SL12B: Ebbe Altberg on Second Life & Sansar – transcript and video

Ebbe and Saffia get ready for the discussion (Jo was off-line at this point)
Ebbe and Saffia get ready for the discussion (Jo was off-line at this point)

On Friday, June 26th, Linden Lab CEO Ebbe Altberg faced questions from Saffia Widdershins, Jo Yardley and the audience in the last of the Meet the Lindens series sponsored by Prim Perfect.

The session lasted just over the hour, after initial teething problems in getting everything working, in which he addressed a wide range of question on both Second Life and Sansar, and offered-up some information of his own.

The following is a transcript of the event, based on a video by Chakat Northspring, which is embedded at the end of this article. My thanks, as always, to North. The official video will be available in due course via the Prim Perfect website.

Are there any things that have changed since you’ve been in charge that you are especially proud about?

[0:01:04] There’s a number of things. I think primarily it’s the relationship between the Lab and residents., I think is much better today. It was a little bit tense, I would say, the relationship, when I came on board; and it seems to be much more casual and fun and collaborative. And I think many of you have seen  more of us than you had for a while previously.

I’m also proud of the quality of the product. The performance, we’ve made lots of strides, the quality; [group] chat had lots of issues. So we’ve made a lot of progress on just making it a more stable product, a better performing product.

And I’m also proud of the focus we have at the Lab. When I came aboard, there were quite a few projects sprinkled about, and today I feel we have what I call four very strong focus areas with really good drive across each of them. So it feels like a healthier environment, not just between us and residents, but also amongst us Lindens, and a better quality product, and a better focused organisation. So I’m pleased with the progress.

Can you tell us what those four areas of concentration are?

[0:02:56] Obviously, continued to improve Second Life is something that we do. And you guys often hear from Oz and Danger on our progress there. We have Blocksworld; a fairly small team working on this neat little app for a younger audience to be able to build virtual experiences on the iPad. We have obviously a very huge investment in Project Sansar, our next generation experience platform, and we’ve worked really hard for over a year now, and we’ll start to get some external customers on-board in just a month or so. Just a few, but it’s great progress.

And the forth one, I would say, is compliance; making sure we run a tight ship when it comes to the linden dollar and who can cash out, and just running a tight ship when it comes to compliance … whether that’s fraud controls, identity controls – a number of things we need to do to make sure we and banks are comfortable with the business that takes place in Second Life.

So those are the four areas of focus.

The last thing you said about cashing out. You’ve said before that you want to speed that up. [do you have] any idea when this is going to happen, or how is the progress to that?

[0:04:39] The progress is good. We’ve now managed to automate a lot of things so that we can see what percentage we would be willing to pay out rapidly in an automated fashion. We’re still tuning the rules as to who we would trust to pay out; to make sure that we’re 100% certain that people who would be paid out should be paid out. And we continue to increase that percentage by continuing to dial the rules and make tweaks,

When we get to a significant enough percentage, then we should be able to start automating the process as well. not just the data saying we could, but we actually will pay out. I don’t have a date for exactly when we can turn that on; but it’s our goal for the vast majority of people that we have a trusted relationship with, to be able to pay within 24 hours or something.

I mean, there’s obviously external processes a well, that we’re not 100% in control of; but on our end, when someone clicks the button that says, “I want to be paid out”, we hope to, within 24 hours, to be able to automatically say, “start the process”, and then like I say, there’s actually some external dependencies for how quickly that actually takes place.

I think it’s interesting that you’ve given on compliance an equal status with the other three projects.

[0:06:14] Well, I wouldn’t say they’re all equal; but it’s a focus area, and it’s very important for us to run a trusted, large-scale business, to make sure that fraud cannot take place. For the sake of us as a business, as well as for the sake of you not having bad things go on. So it is an important aspect.

Danger Linden was actually saying that it’s one of Linden Lab’s advantages coming into Sansar, because all the new virtual worlds that are planning to come on-line, and you’ve got to get this right, and you have a head start.

Ebbe-6_001[0:06:59] Yeah, we are pretty much alone in having had a virtual currency with a floating exchange with cash-out and all these capabilities. There’s no-one else like it. so yes, we have a pretty significant leg-up compared to others if your intent is to have a virtual economy as we do. I mean, there are other business models that one could apply, but the way we are doing it, there’s no-one else really doing it as well as we do. And so that’s something we’ll certainly leverage; both operational experience and [the] technology, as we move forward with Sansar.

Second Life still gets some negative feedback in the media, although it seems like it’s become a lot less recently … but which kind of negative feedback do you pick-up [on], both in the media and form people in second Life themselves, that annoys you the most.

[0:08:15] Well, it’s not that much that annoys me … I’ve only had the opportunity to hear negativity for about a year … but I hear very little of it. whomever I talk to, it’s mostly … surprise that it’s still around, or more neutral. It’s very rarely that I’ll run into people that start off with the negative. So that’s a very small percentage of the population. Usually the negative people tend to be quite loud, but it’s not something I stress about.

