It feels like just yesterday that Mitch Kapor proclaimed that the “frontier days” were over for Second Life. Now, maybe Mitch knew what Mark Kingdon had planned even before M himself – but what’s clear is that while the frontier days may have ended it took from July 2008 when he made the statement to today for the rest of us to notice.
Linden Lab is using what I was going to call a “Shock & Awe” strategy for announcing that Second Life as you know it has ended, but that’s soooo 2000s. Instead, they strapped your avatar in, started the ignition and, like it not, you’re now barreling down the autobahn with no exit ramp.
On this week’s Metanomics Esbee and Amanda Linden were generous enough to let me ask them my meta-questions about the new viewer and Second Life Shared Media and while I probably should have spent more time dissecting their choice of chiclets for chat, I was actually a lot more curious about the process that got them here: how do you ‘renovate’ a viewer that’s been around, with a few tweaks now and then, for almost a decade? How do you decide what goes and what stays?
The Cultural Implications of Change
And most intriguing of all, I thought – do you take into consideration the cultural implications of all of these feature changes and widgets, clickable Web pages and tattoo layers? Second Life is a world, after all, right? Since everyone’s talking about Cameron’s Avatar these days, it’s not like it’s some mystery that the natives have their own culture, values and systems – and that it’s pretty easy to screw it all up.
Amanda replied:
“It’s so interesting actually. I was an art history major way back in the day at Vassar, and I understand how much visual design and experience design can impact culture and vice versa. I think it’s a very symbiotic relationship. I think you’re right. There’s a new element within the viewer that is web based, that allows us to create new kinds of promotional and other kinds of advertising capabilities within the sidebar. That’s something we definitely thought about.
There are a whole host of other capabilities within the new viewer that are meant to make it very easy for users to get to the most used pieces of functionality very quickly. We wanted to surface those things up to the top immediately and viscerally so that people can get to what they need very quickly.
It’s been interesting. I mean we looked at, and Esbee hinted on this too, we looked at a lot of other things besides viewers. We looked at Skype. We looked at IM. We looked at Google. We looked at Yahoo. We looked at a whole host of other kinds of communication technologies. We looked at Twitter. I think that what we tried to do is make it easy, and, by making it easy, I think that, hopefully, it will enrich not only the experience with Shared Media and other capabilities, but will also make it easier for people to communicate. “
Now, in fairness, I put them on the spot a little – we were talking about usability and expanding the user base and I was basically asking if they had an anthropologist on staff.
But what’s intriguing about this is that it speaks to a conclusion that I think we can now arrive at with certainty:
To Linden Lab, while Second Life may be a world, it is not a culture. The more important culture in which Second Life participates is the broader one which encompasses our lives on-line. While there may be sub-cultures that find a place IN Second Life, the frontier days are over, and the sense of it still being “one world” no longer apply.
Design Thinking
Now the Lab will argue that this isn’t really true. They have a canned phrase: “The wonderful creativity of our Residents.” Listen for it – you’ll hear it over and over.
And yet, if the Residents are so creative, then how is it that a complete overhaul of the viewer excluded them until, hmmm, the last three weeks of a nearly 18-month development project?
I won’t argue that this is their platform and they can do what they like with it – and I applaud the new viewer, I jump with joy for Second Life Shared Media, but we’re now talking about a DESIGN THINKING company whose focus is on opening up Second Life to the wider Web (while closing, thankfully, some of the massive gaps IN the world, like the ability to randomly copy other people’s content).
I found it intriguing that Esbee spoke about the Lab as an engineering company. It almost felt like the kind of thing the interface and product development people talked about after work one day, huddled over beer (or chai tea or Starbucks, this being San Fran after all) with Tom Hale telling his team to be careful about the language they use:
“We may be REDESIGNING Second Life, we may be changing the interface, and overhauling the Web site, we may be adding Web-based content, and reconstructing the user experience, we may be applying design thinking to commerce, and creating new channels and on-boarding processes, and we may be creating new ecosystems of business value while we rethink how we monetize the brand – but what ever you do, always tell people we’re an ENGINEERING company!”
