Guest Editorial by Galen: Sansar, A View from the Trenches

Sansar: A View from the Trenches

One Sansarian waxes on Linden Lab’s strategy and life for its pioneers

By Galen

Linden Lab, creator of Second Life, unveiled Sansar to the public late last July with little fanfare. Like some other people, I have been working full bore in Sansar almost exclusively since before then. I get myopic, I suppose. But every now and then I get a reminder that there are other social virtual reality worlds out there, not to mention the venerable Second Life itself. Sometimes those reminders come from bloggers, vloggers, podcasters, and other technophiles. It sometimes surprises me that Sansar seems to be getting so little attention, compared with Facebook Spaces, VRChat, and many other nascent social VR platforms. Some bloggers don’t even appear to know that Sansar exists. But it seems that most of the bloggers who do talk about Sansar are not actually Sansar regulars. It’s interesting to see how outsiders see us, but I think they give readers only a cursory grayscale view of Sansar. I thought it would help to give a perspective from someone on the inside.

Galen Wandering Isles.png
Wandering Isles experience in Sansar by Nya Alchemi; 2017 snapshot courtesy Wurfi

Let me clarify that I don’t work for Linden Lab or get paid by them for any of my work. And I haven’t always agreed with the directions LL has taken, particularly with Second Life. But I’m a techno-optimist. And I fundamentally believe in Sansar’s current direction. So much so that I’ve abandoned all my other major ventures and forsaken the other social VR and similar platforms in favor of investing typically over 70 hours of each week to working full time on developing my ventures here.

I think I know Linden Lab as well as I know any organization. I joined Second Life in 2005 and made a decent side income there all the way through 2011, when I finally took my first “break” from SL. SL won the online virtual world “game” a long time ago and has dominated since. It’s only been in recent years that SL’s dominance has been challenged, largely by contenders carving out the new social VR space. I’ve thrown my lot in with Sansar. Does that mean I am 100% certain that it will win the social VR game and dominate thereafter? Surely not! Although I’m convinced that this is truly a winner-take-all game like the one won by Second Life, I don’t think anyone has any solid reason to believe that any of the major contenders right now will surely win. The ultimate winner may even be waiting in the wings. I just think that Sansar has the best chance of success. Call it my educated guess. And I’m betting my all on it right now.

Sansar is late to the game. VRChat was founded in 2014 and officially opened in February 2017. High Fidelity was founded in 2013 and opened into public beta in April 2016. Sansar has only been open for a bit over seven months now. And other entrants like Facebook have deep pockets. So why would I think that Sansar has a chance of surviving, let alone winning the social VR game? One of the more curious things I’ve heard in recent years from many quarters is that Second Life won the virtual world game because they were there first. The reality is that SL was not at all first. Active Worlds, founded in 1995, was nearly a decade ahead of SL, had a very solid technological basis, and had a vast following by the time SL opened up in beta. Blaxxun got its start around the same time and had quite impressive tech. I believe SL won because it had the best strategy. And I have reason to believe that Sansar may have the best strategy, despite all its current shortcomings.

Galen VRChat.png
VRChat; 2018 screenshot courtesy @CompatibleVRC via Road to VR
Galen Blaxxun.png
Blaxxun BMW Financial Services scene; 1999 screenshot archived at x79.net

Linden Lab’s strategy

What is LL’s strategy for Sansar? Huge question with many small answers. For starters, they dismissed the idea of simply updating SL to support VR equipment like the Oculus Rift and HTC Vive. Although I doubt it, LL claims that SL is just not speedy enough to support them. I believe LL really wanted to get a fresh start on a true next-gen project, leaving behind many of the debatably poor choices they made that may have doomed SL to hit its peak of growth around 2008. LL has often said that “Sansar is not SL 2.0”. I think they mean it, at least in the sense that they really did start fresh. And SL and Sansar will probably never be compatible with one another, technically.