I guess my biggest annoyance is people intolerance for various types of content. and when you look at the content in the real world, and people’s tolerance for that content in the real world. Then suddenly, when it’s in a virtual space, then it’s, “Oh my God!” Then there’s like a different level of acceptance for all kinds of content for some reason.

And that annoys me. So whatever the subject matter is, I can always draw a parallel to how it’s always “so much worse”, or it has just as much interesting stuff going on in the real world as in Second Life, whether it’s art, whether it’s sex, whether it’s whatever it is, all of this stuff is all around us in the real world, so why would it not be completely reasonable and acceptable to also have it in a virtual world. That’s maybe the most annoying part; when people don’t get that.

I know that your family have come into Second Life as well, and you actually have a family home here in Second Life and have actually had that for some time before you became CEO. So presumably, they get Second Life as well. But when you talk to friends … when they’re new to it, how do you explain what your job is?

[0:10:31] Well, it sort-of depends a little bit on the context of whom I’m trying to explaining it to; and it also really depends on their experience with various things. Bit generic when I explain that we’re trying to create a three-dimensional canvas that users can chose how to fill it, and how to populate it with what type of experiences, and that we want to create as much freedom as possible to allow people to create as much stuff as they can imagine.

So, kind-of suggesting that in something like Second Life, you can be whatever you want and do whatever you want and create whatever you want, as long as it’s legal and as long as it’s somewhat appropriate for the rest of us.

But then you can go into the incredible breadth of things that are really already working so well in Second Life; whether it’s education, health, art, role-playing. There’s almost as much variety of hobbies and interests and creations and experiences in something like Second Life like people can enjoy in their real lives.

So yeah, sometimes it’s tricky because it’s so broad. It’s so many different things to so many different people. So usually in a conversation, you usually have to figure-out what is of interest to the other person and figure-out how to relate to them with subject matter that they can get into or understand. And that’s part of the challenge of trying to explain something that’s so broad, because it’s obviously easier to explain a product that is narrow in its application or focus. So it’s usually a bit of dialogue that usually makes it easier than just a simple statement; it’s hard to think of a simply statement that sort-of captures it all for everybody.

Have you had a chance to Look around SL12B yet?

[0:12:50] I was around, I think it was two days ago. I was probably in there about half an hour or so; so I didn’t get too deep into too many things. but I really enjoyed some techno music over in the corner somewhere for a while. saw some fascinating art and creations, all kinds of interesting music, but not really enough time to have a chance to see all of it.

Many Lab staffers have been exploring SL12B and enjoying themselves
Many Lab staffers have been exploring SL12B and enjoying themselves

So, what do you like doing most in Second Life, when you have free time to come into Second Life?

[0:13:57] Mostly travelling and socialising and just going to interesting places and seeing who I can bump into and start conversations and just see what people are making and doing. So for me it’s more travelling and meeting people; that’s what I find most interesting.

Is it usually undercover, or do you prefer to use your Ebbe avatar?

[0:14:29] I’m almost exclusively as myself. I have an Alt, but I rarely use it; I really don’t find a huge need to go undercover. In some contexts, maybe people would talk to me slightly differently, but there’s so many people who don’t necessarily release who I am. I think the vast majority of people inside, don’t even know who I am, so they just speak honestly with me anyway.

I just try to be honest with people when I talk to them so that even that “Oh, it’s the CEO” even wears off real quick, and then we just have a normal conversation like anyone else.

Is there anything exciting coming for Second Life? We’ve already heard from Danger that is not going to be improved any time soon … but is there any other exciting changes and improvement coming in the near future?

[0:15:39] I’m probably repeating what Danger has said, but we are working on a number of areas that are going to improve. There’s still some performance improvement being worked on; we’ll get media on a prim to be completely modernised [a move from webkit to the Chromium Embedded Framework], so that you have proper, current web rendering tech, so you can have HTML 5, WebGL to function on a prim.

Experience Keys are going to be rolled out soon, grid-wide [in terms of deployment], so we can see more people take advantage of those capabilities . and Patch’s team is also working on a neat next fun experience to share with you all; so that will come out sometime in July as well, I believe.

So those are some of the things that are in the somewhat in the near future.

Will last names come back to Second Life?

[0:16:53] We will probably not have last names any time soon. Every time we look at what we can achieve with the resources we have, there’s so many things, I believe, that would have more impact on a broader scale. I mean, there’s a lot of work that’s going to go into the whole acquisition funnel of how users get on-boarded. I think Pete talked a lot about this the other day; about our upgraded landing pages and advertising. Banner ads through [to] optimised landing pages [are] already seeing a big lift of conversion rates from getting those rolled-out.