Thus, Esbee’s response to the methodology for developing the new viewer:
“So we actually decided as an engineering organization, at the beginning of this project, that we wanted to be able to iterate, as Amanda said, much more quickly than we’ve ever been able to before. We want to be able to move into a cadence where we’re more frequently releasing new features and more frequently updating the viewer. And we knew that using the type of design and development approach we’d used in the past that we really couldn’t do that, and so we adopted for those of you who are familiar with it an agile development methodology called SCRUM.”
Um, no – you may have a lot of engineers on staff, but you’re no longer an engineering organization. That’s like saying that Apple is an industrial design company: they may glue together bits and pieces but they’re a design thinking company first and foremost.
We Face the World
And much of the design thinking on the new viewer and Shared Media is now clearly part of a much deeper road map whose question is this: how do we plug Second Life in to social media, training, education and other on-line tools and applications? How do we set this up so that we actually CAN reach a billion users? What would it mean if we were to start wondering what it would take for Second Life to not just be a destination, but the starting point itself?
And so we start to see the unfolding of a strategy to make Second Life more Web like, to bring the Web IN, and to send content OUT into that great shifting tide of social media. And maybe the strategy ISN’T to tap into the poking throngs and Farmville, um, farmers on Facebook but to REPLACE Facebook itself. I mean, why not, right? If you’re going to try to monetize think BIG BANK, as Tom might say.
And so we have the Lab starting to link our actual and avatar identities (and my guess is our ability to use our real names will come before the end of the quarter), as Esbee pointed out on Metanomics:
That’s a wonderful question, and it’s actually a great one to hit on because it was an area of intense debate around the Lab when we started looking at a user profile. Second Life, different types of people are drawn to Second Life. Some of them don’t mind sharing real life information and others who prefer to keep that Second Life completely separate to their real life. And we respect and think that both of those approaches are great. For example, I’m someone who has always shared both my real life and my Second Life alt name, and that’s just the way that I’ve done things. But I have really good friends who keep their personal lives anonymous, and that’s great.
In the design of this particular panel and the way we approach user profiles, we wanted to give them the same weight because of the way that we’re going after new users. This viewer will become a viewer that not just our existing residents and the new users that we attempt to target, but also large areas of business and education, and those are users who will probably value being able to share more of their first life, if you will, or their Real World information. And so we wanted to give that a little bit more prominence.
We are doing a lot of work right now that I can’t talk about as much because I’m not involved in the project, but we’re talking a lot internally about identity and what that means in a Virtual World, and it’s actually a huge area of focus in the Lab right now. So the profiles as we see them today, we wanted to surface as much information as we could at once, without creating a lot of clutter, and it will hopefully hint to new options and approaches to identity in Virtual Worlds as we move forward.
And while the Lab respects that not everyone will want to share their real life identities, they certainly won’t be discouraging it when ads targeted to new users look like this:
Now….not ALL ads will look like that. I’ve seen Steampunk and Goth banner widgets which are clearly aimed at a more (optionally) anonymous crowd, but with this type of advertising, it’s yet another cultural change that may happen in subtle ways, yet happens nonetheless.
Open Sourcing Interface?
Now, on one topic I’m not sure what to say, but as I understand it the Lab has done what I long predicted they would do when they launched the new viewer: parts of what was once an open source client would become “closed”. And to the best of my understanding, that’s exactly what they’ve done: while the code behind the viewer interface is open, there are things about the interface itself which are not.
And I’m happy to be corrected on this, of course – but the code repository for Snowglobe represents the open source version of the client, and yet it doesn’t include all the “bits” that make up SL 2.0.
As the Lindens said in response to my question (emphasis added):
AMANDA: Exactly. So Snowglobe 2 basically has all of the bits that you need. It doesn’t look exactly the same, but it is essentially the same as Viewer 2. And it’s now available in Windows and Mac and not yet on Linux. I think those guys are still working on that.
DUSAN: So what we see now as the release of Open Source Snowglobe, well, I mean it’s all Open Source, but what we see now as Snowglobe’s Open Source code is Viewer 2.0’s Open Source code.