I do believe SL will live on indefinitely, alongside Sansar. But I predict that SL usage will wane as Sansar ramps up its technology and its base of users. Assuming LL doesn’t give up on Sansar, I would predict that Sansar’s basic stats of regular usage will surpass those of SL before the end of 2020. And SL’s numbers will decline; slowly at first, then snowballing faster, and then smooth out into a long tail as LL retires most of its sim servers. This is essentially what seems to have happened to ActiveWorlds (yep; it’s not technically dead yet!) In the meantime, SL is going solid for now and LL is investing in improving its technology. But that’s where LL makes its money, too. For now. And it probably will be a substantial revenue source for quite a few years hence. I can’t prove this; it’s my speculation.

Next, LL made a critical decision to roll their own proprietary video game engine. Many modern 3D games and VR systems rely on the excellent Unity and Unreal engines. This gives them a huge head start in features and makes some people question LL’s sanity for doing it the hard way. But I submit that Sansar’s “experiences” have a certain polish that is a direct result of the hard work they’ve already done, while VR platforms that rely on Unity seem to have a similar flavor with just a little less polish. LL’s choice to use the long-popular Havok instead of rolling their own physics engine has been one of the most significant sources of system instability in Sansar’s early days. Just throwing a ball might have crashed a scene a couple months ago. But it illustrates how using preexisting tech isn’t always a panacea.

LL’s strategy is also to make it as simple as possible to create content in Sansar. Not by using the old prim-building paradigm of SL, though. With so many mesh creation programs available, including the Blender freeware, LL decided that Sansar’s own scene editor would be more about placing assets like 3D mesh objects and audio files created outside Sansar and configuring some of their properties. Contrast this with Unity’s rich building environment, which seems to come with a much steeper learning curve.

Galen Unity.png
Unity Editor; Screenshot courtesy Janardana via Giant Bomb

In keeping with the ease-of-creation goal, Sansar has a centralized model, just like SL. Linden Lab intends to host the servers. High Fidelity, by contrast, is betting on an open source model, with content creators hosting their own servers. LL is betting that this will create added hurdles to adoption of HiFi that Sansar won’t face; and not just because of the ramp-up challenges for users.

I believe that Sansar’s real secret weapon lies in its nascent economy. I have long believed that SL won the virtual worlds game primarily by providing the first practical combination of permissioning and micropayments. This one-two punch enabled content creators to have enough control to be able to meaningfully own and resell their creations. The result is a market with hundreds of thousands of products for sale, a vibrant virtual real estate market, and thousands of people working for pay in SL. But I also believe that one reason SL’s economy hit a plateau nearly a decade ago is that its object permissioning system never continued to evolve. Most products sold in SL are still created, sold, and marketed by single individuals and tight partnerships. In order to sell a component to be used by other content creators to make their own secondary products for sale, one must typically give “full permissions” to their creations. Full perms items are vulnerable to abuse because as soon as one person buys the product, they can easily distribute it in an unlimited way at no cost. It can be very difficult to find and punish people who abuse full-perms items, so most creators avoid selling them. It’s hard to profit from doing so. This creates a barrier to wide-scale collaboration.

By contrast, LL has announced a forthcoming “supply chain economy” for Sansar. The key concept is that someone can buy a component, include it in their own product, and then sell it, with a portion of their proceeds going to the component’s creator. For example, you might design and sell a streetlight. I might then design a city street scene that incorporates your light. When I sell it to someone, a portion of my sale price will go to you, the streetlight designer. It will probably be exactly the same amount as if the customer had purchased your light from the store outright. And you won’t have to cede unlimited rights to your light for me to build with it. I am convinced that the natural result of this will be an explosion of creativity and a heavy flow of revenue among creators. Many items for sale will have hundreds or even thousands of creators behind them, all earning their cuts.

In the name of the supply-chain project, LL has held up a lot of expected features, much to the chagrin of many creators in Sansar. But I’m convinced this will be the uber feature that distinguishes Sansar more than any other. Not to most consumers; they won’t know or care to whom their payments are disbursed. It is creators who will love this feature. Instead of having to be super creators who can design all parts of a product from scratch, they’ll now be able to collaborate with more creators than they can count without even talking to them. As a scripter, this has great appeal for me. My sense is that in SL, most of the really good scripters are all-in-one programming and entrepreneurial virtuosos or are tied as employees to single designers doing custom work for them. I think few scripters can make money selling components of systems in SL. Supply chain will also open up the possibility of retailers working together with 3D designers so that designers no longer have to also be sales and marketing experts. In short, specialization will increase dramatically.