Then we also want to make sure that the transition into Second Life becomes an easier one. And whether that’s making sure that SLurls get straightened out so that you can send those around and expect people to successfully show-up at the end of that SLurl, rather than get lost in some weird labyrinth of pages.

Experience Keys were added to the new user experience earlier in 2015 as part of the Lab's continuing examination of the on-boarding process, with the promise of more A/B testing and experiments with gateways, etc., will be forthcoming
Experience Keys were added to the new user experience earlier in 2015 as part of the Lab’s continuing examination of the on-boarding process, with the promise of more A/B testing and experiments with gateways, etc., will be forthcoming

We’ll also experiment with some gateways, and continue to do all sorts of A/B tests for what way can we allow you guys to ultimately bring people more easily into Second Life. That’s probably where we’re going to spend more of our energy than anywhere else, beyond what I mentioned earlier.

Do you agree with people in the media, although some a sceptical, that VR is about to get really, really big. that we are at the renaissance of VR and that soon, perhaps every house will have a VR headset, and Second Life, of course – and Sansar – will be able to use that?

[0:18:38] I’m a huge believer in VR, but it’s going to take time. But as we build Sansar, we obviously have to have the expectation that Sansar is going to be around for decades. I mean, look at Second Life; it’s 12 years in, and it’s still doing really, really well. And Second Life is going to be around for much, much, longer.

We have to invest in something that we think is going to be relevant 10 years from now, and longer. and in that time frame, three years, five years, ten years, absolutely. VR is going to make huge impact in how people interact with one another and the world around them. and I can hardly think of any particular business or use case that will not be impacted by VR. Whether it’s architecture, education, travel, health. All of them are going to find ways to do things in really interesting ways because of VR. So yes, we’re making big bet that Sansar is going to be a great way for people to create experiences that take advantage of VR in a really great way. Absolutely.

Second Life has lasted for 12 years; presumably, you’ve got to plan for technologies that haven’t even been created to access it yet … I guess there’s going to be all sorts of VR headsets and ways of interacting coming down the line. You’re presumably going to have to work a bit in the blind on that with Project Sansar?

[0:21:07} There are some things that are going to happen that we can’t predict; but there’s also a lot of things going on that we can predict fairly well.

It will also take a while for the VR market to sort-of shake out: who are going to be the winners and losers, and what types of devices, whether it’s the HMDs, or head-mounted displays like the Oculus or the Vive from Valve or the Morpheus from Sony, or whatever Google’s cooking up going beyond Cardboard. And some of those will make it, and some of those will not make it; some of them might become quite niche and some of them might become broadly applicable; some of them will be easy for us to add support for.

Hopefully, over time, things will settle down a little bit on what types of input devices and what types of devices most people prefer to use. So we have to sort-of pick our battles on what things we want to support and not to support. Input is a complicated one, because there are lots of ways to interact with the world when you have an HMD on.

Oculus touch is just one of the many input controllers emerging as a part of consumer-focused VR the Lab needs to keep in mind (images: Oculus VR)
Oculus Touch is just one of the many input controllers emerging as a part of consumer-focused VR the Lab needs to keep in mind (images: Oculus VR)

You have things from Leap Motion where you use your hands, to these new touch controllers coming out from Oculus, to the controllers coming out from Valve to the sticks that are coming out from Sony. Several of them have sort-of similar feature sets that you can have a sort-of generic technology that allows you to support more and more of those. But some of them require a different user interface, so that’s where it gets really complicated because we don’t want the application to have too many different user interfaces for too many different use cases, because it gets really cumbersome to support all of that.

So for now, we’re just focusing on Oculus, and we’re still a year away from consumers starting to buy this stuff. so it’ll be interesting to see what shakes out and what becomes popular, and therefore what things we should latch onto.

It’s just something that we have to make sure that we build Sansar in an intelligent way, in a modular way, so it’s easy to extend support for new technologies over time, or if some portion of Sansar is to be completely re-done, you can re-do that portion without having to re-do everything else. So it’s the right level of modularity and architecture, and we’ve spent a lot of time thinking about that stuff up front; which is why it’s taken quite a bit of time to get the start of Sansar out to the world, because it takes a lot of architecture and plumbing to make sure we build a good base that we can comfortably then invest decades on top of.

What kind of name system can we expect for Sansar?

[0:24:56] I wouldn’t say that’s completely finalised yet. We’ll probably have to do something similar, because we want people to be able to, at least so far, our goal is to allow people to be able to use their identity across Second Life and Sansar, so you can log-in with your same credentials on both sides.

So for starters in Sansar … you’ll be able to have a display name and you’ll have an e-mail and password , and we’ll start very simple. But we’ve also architected-in what we call “personas”, so that over time you can have multiple personas associated with your account and they choose which persona you want to use in any particular context. But that’s further down the road.

