to process in real-time continuously. It’s quite likely that VR habitats will employ simulation as the basis of procedural systems to generate various features. It might be very nice, but it has nothing to do with the point to VR itself. It’s just another one of many ways to create a particular virtual space for a particular purpose. Why would you design a whole VR platform around this one simulation and its constraints? Why limit everyone’s options to that?
Most certainly, people are expressing themselves in the creation of architecture and environments around it. In SL this certainly represents a lot of man-hours of investment creating a lot of value. But it represents a minority of the actual amount of time people spend in SL. The majority is social interaction, most of which revolves around avatars and never involves but a tiny fraction of the architecture in the environment. I’m in no way suggesting this architectural creativity is somehow wrong or that there should be no participatory process to crafting relatively large contiguous public spaces. I’m actually arguing that this is unnecessarily constrained by the introduction of an artificial scarcity of space created by all-inclusive contiguous environments. Why should I be limited in my creativity to a little lot? Why shouldn’t I have all the space I want when it’s just bits? Why shouldn’t I have as many spaces as I want? In that all-inclusive space, the moment you put locked doors or a fence around your creation, it has ceased to be expression, communicating openly. You have stolen and enclosed a piece of the public space, the commons, to fill up with junk no one ever sees and if what you’ve created is surrounded by an ugly fence and is aesthetically incongruous, you’ve created an eyesore. You’ve committed vandalism. Right now, with the architectures of so many VR platforms based on single contiguous spaces, this is inadvertent. But it doesn’t have to be that way. VR doesn’t have the limitations of the physical world unless you deliberately engineer them in. My argument is that it is fine to impose those limits on the discrete environment for the sake of its specific purpose or theme, but not on a whole VR platform. A platform is supposed to function at a higher level. The VR platform itself should be application-agnostic. A multiverse simultaneously supporting as many possible kinds of space models as possible, persistent or temporary, public or private, without conflict.
The mistake here is failing to understand that, as a social space, the primary purpose of VR space is to be a stage for the performance of users to each other, and in the private space context, an extension of the avatar. This is illustrated by the invention of the skybox script in SL. A skybox is a script-generated virtual box that is put high in the SL sky, beyond where users can normally fly their avatars, and created with a teleport link that people can pass to friends or place on a public object to access it. It was inspired by the runaway gentrification in SL and the realization that, logically, private space is only temporary in function, its limits of scale are arbitrary, and it doesn’t need to exist except when you and others are using it. SLs primitive design offers no option for creating on-demand pocket-spaces for this purpose. If it did, its real estate racket would collapse. The only option is to buy or rent space in the contiguous environment and put up walls and fences around it. And, whether or not you’re using it, this fenced-off space stays there, cluttering up the environment. But some clever users realized there was this infinite amount of space in the SL sky no one could reach or see and, with scripts giving one the power to auto-generate structures in that location on-demand, they could create those on-demand temporary private spaces, giving people all the benefits of private space without having to buy or rent it. These scripts are attached to the avatar. They, and the structures they temporarily create, are extensions of the avatar. There can be any number of them, each designed to suit a particular activity, theme, or mood. While they exist, they can be as open or private as one wishes. Some people realized they could use this in very creative ways, like making their own versions of the Tardis that seem bigger on the inside and whose controls can generate different exterior ‘stage set’ environments outside the Tardis door as different imaginary travel destinations. You could create vast temporary complexes of spaces with elaborate integral behavior scripts, for things like dungeons in role playing games, and never have to buy any space. The skybox represents how space in VR should work, users intuitively understanding this and devising a clever hack to make it work–in spite of the developers. SL should be nothing but skyboxes. Unfortunately, scripting remains too clunky for most casual users and so skyboxes became an appliance with more-or-less fixed features. SLs developers should have gotten the message, but apparently never did.