By stark contrast, HiFi’s open source model presents a gaping hole for intellectual property (IP) to pour right through. HiFi has recently introduced a convoluted solution to this problem in the form of a blockchain-backed IP registry. I honestly don’t see how this is genuinely going to stop people from stealing content left and right. And the fear of this prospect has scared more than a few content creators away from HiFi. So HiFi is trying to patch the IP rights hole in its market and VRChat doesn’t have a market. Meanwhile, Sansar has had a market from the start, actively polices IP violations, and is about to introduce its supply chain economy. I think this speaks volumes about LL’s strategy for Sansar. Note that competitor Sinespace has its own market, too.

Galen Sansar Store.png
Sansar Store

LL’s strategy for Sansar also involves a drastic reduction in what might be called “land fees”, compared to SL. A full region in SL can cost US $200$300 per month. By contrast, Sansar residents can get 3 full experiences – the rough equivalent of 3 SL regions – for free and pay just $10 a month to get 2 more. This radical difference is achieved mainly by turning off scene servers when nobody is visiting them. At this moment, Sansar has 920 experiences listed in its Atlas. Yet only 4 of them are occupied, which means only 4 scene servers need to be running. In SL, all 23,311 regions must be running in servers all the time. Unless things have changed since 2011, one server can host 4 sims, so there must be over 5,800 servers constantly running, even though only a small fraction of them are occupied at one time. I predict that within a few years, Sansar will top 100,000 unique experiences, yet the vast majority of them will sit idle in low-cost hard drives while maybe only a few thousand experiences are loaded into memory because users are visiting them. LL will have to curtail the use of bots in scenes or charge extra fees to those who keep scenes loaded using bots. Otherwise they will see server costs soar. Perhaps they will create a mechanism for inviting bots to join scenes as they spin up instead of staying permanently resident.

Galen Sansar Subscription Plans
Sansar’s subscription plans

One part of LL’s strategy is to make money by taxing transactions, including sales of items (15%), conversion of Sansars (aka “Sansar dollars”) to US dollars (10%), and bank transfers ($3 – $15). These could really add up to a hefty portion of one’s sales. If, for example, you sold US $1,000 of merchandise, you could wind up with only $750 of it in the bank before you even get to pay your own nation’s income and VAT taxes. I suspect LL is going to face a lot of pressure from residents to lower the component rates. But for now, this stands to be a huge money maker for LL. Even residents who take advantage of free experiences will be paying LL each time they buy items in the store.

LL’s strategy has also been to “go deep before we go wide”, by which CEO Ebbe Altberg means they want to introduce a lot of what they consider essential features before they make Sansar available to a wider audience. Many people have requested the ability to run Sansar on MacOS, for example. And to use VR equipment beyond just the Vive and Oculus, like the newer Windows Mixed Reality headsets. And many have loudly begged to be able to run Sansar on older machines. LL’s strategy thus far has been to eschew these and other options, not because they never intend to, but because the staff they would devote to those adaptations would be taken away from “going deep” in features. Plus various technical limitations that make some of these options less feasible.

Somewhat controversially, LL has also chosen to keep Sansar monolingual (English) and primarily oriented to people without sight or hearing impairments, at least during beta. As with the above, I think this is a smart strategic move for now. But Sansar’s bent toward VR equipment and voice chat are going to present technical, public relations, and possibly legal challenges in the future. I actually believe that in the long run, most of these will be solved by technological improvements, like making VR headsets lighter and easier for people in wheelchairs or with motor or cognitive impairments to use and more exotic technologies like brain-computer interfaces. I predict that getting around in VR will soon be much easier for most with disabilities than getting around the real world. And innovations from Google and other companies in the realm of speech translation will break down language barriers.