One thing that a lot of businesses, when they’re run by more than one person, would find very useful is the ability for two separate accounts to share one access, like a business manager account. Could you see that happening?

[0:26:26] Well, with personas, we imagine that you’ll be able to log-in once with your credentials [“master account”], and then have multiple personas underneath that account. Each of those personas could have its own specific avatar and account, so that you can transfer currency amongst them; you’d probably still only have one US $ account, a real money account, but as far as virtual currency, you could sort-of move money between these personas.

Ebbe enjoys exploring Second Life, preferring not to use an Alt. I've bumped into him a few times, including a visit to Petrovsky Flux
Ebbe enjoys exploring Second Life, preferring not to use an Alt. I’ve bumped into him a few times, including a visit to A Petrovsky Flux in mid-2014

And so you would log-in and chose, “I want to be this persona, because now I’m going into 1920s Berlin, and I want to be that character, dressed that way and whatever, and then you’d come out and go somewhere else, you could switch personas without having to log-out and log back in again on a different account.

So we’ve architected to support that, but it’s not something we’re going to have from day one, because it’s going to take quite a while before that becomes a key requirement for Sansar; well over a year from now before that becomes a critical aspect. That becomes more critical when you have a rich avatar system, where people can have a really strong, personalised identity associated with their avatar.

And that’s not something that’s necessarily going to be there from the beginning in Sansar. In the beginning, we’re focused more on how you can create a great environment, a great experience, and then later on we’ll start in earnest, probably around next year, building up how you can create a really strong avatar and persona.

Danger Linden also talked about every new user is going to get a certain amount of Land for free with the basic account. Any idea of how we have to imagine this? How much land it will be, and if the people who rent out land at the moment, the landlords, should be worried?

[0:28:34] Well, I guess a lot of different types of businesses could be worried; or they could just look for opportunities.

The business model will be different in Sansar. I’ve continually said I believe land is too expensive in Second Life, and that we need to bring land prices down, and subsequently for us, to make sure we make a reasonable amount of income for the Lab, we need to bring taxation on the GDP up. So we have to think about what taxation is there on transactions in the system; we have to mix that with different types of subscription models for different types of capabilities…

So, we’re constantly trying to think of how far we can push land costs down and make sure that we can make up sufficient revenue in other ways.

One thing that bothers us a lot is that the cost of land in Second Life is actually preventing a lot of great content from being created and kept around. And if you’re in the user-generated [content] space these days, putting a limit on how much content can be created is not a good idea.

How far do you think you can push it down?

[0:29:56] I don’t know. but I hope a lot. And it also depends on what capabilities you want to have on that land; what kind of concurrency and what kind of capabilities, and we’re still playing with a lot of those variables; and I’m sure no matter how much we think we have it figured-out, I’m sure we haven’t, and when reality sets-in, and we start working the product with real users, that we’re going to keep fiddling with it.

But we want, like Danger said, to make it easier for people to come on-board and start to be able to create stuff without any significant hurdle. Now, how much land, how persistent will that be – for example, something we’ll do in Sansar that we don’t do in Second Life is, that if there is no-one on [your] land, then we can just store it away; and when you come back, we can load it back up.

In Second Life, everything, whether someone is actively using that land or not, is occupying CPUs and GPUs, just taking-up server space. And so we can be more efficient with that. And so if someone wants to come in and tinker, and then they leave, then maybe a minute later, we just save that out on disk, and that CPU can then be freed-up for something. So that can help us scale a lot.

So maybe if you always want that land to always run “hot”, anyone who visits it gets instant access as opposed to having to wait for it to load, or something; well, maybe you have to pay a little bit more to have it run hot all the time.

We’re still futzing with this; but just in general, our goal is to have property tax to come way down and sales tax to come up and then figure-out where is the equilibrium, where is the right balance between the two, plus other opportunities to generate revenue as well. the cheaper we can make land, in my opinion, the better.

In reference to Second Life, could we get more land for Premium members, 1024 sq metres, rather than 512 sq m?

[0:32:312] We are exploring opportunities as well in Second Life for other types of land packages, or land products. and we are thinking about if there are possibilities for us to reduce prices of land. We’re also exploring whether it’s possible for us to introduce other types of land with other dimensions and prim limits, or whatever, that would be appealing to users who don’t want to fork-out for a full region today.

The Lab does consider the issue of land costs in SL and what might be done to ease matters, but the issue is fraught with problems (image: Alirium)
The Lab does examine the issue of land costs in SL and what might be done to ease matters, but the issue is fraught with problems (image: Alirium)

So we spend a lot of time discussing these things. Sometimes we get a little bit nervous about how many more users would we get; how much more activity would we get? If we reduced land price by X%, would we get that back, or would we just make that a pure donation to all of you; which I’m sure you would appreciate, but won’t necessarily help us at the same time!