AMANDA: Correct. That’s right.
DOUG: As far as you know. Okay.
ESBEE LINDEN: And we are working right now on a plan for how we synchronize our branch that we developed Viewer 2 in, with the Snowglobe branch moving forward so that we can really more frequently interact with the Open Source community and make sure that we have a much closer dialogue with them as we proceed further in our development this year.
Maybe someone can tell me what “doesn’t exactly look the same” actually, well, means?
At the same time, the Lab launched Twitter OAuth into Second Life, which brings Tweety goodness to the Grid:
“What this means (as you may have read) is that you can now let people Tweet from within Second Life in a safe and secure way, without having to set up external Web servers, and without requiring Residents to re-enter credentials if they want to use Twitter from inworld. OAuth is an open-source protocol that provides secure API access to login credentials. This means you can give Web sites or inworld objects the ability to update your Twitter account without actually having to give out your username or password. The Twitter OAuth Library we created also allows for fine-grained control over which Second Life objects can send updates directly to Twitter.
Tweet away!
What’s Next
So if your head isn’t spinning already, the Lab isn’t done yet. Amanda confirmed, as T Linden (Tom Hale) has done on the forums: mesh imports are coming. But why stop there? As noted above, the Lab is also working on:
- A new orientation experience for new users
- A new system for linking your avatar and real identity
- Inventory overhauls
- Desktop sharing (maybe)
- Links to social networks and other Web content (via Avatars United somehow)
- Continued changes to the Linden and commerce in Second Life (for example, this week’s announcement that your avatar “wallet” is now connected to the Web – which opens up, well, some pretty stunning possibilities as Gwen has pointed out)
- A new ad campaign
And….well, and there’s tons more “ands”.
The frontier days are definitely over. As I predicted before, I think we’ll be at 200,000 concurrency before the year is out, and I’m pretty confident that it was a conservative prediction.
And as the new Residents come back or come in to have a look – the world they’ll be entering will be nothing like the one that was there before.
Oh God save us….never mind…too late
(
[...] We are doing a lot of work right now that I can’t talk about as much because I’m not inv… [...]
Thanks for the post Dusan. A few clarifications:
Dusan Wrote: Maybe someone can tell me what “doesn’t exactly look the same” actually, well, means?
>>>What it means is that snowglobe 2.0 is the source for the current 2.0 beta viewer now in wide release, but it doesn’t include some of the features from the Snowglobe 1.32 Snowglobe branch yet. The reason for this is that we wanted to ship the Snowglobe 2.0 open source bits at the same time as the public beta of Viewer 2.0. We expect that going forward “experimental” features will make an appearance in Snowglobe before they show up in the mainline viewer, and some features might never make it all all. But the point is that we’re not holding anything back from Snowglobe – it’s the viewer 2 source. More info here http://wiki.secondlife.com/wiki/Snowglobe
Dusan Wrote:
And yet, if the Residents are so creative, then how is it that a complete overhaul of the viewer excluded them until, hmmm, the last three weeks of a nearly 18-month development project?
>>> Lots of residents (both inside and outside the lab) were involved in the viewer redesign. We drew inspiration from results of the viewer redesign contest (see http://dusanwriter.com/?p=557), from research and user feedback studies, from SLViews sessions, from usability tests (both in paper and in software) and from a private beta program. What we didn’t do, was make a small step forward, then try to get every Residents’ feedback. Why did we do this? One was because we were focused on solving for new Residents, another was because design by committee results in incremental change, another was because the problem of designing a viewer requires holistic thinking to balance various constituencies and interaction models. The viewer is a chat tool, a voice tool, an authoring tool, a “browser” of sorts, an admin UI for land/groups, and a file management system.
Hope that clarifies things for you…
[...] To Linden Lab, while Second Life may be a world, it is not a culture. The more important culture in … [...]
Thanks for the clarification T.
My point about the design process wasn’t to criticize the methodology but to make the broader point that the Lab has shifted to a more design-focused approach to product development which doesn’t rely on things like the JIRA and employees deciding what they want to do in the morning as the primary drivers for how the business evolves.