Speaking of the expensive hardware you need right now to use Sansar in VR mode, I think LL’s strategy is to count on innovations over the next few years to keep bringing these costs down. I predict that by the end of 2019, good VR hardware will be available for under US $100, a Sansar-capable PC will be available for under $500, and all-in-one wireless devices capable of running Sansar will be available. I also predict that the era of bulky geekware won’t last long. By the end of 2020, we should have slim augmented reality (AR) glasses and clip-ons that even look stylish, but cost loads less and are easily pocketed as needed. Moreover, I think we’ll see more body motion capture using video cameras instead of wearable sensors, right down to our fingertips. In short, I’m convinced that today’s generation of HMDs are just clunky prototypes and that low-cost, unobtrusive VR and AR are just around the corner.

Galen Magic Leap.png
Magic Leap One AR prototype

So wait; is VR finally here to stay? Yes! Isn’t it just another fad that has come and gone in cycles since the fifth century? No. Is everyone going to have VR equipment in the next year or two? Of course not. Let’s move on.

Another part of LL’s strategy is to enable fashion designers early on. Sansar’s “Fashion release” in December introduced a tight integration with Marvelous Designer that lets Sansar leapfrog other social VR platforms by allowing users to buy individual shirts, dresses, and other clothing and even to use cloth simulation during the dressing process to adjust them for best fit. By contrast, the other platforms mostly enable users to upload entire avatars created using outside programs. Some allow rigid and semi-flexible objects like hats, watches, and rubbery hair to be attached to avatars. But as far as I know, only Sansar has this unique ability to customize the fit of clothing, including to avatars with added muscle or other bulges. Users in VRChat may have easy access to hundreds of imported (and often copyright-infringed) whole avatars, but soon Sansarians will be able to craft millions of unique avatars just by combining different attachments and garments and adjusting them to taste. Second Life’s own marketplace nets residents tens of millions of dollars annually, a large fraction of which goes to fashion designers. Sansar is already headed down that same road.

Galen Marvelous Designer.png
Outfit created with Marvelous Designer by Debi Baskerville

LL has made the strategic decision to focus on nailing down stability and performance early on. Second Life arguably suffered in its early days from over-enthusiastic engineers churning out features very fast without ever quite finishing them, fixing bugs in them, or any overarching themes. The strategy seemed to be: do all the things! It wasn’t until an apparent management change around 2007 that the focus changed to increasing platform stability and battling the “griefers” who were enabled by the poorly considered feature firehose. By contrast, Sansar’s features appear to be developed in their own sandboxes until the team feels they are stable enough and reflect LL’s strategic goals. Then they go through (not always perfect) QA testing before being delivered in roughly monthly updates.

Which brings up another key part of LL’s strategy: communication with users. As new features are released, residents have quite a few avenues to let LL know of bugs or feature requests, whether through its Zendesk-powered public forum, request submittal form, Sansar’s own Discord server, or even weekly in-world product meetups and daily community meetups. Residents regularly get to hang out with and talk with product managers, engineers, and even CEO, Ebbe Altberg. While we don’t always get to be as involved as we wish with LL’s typically secretive development process, we at least get opportunities to let them know what we want and what we think of what they deliver. Both often result directly in changes in LL’s development priorities.

One dubious aspect of LL’s strategy for Sansar is its seeming desire to control everything. In my opinion, a big part of why SL took off was its near anarchy from the start. Residents felt free to explore whatever they wanted with only the most basic restrictions in its terms of service. The dark side of that was an explosion of scandals early on, including child pornography, unregulated gambling, banking scams, and more. Over time, LL banned and regulated ever more behavior in response to threats of government intervention and a growing public image of SL as a haven for criminals and perverts. One of the most significant regulatory changes was requiring adult content to be labeled and segregated, partly in order to merge the old teen grid into the main one and perhaps mainly to push adult content into the back of SL’s closet, more hidden from public view. So it’s no wonder that LL has purposely chosen to ban adult content from Sansar, at least for now. This skittish concern about how Sansar will look to a wary public seems to pervade much of LL’s process at present. Many of the technologies in Sansar stop short of enabling residents to create and host content that might besmirch Sansar’s image. In-world scripts cannot communicate with the outside world. Creators cannot replace the default avatars with custom ones, partly because the default avs have modesty panels covering sex organs. Residents cannot easily give money to one another. The list of technical restrictions on risky behavior goes on. But I worry that this pattern will also hobble Sansar’s community growth if LL doesn’t loosen its iron grip fairly soon. Sansar’s residents have been almost wholly sympathetic with the “we’re still in beta” explanation so far, but that won’t last.