We want to make sure we get a return on that investment of introducing additional products. We also have to think about if we introduce new land products, what would that do to other land products? What percentage of people who have full regions would go to these new products, and what impact would that have on our revenues.

It’s complicated, but trust me, we debate this all the time. And also, there’s different use cases and different users. Is there even more we can do for educationals, non-profits, so this is a continuous dialogue amongst us. And I hope we can figure-out something to test in a reasonable time frame to provide other land products to Second Life.

Have you thought about hiring economists to look at the potential economy and craft a working model?

[0:34:24] I know the Lab has had economists on staff or in advisory roles in the past. We have a fairly smart group of people from the CFO to business operations people to people like Patch, who have been doing this for a long time that we can probably figure-out most of this among ourselves, but if we feel the need, we would. But it’s a fascinating puzzle, that’s for sure.

Streaming was a wonderful service we had from SL Go, and it’s now being experience by Bright Canopy … Is there are chance that Linden Lab are looking into it?

[0:35:44] We are looking at it, and we’re talking directly to some of the suppliers of those capabilities. We obviously have a close relationship with Amazon, and they have a product in this area.

Unfortunately, running a server with high-powered GPUs comes at a high cost. And so to offer something like SL Go did – that obviously didn’t work-out for them, they had to sort-of bail-out and just get bought for patents, because it was tough to make a business model around it where users are willing to pay the costs that it takes for a company to not only cover the costs of all those servers, but also to then make some margin on top of that. So it’s a tough one.

I think that over the years, those cloud-based GPU services will come down in cost and they will reach a point by which it will be a completely reasonable way for us to distribute our product; I’m not sure if it is there today. If we wanted to cover our costs with that, we’d have to stick you with a pretty hefty bill, and I’m not sure enough of you would be willing to pay that to cover the costs for us; and then if we can’t even get enough users to buy it, then it would be a big investment for us for a handful of people to enjoy.

Proviing a streaming solution like SL Go can come at a price premium; while the Lab continues to evaluate this space, it's unlikely there will be any kind of offering from them in the near future
Proviing a streaming solution like SL Go can come at a price premium; while the Lab continues to evaluate this space, it’s unlikely there will be any kind of offering from them in the near future

But it’s clear that for a number of use cases, including mobile and lower-powered PCs, this is a great solution to get a well performing product. so it’s something that we’re going to continue to evaluate and continue to have dialogue. I think we’re going to see more and more of these capabilities being offered by Amazon, Microsoft and others. I know Nvidia’s working on stuff that some of those other guys probably collaborate with too.

Hopefully at some point [they’ll] make this of sufficient scale and cost that it makes sense for us to make a product on top of it. Today, the math doesn’t really work out … And you saw where [Onlive with SL Go] started with their pricing models; those pricing models started out to cover their costs, and then they realised that cost was not something you guys were willing to pay for, and they ultimately went to $10 for all you can eat per month, and it’s not clear to me that would have reached enough of an audience … So anyway, it’s something that we continue to evaluate, and certainly for Sansar we’re going to continue to look at it as well. It’s one way to deliver a mobile experience in a performant way; but we also have a lot to do to make sure that we can get a really performant client that can do a lot of this stuff. So I believe it’s in our future; just not today.

Will you be able to sell goods in Sansar without payment information on file? Are you looking at ways to protect against copybotting in Sansar?

[0:39:48] Will you be able to sell without payment information on file? We don’t know yet. What we want to do with Sansar … is we want to have a more sophisticated system; the more we know about you, the more we trust you, the more we let you do. So whether payment on file is a requirement to sell or not, I don’t know what decision we’ll come to. and even if we had made a decision today, we may find that is wrong two years, one year from now. So it’s hard to say.

As far as copybotting, ultimately there is no 100% copy-proof system. Just look at DRM with regards to music, to video games; with anything on the Internet, someone can always find a way to steal. So there is no 100% waterproof copy protection.

We will obviously try to make sure we do everything we can to make it a worthwhile investment for content creators to create their content and make a market where people can run successful businesses. Our core is the creator is king; our products are mainly targeted for creators to be able to create experiences that their users can enjoy; and we cannot have successful creators that can run large, successful businesses if there are all kinds of problems associated with that, including stealing and stuff like that.

So it’s going to be important; but can I guarantee that everybody’s stuff will never be stolen in Sansar, ever? No. Because it’s impossible. But I think we’ll do enough to make sure that it’s a very successful platform for creators to run businesses on top of.

Will it be possible to have sound files longer than 10 seconds in Second Life, and will there be any other media formats than the .mov format?

[0:42:42] We’re already making good progress on that in Sansar; there’s already so many things we can do completely differently, in modern ways, so streaming content, having more control over your audio, better CD sound, much better understanding of space and materials and how that affects sounds.

