Where We're At
Second Life is an innovative virtual world populated by human-controlled avatars, considered by many to blur the lines between "real" and "virtual" life. Although Second Life members act only through digital avatars, the interactions have real (often tangible and measurable) impact on our lives--at its most crass, Second Life allows us to accumulate play-money that be converted
into American dollars; at its most sublime, humans at opposite ends of the globe can affect each other's emotions
through Second Life's communication tools.
Second Life is always changing, its residents constantly anticipating new developments and improvements. While I'm interested in Second Life's future, after a year of residency I find myself asking "Where are we now?" The simple answer is "Not where we should be."
Where We're Not
Second Life hasn't weathered well. It might have been utterly cutting-edge back in 2002 when it was first demonstrated, or even in 2003 when it was first commercially-launched, but in 2005, things are different. Second Life is, at its core, a "massively-multiplayer" online environment, and although isn't a game, is related to massively-multiplayer online games (MMOG) such as Everquest
, The Sims Online
, and World of Warcraft
. The MMOG market is hotter in 2005 than ever, with subscription-rates in the millions and new games sprouting up all over the place. Most MMOGs have a limited shelf-life. It's unlikely, for example, that World of Warcraft will still be going strong in another three years. Probably another game will have taken its place. In Second Life's case, iterative improvements are made many times during the year, but the "engine" that runs the world of Second Life is still dated. At a certain point, it just doesn't get much better. From a technical-performance standpoint, compared to recently-released MMOGs, Second Life can't hold its own.
The fewer the apparent technical limitations, the more enjoyable the virtual-world experience. Second Life is both ahead of its time and behind the times. It is ahead of its time because the computers and networks required to experience Second Life optimally are out of reach of most consumers. Perhaps when Internet2 rolls out and we are all using quantum computers, we'll get a really fluid, seamless experience. Second Life is behind the times, because even though its engine has extreme hardware requirements, it isn't the best engine available: the features the engine allows are good, but there are better engines available (and in fact, are being used in today's games). To put it practically, if my computer can run the graphically-intense MMOG World of Warcraft
without so much as a hiccup, why should it choke on Second Life? Given the choice between a gorgeously-crafted, fluid game experience and a haphazard, stuttering virtual world, which do you think I'd prefer? Most hardcore Second-Lifers would at this point be pointing out that comparing World of Warcraft
to Second Life is like comparing a video game to a chat-room, and I agree. The problem is that Second Life is marketed like a video game, but isn't.
Truth in Advertising
"Bungie jumping, sword-fighting, snow-mobiling, water polo -- try some Resident-created activities or make your own," declares the Play
section of the official Second Life web site (April 18, 2004). In its history, the site has shown many screenshots depicting scenes of apparent video-game style action, even promoting the user-created "U:SL
" project, an attempt to bring first-person shooter game-play to Second Life. The fact is that Second Life makes for great screenshots, but its action-based game-play doesn't even come close to any decent video game made in the last ten years. There is simply too much overall lag (and I'm including both network latency and lost frames in this) for responsive game-play of any kind. I originally joined Second Life with the expectation (based on the available marketing material) that I'd be able to play such games in-world, but in fact, Second Life's performance problems prevent most forms of responsive action from taking place. While it's true residents bungie-jump and sword-fight and snow-mobile, it's also true that Second Life sword-fights might as well be performed drunk, blind-folded and underwater; snow-mobiles, like other in-world vehicles, have a tendency to get stuck, can be difficult to operate, and have been known to lose their driver while crossing between servers.
Resident culture is such that performance-issues are not addressed directly and openly--most games and activities are, of course, created by other residents (i.e. don’t rock the boat, be careful who you criticize, etc.). The only fault (if any) with the creators of these diversions is that they've tried to make a Lamborghini out of a lemon--you can give the lemon a great paint-job, but at the end of the day it's still bitter and immobile. The fact that the U:SL project was actually undertaken boggles the mind. It looked fabulous (i.e. made for great screenshots), it was heavily-promoted by Linden Lab, but it was utterly unplayable. On the flipside, Second Life games that don't rely on accurately-responsive action, such as the puzzle-game Tringo
, have been a massive hit. Tringo now dominates Second Life's daily activities, and has been licensed by a real-world company for distribution. If Linden Lab is going to paint Second Life as a place to play, it should shift the focus from action games to games that can actually be played as one would expect.
Content is King
One of Second Life's greatest strengths is its user-created content. Some professional-level content has been generated by residents, but unfortunately, like the real world, ninety percent of everything is crap. Because Second Life operates on a free-market economy, there's a chance that even if a piece of content is exceptional, its creator might not have the funds to promote it. It's possible to find content based on popularity, but this operates on a power-law
, creating an inner core of A-list content while there may be comparable or better work on the sidelines. Furthermore, some creators aren't even interested in publicizing their creations. It all adds up to a scavenger-hunt for decent content, through which reams of garbage must be sorted. No commercially-successful video game requires its users to supply all the content--even MMOGs like City of Heroes
gives its players a predetermined palette of configurable components
with which to build their own superheroes. As disappointing as The Sims Online
might have been, they didn't require users to build the social-space from the ground up. To be successful, you have to put your best face forward (and offer something behind the gloss). Second Life's resident culture generally demands that Linden Lab refrain from "picking favourites," and so the company has stopped choosing hot-spots, instead auctioning off "Sponsored Links," which rewards wealth, not creativity. My recommendation is not only that Linden Lab pick all the favourites it wants, but commission skilled residents to build "official" sites that display the best of what Second Life has to offer.