She Works Hard For The Money
Escort Oasis in Second Life 2017

The desire to control everything also dovetails with LL’s desire to make it as easy as possible to do most things in Sansar. One of the more common complaints people have is that the basic avatars all look about the same. Newcomers familiar with SL’s own detailed appearance settings are often shocked to discover that just about the only things they can adjust on their Sansar avatars are sizes of face features and skin tones. There are distinct benefits to these limits. For example, experience creators can design realistic indoor spaces and doorways with the knowledge that there are not overly tall avatars walking around just yet that won’t fit. Configuring and outfitting avatars is very easy for now. And yet in spite of having one of the most technically and visually sophisticated basic avatars around, newcomers to Sansar often complain that the avs all look cookie cutter and even ugly. This is because in Second Life, people are used to going through av styling processes so tortuous as to make real world fashion seem easy by comparison. But they do it because appearance seems to matter more than ease of use. This trade off is bound to smack Sansar just the same. In the process of trying to prevent the flood of quirky solutions to style problems, LL’s ease-of-use initiatives may actually backfire.

One curious aspect of LL’s strategy thus far is its very modest focus on interactivity in early development. Newcomers to Sansar often get the impression that its experiences are like beautiful museums full of static content. Although the basic scripting engine and API have been in place from the start, that API gives scripters only limited abilities to detect and manipulate things. Scripters familiar with SL’s proprietary LSL language and API are often surprised and appalled at how little of the feature set they are used to is available in Sansar. However, I’m convinced that this was a strategic choice on LL’s part. They wanted to cover more basic capabilities like creating an excellent rendering engine and getting fashion designers going first. As a scripter, I’m seeing and welcoming a sudden shift in emphasis toward interactivity. And although my focus has been almost exclusively on scripting, I actually think LL’s strategic choice to largely put it off until now has been a smart move and not a big oversight.

Similarly, LL has made a strategic choice to avoid heavily publicizing Sansar for now. Until the end of 2017 their focus had been on getting the basic plumbing of Sansar’s platform in place. Ebbe has repeatedly stated that in 2018 they are focusing more on adding lots of features on top of that basic platform. Many of those features are oriented toward another key strategic goal for 2018: increasing retention. New residents can now more easily find other people in-world, for example. Resident-driven projects like HoverDerby and Atlas Hopping are creating some of Sansar’s first persistent communities. And LL is just starting to draw larger entities like Warner Brothers and Intel into Sansar via tie-in experiences. This process worked before to help popularize SL, like with the famous CSI: NY tie-in. The stunning Aech’s Garage experience is drawn almost directly from the forthcoming Ready Player One movie. Sansar may not have much public visibility just yet, but LL’s ability to connect with larger corporations will no doubt lead to some spectacular efforts to reach big audiences soon. But they are holding back for now while engineers and residents work toward making it worth people’s time not only to visit Sansar, but also to stay here.

Galen Aech's Garage.png
Aech’s Garage scene by Sansar Studios; Screenshot courtesy Inara Pey

All in all, I’m convinced that Linden Lab has a very clear, detailed strategy guiding how Sansar is developing now and for the next few years. Second Life’s early years seemed marked by a potent vision but no real business strategy. It toppled all its competitors in spite of that. I think this time around, its strategy will give Sansar the best chance of winning the social VR game.

Being a pioneer

So what’s it like to be a creator in Sansar right now? The simple answer is: fun and frustrating. For starters, many of us look with envy at what is possible in some of the other social VR platforms. Those that are built on Unity, like VRChat, immediately benefit from a rich toolset. Users of Sinespace can rove around vast landscapes thanks to innovative LOD. Avatars can actually sit down in HiFi. And they can play cool video games in Rec Room. Upwards of 200 people can be in the same region in Sinespace, versus Sansar’s 30-ish limit.

Galen Rec Room.png
Shooting game in Rec Room; Snapshot courtesy James Bricknell via VRHeads

That said, most of us are pretty excited about creating content in Sansar. The most striking thing for most of us, as for outsiders, is still the polished beauty of many experiences and products in its store. LL has even given away tens of thousands of dollars to winners of two rounds of contests for creators, spurring construction of some amazing content early on.