In VR, doing correct audio is extremely important; it’s part of the immersion. If you have someone talking from behind you and you can’t clearly tell it’s coming from behind you, that breaks the immersion. So making sure we get spatial audio right, and all kinds of other audio correct is absolutely essential to Sansar.

Can you say anything at all about the minimum technical requirements for running Project Sansar?

[0:44:49] I don’t think we know yet. I mean, obviously our goal is to enable it on as many machines as possible. How Sansar will ultimately perform, compared side-by-side with Second Life on the same hardware, I think it’s too early to tell yet. Our goal is not to require anyone to have to buy new hardware in order to use Sansar if they can successfully use Second Life today, but I can’t guarantee that’s going to be the case.

I mean, some of the things we’re doing will make Sansar way more performant than Second Life, just in how we’re doing the technology and the pre-baking of content and the stuff we can do. But it’s still hard for us to say, in order to get all the capabilities we want, that the same graphics card you have today will be an ideal solution for Sansar.

system spec
If you want to use VR devices with Sansar, you’ll need to keep an eye on the system specifications recommended by the VR system manufacturers (image: Oculus VR)

Now, if you want to use VR [with Sansar], then there’s performance requirements we’re also not fully in control over. I mean, Oculus has said in their specs what they deem to be the minimal spec; so that’s maybe a level that’s beyond most of us today, since their graphics card requirement is a really new one, and so most of us are probably not there yet, unless you already own a Titan or something!

But we’re not going to have a requirement that you need to have that spec to run Sansar on a PC. So the goal is not to require new hardware, but I can’t guarantee that will be true for everybody.

OpenGL or DirectX for Sansar?

[0:46:04] Ultimately, both.

Will inventory be free on Sansar?

[0:46:17] That to me would be one of the ways to reduce content creation, and anything that gets away from content creation is not in our interest; and storage is not that expensive. so charging people for storage, those days have come and gone.

Will Second Life and Sansar become more interesting to education, for example, by allowing them to build a sim where people can’t get out or can’t get in, so you’re quite safe … without someone suddenly teleporting in, or having half the class suddenly teleporting out to an X-rated sim?

[Second Life already has strict access controls which can be employed by education institutions to control inward / outward movement of avatars and students].

[0:47:11] Will Sansar be a more appealing product for educators? Yes, I believe so; not just because of the quality, cost, a number of factors. But I don’t know if we’ve figured out exactly how much can a particular experience can constrain [people]. at some point, you almost get to a white label business, where each experience is a standalone thing. Maybe that will be offered some day; but maybe it comes at an extra price, because part of the value for us is to have a network where users can meet each other. So a school that gets cheap land to build this really great experience on, and there’s 20 students coming in and out, but those students never go out to the rest of Sansar or Second Life to do some shopping and become users of other things; that ends-up having less value for us.

But these students, once they’ve had their lessons in Sansar, may well return as permanent users.

[0:48:24] But that means that’s part of the process. you want to make them aware of what else they can do. So… maybe that comes at a price where we can do lock-down, so it almost feels like an exclusive, white label experience. At some point you can start getting into true white label, where they can have their own name space of users and identities and the whole shebang; but that’s kind of a different business model. We’d rather have one name space of users to make it easier for people to move about; and yes … we’re going to do things to make it easier to contain certain types of content to certain areas … Because we’ve said already that we want 13-year-olds and up to come on-board, and we don’t want those users to come in contact with content or users that are here for a completely different reason. so those are things we’ll sort-out with time.

It's a long list of third parties that are a part of the Sansar equation
It’s a long list of third parties that are a part of the Sansar equation

What physics engine are you using in Sansar? What path interpolation?

[0:49:55] We’re licensing tech and in negotiations with various parties about different things, so we don’t really reveal yet what some of the innards are. But we’re using a very modern physics engine, let’s put it that way!

Are you working with Nvidia?

[0:50:19] Well, we talk to Nvidia a lot – primarily about driver issues and stuff like that. In fact, I don’t know if Nvidia have a technology that we’ll directly use in Sansar. But we do talk to a lot of third parties about a lot of technologies. There’s no way we can build everything ourselves, so we constantly try to make the decision between build, buy or rent when it comes to technology.

It’s a long list of third parties that are a part of the equation, whether it’s physics or lighting or facial expression, and you can go on and on and on, with people we either have contracts with and we’re totally dependent upon at this point in time, or people we’re having on-going conversations with to evaluate ways we can bring their technology into Sansar.

It’s a year after the launch of Sansar. What’s your idea of Sansar being a success?

[0:51:28] Well, it depends on your definition of “launch”. We want to launch early, in just a month or so, with just a handful of hand-picked users. so that’s a launch for us, to start basically having an on-going dialogue with those users. We’re also starting to have dialogue with users who might be a little bit further out than that. And we also have to think about what use cases are relevant in Sansar, at what point in time.