The concept of virtual "zoning" is still a hot topic of debate among residents, some of whom desire the consistency of a themed area, and others who see the beauty of chaos. There are no successful MMO environments I know of that are so chaotic. Successful virtual worlds are designed on purpose, not by accident, and I think it's important for Linden Lab to provide newcomers an expected environment, at least at first. The newbie "Welcome Area" should be expansive, relatively free of advertising, packed with top-quality content, and populated with official, litter-free themed builds.
Subscribe to Everything
Second Life could be compared to a gated community surrounded by a high stone wall. Only card-carrying residents and approved mail can pass the gate freely. Just outside the gates are billboards and leaflets selling the features of what lies beyond the gates, displaying pictures of joyful and active citizens. An outsider might wonder why, if life is so grand inside, the wall isn't low enough to peer over. I wonder this too, and I think there's a lot to be gained by increased visibility. I'm not proposing that Linden Lab let outsiders stand outside resident windows--visibility still needs to be controlled. I am suggesting that Second Life pique outsider curiosity by opening up its walls and allowing some information to pass through. Not only this, but provide the outside world some useful information. Without getting caught up in exactly what technology would permit this, what I'd like to see are official RSS feeds that would allow subscriptions to numerous forms of safe data. Feeds might include data such as the number of residents on a given parcel of land, the list of PG-rated community events, the list of land for sale, or the amount of Linden Dollars changing hands. Information that is not only of actual use to residents, but possibly the rest of the world. Public feeds would not only allow non-Linden and non-resident programmers to create functional tools for analyzing data, or even just visualizing data. Additionally, the official Linden weblogs, which are infrequently-updated could at least be given some life with feed data--Robin Linden's blog
could, for example, give us a ticker displaying events and the number of avatars attending each.
Power-Users: Second Life As a Creation Tool
Second Life has a number of professional and "serious" applications, mostly related to education and health. Linden Lab could expand the use of Second Life into more industries by allowing in-world creations to be exported, if not created and modified offline and imported. I've spent hours fiddling with Second Life's "prim" building blocks, but the most I can do with the creations afterwards is take a screenshot. I can't get that creation out of Second Life. Once the excitement of creating objects wore off, I stopped making new things, because they have no use outside of Second Life. Forcing me to view my objects exclusively in Second Life didn't encourage me to do so, it did the opposite. I am a professional designer, but Second Life is the only tool in which I've been able to easily create and texture simple 3D models--it's great for so-called "rapid prototyping." If Second Life models could be exported in a common 3D file format, I'd be able to use Second Life as a professional production-tool. But I wouldn't want to.
Creating things in Second Life is time-consuming and subject to distraction. One's attention, in-world, is constantly being divided between the task at hand, and reams of miscellany, such as instant-messages and random visitors. It is interesting to see people building objects in-world, but this shouldn't be considered a form of entertainment. There are existing ways in which to secure some privacy, but I don't consider these methods very elegant. Why limit creators to the confines of Second Life? I'd love to see an "advanced" editor for building that isn't part of the main Second Life grid at all (or any grid, if possible). Give me a skinnable "ground" plane as reference, give me my building tools, and let me build in peace. Then, let me click an Upload button and send the thing into my Second Life inventory. A copy would go into Second Life. The file for the master object would be sitting on my own computer. Ideally, an "advanced" item would be able to be "fused" on upload, meaning that the primitives are all joined to form one object and hidden faces are removed to keep polygon counts low (thereby making the object more efficient). Only uploaded copies are fused—the master stays on the client side.
As a side-note, keeping a master-object on the client side as backup would help to correct the build-killing scourge of "Prim Drift," a bug in Second Life that has yet to be fully corrected, and which is one of the main reasons I don't build anything.
Second Life assets (items, textures, sound-effects, etc.) are all files. Some residents, particularly the more active ones, tend to accumulate many assets. The current method of organizing these assets is a vertically-stacked "tree" system, not unlike Windows Explorer
. The problem as I see it is that while this system resembles a "standard" file-system, it is not similar enough to either Windows or OSX, and is frankly too cluttered most of the time. I'd like to see inventory managed in "basic" and "advanced" modes. Basic would include a highly-streamlined, abbreviated in-world inventory system, and the Advanced mode would allow fine control over the files and perhaps even utilities like the removal and/or consolidation of redundant items, and batch-renaming files.
Since its launch in 2003, Second Life has had numerous substantial improvements. In 2005, however, its virtual-world-engine just can't compete with the recent crop of massively-multiplayer experiences in terms of responsiveness and overall quality. Although it still surpasses most MMO environments in terms of social and creative opportunities, Second Life won't be in a position to compete seriously until Linden Lab rebuilds its engine. Moving Second Life from the "Havok 1" engine to the "Havok 2" engine is currently one of the most-desired features
out of 243 proposals put forth by Linden Lab for resident voting
Until the new engine is implemented, the Lindens could set Second Life in a league apart by publishing a plethora of public RSS-style feeds for leverage by the community and outsiders. Any major new implementation of Second Life needs to address the needs of power-users in order to progress further towards "serious" and professional applications.
Finally, Linden Lab is getting bogged down with democratic and consensus-building efforts in its relationship with residents. There are some tough decisions ahead, particularly where quality of content is concerned. With a stated goal of hosting one million residents by 2007, any short-term community upheaval resulting from executing these tough decisions will seem like growing-pains on hindsight. It's time for a prescribed burn: eliminate the dead wood so that the forest flourishes.