For me as a scripter, my first interest was in bringing interactivity into wider use in Sansar, especially by reaching out to non-scripters. Within a few days of arriving I created the first operational door opener. Soon after I created several generations of Reflex scripts that let creators add some basic interactivity to their scenes, like responding to a user pressing a certain keyboard key or stepping in some area by playing a sound or making a book fall off a shelf, for examples. I also created Clockworks as a way to add complex, script-driven animations. Most of the scripts I’ve made and consulting I’ve done so far have been free.

I have also made a little money here and there. I do sell some of my more complicated scripts (e.g., US $20 for Clockworks). I have taken small commissions (e.g., $50). And some Sansarians have generously donated money in thanks for my free help. I probably have made around $300 thus far, or about $43 for each month I’ve been here. My strategy thus far has been to invest my time and efforts into Sansar without expecting any serious compensation. I have focused largely on fostering interactivity and creating technology demonstrations. Like many Sansarians I have been predicting that there will be a rash of economic investment by companies and individuals willing to pay their way into the gold rush soon.

I believe that the financial success we creators may find in the future will come more from collaborations than from individual efforts. To that end I’ve chosen to specialize mainly in scripting and not to try to develop all the talents needed to foster full ventures. For example, during the Halloween contest I partnered separately with Jasmine and Alfy to create two different experiences: Miner Difficulties and Subway Nightmare. More recently I’ve worked with Jasmine and Draxtor to start up Sansar’s first team sport, HoverDerby.

Galen HoverDerby.png
HoverDerby experience

One area hotly debated among residents and creators is over how the avatar should look and function. LL’s answer thus far has been to create and control this to a large degree. Their main rationale has been to make it as easy for residents to configure as possible while still allowing vendors to create custom clothing and attachments like helmets and ray guns. The Lookbook feature acts like the scene editor in taking the user out of the scene in order to work on their av’s design and then “baking” the resulting settings into a full av before returning to the scene. In the Lookbook users can edit skin tone, some settings on the face, a hairstyle, and some other basic aspects, but this is severely limited for now. Beyond the head, one cannot change any aspect of the body’s shape other than to choose from male or female genders. This has led some, including me, to ask LL for the option to replace the entire mesh that makes up the visible body shape to allow for endless variety in the belief that no matter how many custom body settings LL adds to the Lookbook, it won’t be enough. Others have argued against this for fear of creating a dizzying number of clothing items and attachments to deal with all the various body shapes that will emerge. Some users have already created test attachments that fit like sleeves over the current shape and have large muscles and even nonhuman features. They show that MD-created clothing can actually resize nicely to fit alternative shapes, so I think replacement shapes will be inevitable. But I also sincerely hope that LL will continue to add more “sliders” for adjusting face and body settings. That way shape designers will be able to create basic shapes that allow for infinite variation by their customers. This approach will also make it practical to create a wide variety of nonhuman avatars that are simply not possible right now. Failing to allow body shape replacements will mean that avs have that cookie-cutter feel indefinitely, and that users will resort to awful alternatives to torture their avs into looking different.

One of the biggest challenges in Second Life is the massive cost of avatar rendering. SL has always had limits on how many primitive shapes (“prims”) like cubes and spheres can exist in a region, thus creating an incentive for designers to create “low prim” in order to keep regions fast to render. But there isn’t a limit on the number of prims and total size of images (“textures”) on an avatar. As a result, single avatars often consume more graphics resources on a user’s computer than the scenes they visit. Sansar introduced limits straight away. There is a limit to how many “trigons” (one-sided triangles that make up a mesh) an attachment can have and there are a limited number of attachment points (5) and clothing slots (5). Some of us have argued for a “trigon budget” as an alternative. There would no longer be limits to how many things an avatar can wear. Instead each resident would have a total number of trigons their avs can be made up of. One hazard of this is that residents would have to pay attention to their trigon budgets when purchasing clothing and configuring avs. But then, mixing and matching already requires users to be cognizant of what attachment points they have available. And designers now have the incentive to gang together multiple items in order to reduce slot usage. This small debate illustrates the complex balancing act LL must make between maximizing flexibility and minimizing confusion and frustration on the part of new users.