Late this year / early next year, Sansar will be relevant for some use cases and be really cool for some use cases. And for other use cases, we’ll be completely not ready at all … So it’s not one of these things where it’s sudden born one day and has everything. I mean, it’s going to start out by having almost nothing, and then it becomes meaningful to someone.

Right now, we spend more time talking to architects than talking to others, because the ability to just walk through a space, and communicate in a space, that’s something we’re going to be quite advanced on, early-on. But then, like I said, a sophisticated avatar system, with high-customised [capabilities] and a marketplace recording and all that; that’s not ’til way later next year. So that means someone who has a dependency on avatar customisation, there’s no point in talking to those kinds of users for a long time.

That’s why, when you say “launch”, it’s quite fuzzy. We’ll be launching this year, but for other people, we will not be launching until next year, and for some people, maybe not even ’til the year after that.

But how would you consider Sansar to be a success?

[0:53:27] Well ultimately we won’t know that for a few years. It’s successful when it’s easier than anything else to create a virtual experience. and when I talk about a virtual experience today, your options to create a sophisticated virtual experience, if you don’t have something like Second Life, or Sansar is: hire an engineering team, license Unity or Unreal or something, and then start writing C++ code.

We want to eliminate those kinds of skill requirements in order to create a meaningful experience that people can share together. So when we can see that creators – and in the beginning, those creators will have to be more sophisticated – because like I said earlier, we’ll start with Maya as the authoring tool until we can start introducing our own authoring tools into Sansar – and then, over time, figure out ways to lower the bar for how people can enjoy creating content.

So it starts with something like Maya, and then we start creating in-world tools for layout, and support more third-party tools, and at some point we start to create in-world creation tools. And we have to do that and choose; it’s no point us trying to compete with Maya, Blender, 3D Max, Sketchup and yadda, yadda. If you want to use those, fine.

At some point we have to find the space where we can create tools that don’t necessarily compete with those, but add additional use cases where people can create content. and whether that’s in-world voxel tools or in-world other sort-of simple tools for creating content – that’s something we’ll be working on for the rest of our lives!

Will you be developing voice-to-text / text-to-speech in Sansar?

[0:55:57] I don’t know that we will develop it, but we will probably be able at some point to introduce it without too much headache, because there are probably a service offered by Microsoft or Google or someone, that makes it easy for us to send them text and back comes voice or send them voice and back comes text. So when there’s a service out there that has a reasonable costs associated with it, then it becomes easier for us to integrate. But I wouldn’t see us as being in the voice-to-text or text-to-speech business ourselves. so that would be something we’d have to license or partner or something like that.

Voice-to-text / text-to-speech may not be something the Lab develop themselves for Sansar, but it could potentially be integrated into the place via a third=party service
Voice-to-text / text-to-speech may not be something the Lab develop themselves for Sansar, but it could potentially be integrated into the place via a third=party service

So based on how good I sees these guys getting with doing things like this, it becomes, “when is the service and the cost appropriate for us to integrate”, and it doesn’t seem to me that it should be a really complicated thing to do.

How will the selection process work with alpha testers and beta testers for Sansar? Will you have an application available, or is the Lab just going to hand-pick or approach existing residents and content creators in Second Life?

[0:57:15] That will continue to morph over time. In the beginning we can just start with people we already know, that have the skill sets and also the patience and dedication to invest time in it; and it’s going to be frustrating. I mean, things are not going to work, and it’s going to be buggy, and then one day we say, “we’re sorry, we have to wipe everything and launch new servers and all your content had to be eradicated; we’re starting over.” We don’t want that being something a lot of alpha users have to suffer through.

We’ll start with some people that can dedicate themselves knowing all that pain that’ll come with it. and then when we see the reliability and the number of times to we have to break content compatibility and sort-of wipe content out, and it starts to stabilise; well then we can start to open up the door to more and more and more people.

And I want to get to the point as soon as possible where we don’t have to screen or something like that; but we also don’t want to have thousands of people coming in and spend five minutes and then go blog about how shitty it was. That doesn’t help anyone. so I’m hoping that later this year we can have a much broader set of users, but probably not ’til the first half of next year where we have sort-of full access for people to come in and monkey around.

How far the avatar customization would go in Sansar? Would there be only humanoid avatars natively, or some more sophisticated skeletons altering functions as well? Four-legged avatars are quite popular here in SL, for example. Would Sansar offer the same?

[0:59:30] Well, everything we’ve talked about so far has been to be much more flexible and scalable than Second Life has been with customisable skeletons and all of that. But we’re barely started. We already have avatars in Sansar that we use internally here that look very nice; but in order for you to figure out how to severely customise them and then also have a marketplace where people can buy and sell clothing and have that fit; it’s a really complicated system we have to build.

But yes, we do want it to support arbitrary legs and limbs and whatnot, but like I said, it’s not ’til next year [that] we really get going on that. and it’ll probably be over a year from now when we can sort-of say that’s at a place where it starts to be where we want it to be. So, we’ve got quite a bit of time to chat about that.

Recently in a post on you were quoted as saying that a reduction in land fees/tier for project Sansar would be leveraged by a tax on revenues (to store owners), yet on other occasions it has been stated that a new tax would apply to all transactions, like a sales tax, to leverage the lower land costs. Can you please clarify?

[1:02:01] There’s more to be figured-out. I mean, do you want to tax gifting? but we do somehow have to get a bigger piece of the GDP than we’re doing in Second Life. That’s where we have to make up a lot of the difference in us being able to reduce land costs significantly. The other things we’re going to do will be operationally more efficient so that Sansar is a cheaper technology to run at scale than Second Life, and we can pass some of those savings along.

Today, when we collect 5% on Marketplace and 0% on in-world transactions; that’s not going to be really sustainable if we really want to move land prices down. So that’s as much as I have for now.

Are there any “remote” employment opportunities with Linden Labs for those of us in the United States or Canada?

[1:03:38] We’re a fairly distributed organisation already. We have offices in San Francisco, Seattle, Boston, Virginia, and also quite a few people working from the “Moon Lab” at home. but having offices outside the US creates all kinds of additional complexities, so it’s not something we’re looking to do at this point.

What’s the reason for not wanting third-party viewers in Sansar?

[1:04:23] Well it’s not “not wanting” them. It’s that we wanted to reduce the complexity of what we’re trying to do, and open-source is an additional complexity; certainly early-on. Four years in, will we decided to go open-source [which is roughly what happened with Second Life] , I don’t know.

It’s not anti open-source or anti third-party viewers, but it does create a lot of additional complexity, and we’re trying to minimise those complexities early-on so that we can run faster, and ultimately, maybe focus more energy in thinking of extensibility of our viewer, versus fully fledged third-party viewers. Because back in the day, some people took advantage of that in all kinds of weird ways.

So from a security point of view, and a speed point of view, we just felt it was better to no go that way starting out. And three years from now, I have no idea what we’ll decide to do on something like that. If we can provide the right level of flexibility and extensibility to our viewers so that people can add third-party tools and capabilities that we don’t have time to develop, in some sort of plug-in system, or something like that; that might be more appealing to us than going full open-source. So it’s a decision for us now to get to like a “1.0” faster.

[1:06:20] We want to make sure we can make a very customisable user interface. If you look at the HUDs in Second Life, that’s pretty limited and pretty lame. making it possible to extend, so that when I walk into an experience, it’s not just the experience, it’s the user interface around it, that’s highly customised for this experience. Then you kind-of have to supply the flexibility to allow for third-party capabilities to start to seep in there somewhere. so it’s something we will consider.

A lot of work is going on now with the architecture and the plumbing of an extensible user interface system that not only is easier to extend for us, but is also easy to extend for the users, and then can also run across all the platforms; it’s a mega-challenge. but we’re on it.


At this point, the session wound down, with a suggestion – originating from Ebbe – that a way is found to hold similar events more regularly in the future, with a possible monthly Q&A session then being suggested, as well as finding other channels by which information could be exchanged.

10 thoughts on “SL12B: Ebbe Altberg on Second Life & Sansar – transcript and video

  1. That’s really interesting and I’m liking Ebbe Altberg a lot, he cares both of the social and technical sides, his words let me think Sansar it’s going to be well planned and LL is thinking ahead (so hopefully in future you won’t have to stick with an hammer new stuff for which your world wasn’t designed for, as it happened with SL, or at least you won’t hammer too much), and he seems to know SL pretty well.

    There is a couple of extra things I’d like to be able to do with a virtual world though. Let’s say I’m designing real cars and I’d like to show to my costumers how does it feels to sit inside my vehicle; so in my website there may be an embedded Sansar plugin, my costumers just come to my website, wear their own VR glasses and experience my project in VR. This would happen in the same way now someone watches an embedded video, and the costumer would interact naturally with the VR environment, without the need of learning complicated controls. This would have a lot less barrier than SL, and it would be as accessible as Youtube or Facebook. It would be interesting to architects too, but think if you are a tourist instead: you go to the hotel page, you start the embedded Sansar VR, to view the VR content made and uploaded by the hotel owner (or by the company that made the website), and you get a better idea how the hotel is and how its rooms are.
    Of course you can still just put web-links to Sansar as with SL, instead of making a plug-in or a web client; but clicking on a link to open VLC or another media player isn’t the same as watching it directly. It’s not as simple, immediate and accessible.
    So in this dream themed SL12B, I have got this little dream.


  2. I found this not only informational but enjoyable to read. The near future looks to be interesting. Thanks Inara..


Comments are closed.