Sometimes there are setbacks. Recently LL announced that they would phase out the terrain editor, a tool introduced last September for sculpting pieces of ground when editing a scene. LL indicated that the editor had been a significant source of client crashes and that the terrain objects themselves caused significant performance problems. Thus the plan to phase them out. We all anticipate some future replacement, but LL has yet to tell us what it might be like or when we might see it. This is the biggest recall of a feature in Sansar’s history thus far. Yet I consider this a sign of courage on LL’s part to do painful things sometimes in order to create a better platform. Few of the features added to SL were ever removed, in keeping with a never-look-back attitude. I’m hopeful that LL will continue to take this more balanced approach with Sansar. The initial reaction to the stunning terrain editor announcement was surprisingly not negative. People were disappointed but overall sympathetic to LL’s reason for the change, mainly because they brought us all together to discuss it openly.

Probably the most frustrating thing for creators right now is how aware we are of Sansar’s limited feature set. Just to name a few gaping holes, creators are anxiously awaiting custom avatar animations, the ability to sit on furniture and vehicles, particle systems, simulation of river and ocean water, improved basic movement and  inverse kinematics, better camera support for custom machinima, mouse-click detection (talk about basic), script-mediated monetary transactions, vehicle simulations, and HTTP requests to outside web servers. Many curious potential creators have visited Sansar and decided to wait it out for now because they perceived a feature poverty by comparison to SL and some of the other social VR worlds. Being in the trenches from day to day, it can seem like major features are very slow in coming, but my own sense as a professional programmer is that this is progressing rapidly. It’s probably going to be a few years before we see all the above features and many more we anticipate.

One pervasive feeling about Sansar right now is that it is full of creators. Visit VRChat and some other social VR platforms and you’ll find that most people are end consumers who are not there to create content. And Sansar will eventually get there, too. But for now, if you meet someone in Sansar, odds are they are already working on at least one project here. And while there is always an undertone of competition among these often ambitious creators, there is also an overwhelming sense of camaraderie. Creators share a lot of what they learn and help newcomers get up to speed fairly quickly. That positive spirit keeps Sansar fun.

All in all, I’ve been excited about Sansar from the start. And it’s getting better by the day. I continue to spend almost all my time creating cool new things with great people. And although it is not a foregone conclusion that Sansar will “win” the social VR game, I think Sansar is off on the right foot. I believe it’s the right place for me to invest my time and treasure in. And I see a rising number of people who believe likewise.

Galen The Maze.png
The_Maze experience by KungFury
Advertisements

5 thoughts on “Guest Editorial by Galen: Sansar, A View from the Trenches”

  1. Wow great early history of Sansar with some excellent commentary on where they are probably going. I often have a hard time getting my head around Sansar and the LL strategy I think mostly because I didn’t have the SL background that most others in Sansar have. Being used to developing in Game Engines I still get confused by what Sansar is from a indie game developer perspective. It is way to closed to be a game development platform like Unity or Unreal and feels much more like a MMORPG that has a nice mod development kit associated with it. That being said I like Galen have cast my lot with Sansar and am encouraged to some extent to what is being developed. Although, I will always be pushing for a more open API, plugin architecture and other ways to allow more experienced developers a tool for developing their “games” inside of Sansar. Once again an superb, well thought piece Galen. Thanks for posting Ryan.

    Liked by 1 person

    1. Excellent op-ed Galen. I would add ‘HUD’ capabilities to the list of desired Sansar features. Also, one important challenge to LL is they are developing a platform that works both for desktop and VR (goggle) mode. This is an exceptionally difficult technical feat. VR mode requires something like 8X the data stream that desktop requires for the same scene. There is a mix of desktop and VR people in the same scene at the same time. The goal for LL is to make each mode top quality. So far, they are close to accomplishing that.

      Like

Leave a Reply

Fill in your details below or click an icon to log in:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out /  Change )

Google+ photo

You are commenting using your Google+ account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )

Connecting to %s

This site uses Akismet to reduce spam. Learn how your comment data is processed.