As I’ve posted elsewhere, SL remains one of the most robust platforms anywhere (aside perhaps from the Web itself) for people to develop their own “stuff” upon – scripts, content, their own viewers, etc.
I’ll tell you that my own preference is AGAINST design by committee and I believe that most crowd-sourcing is either cheap labor in disguise as something communal, like a barn-raising. So don’t get me wrong and think that I’m criticizing the methodology.
I am saying however that the Lab has moved from a more fully open development culture to one that’s more centralized and design-focused.
Have a look at what Philip is doing – he’s giving away his new stuff before he even starts, posting code in bits and pieces as he builds it. Compare and contrast.
So, again – applause from my end, and hope that clarifies my own point somewhat.
^– What Brinda Said.
To be honest, what I have seen and read about Viewer 2 doesn’t make me even want to try it out myself. Seeing the reactions to search for example (this time not the usual 90 percent negative feedback but more like 99 percent), and on the other side the typical PR stunt “Look how great search in Viewer 2 is!” – ouch! Even for LL standards that is a new top.
T Linden writes as comment on your blog “The viewer is a chat tool, a voice tool, an authoring tool, a “browser” of sorts, an admin UI for land/groups, and a file management system.” I learned that usually things that try to do *everything* do *nothing* really well.
Personally I hope that there always will be 3rd party viewers that are totally different from SL Viewer 2. Hopefully with some of the positive new things, like Shared Media.
Ah, I wish to thank T Linden for the clarification, too. When I first read your post, Dusan, I got the impression that Snowglobe was not the codebase for SL 2.0, which sounded incredible, since it has exactly the same functionality, and a few nasty bugs corrected on top of that (and I still don’t understand why it’s faster than SL 2.0 Beta). So having two internal code branches that produce exactly the same experience made me go “wow”. It would be as unexpected as suddenly seeing Microsoft announcing Lindows (or some other brand name which has not been taken), which would be a Linux-based, open source version of Windows, but which looks exactly the same and runs exactly the same applications. It would be extremely unlikely, but I have to admit that Linden Lab quite often surprises me in a positive way, and it would not be impossible…
But clearly that was not the case (whew!…. my world stopped spinning
).
Well, as to the rest of your post… I think that best news for me is that all the predictions I’ve read at the end of 2009 will fall short — by a huge margin. Linden Lab is doing so much more in 2010. It’s amazing. I’m sure that we have to thank the new management for this new vision — a vision of pushing limits again. Oh, sure, I know, it’s not the same vision of the pre-2006 days, and oldtimers will always complain about how LL “lost their vision”. Those kinds of comments are absolutely unavoidable. They’re also — with due respect — immature and infantile. The world changes, all the time. Successful companies change with the world. Being stuck with old ideas and old concepts means a SL with 15,000 registered users and lots of happy geeks. We’ve outgrown that phase — it seems so distant now as to be just a dream. And, indeed, the past is nothing else but that: a dream. Memories. Things to talk about to our grandchildren, but nothing more than that.
For me personally, the past few years, which I call the “technical stability” era, have been quite painful. I obviously understand the need for stability, and it was nice that the former management was so keen on stability for the benefit of the vast majority of users. But until 2007 I was able to successfully push Second Life as being all about innovation — innovation in business models; innovation in financial/economic models; innovation in tapping crowdsourcing to push an “idea” ahead; and lots and lots of technical innovation, being dumped at us from all sides — open source viewer, WindLight, sculpties, flexiprims… *breathes deeply*. Oh yes, that certainly was part of the hype cycle: we were all excited by the possibilities, and dreamed about what would come next — LL was happy to provide us with new innovation, new ideas, new dreams every month, and we managed to sell those dreams to others.
Then the “stability era” settled in. Well, of course we have to have stability. And I understand it takes time. A lot of time. But from my perspective as a seller of the “Second Life concept” it was hard to explain to prospective clients that SL hadn’t really changed much in almost 3 years. Sure, it was more robust, more stable. Sure, gambling had (almost) gone away. Sure, griefing has remained low (the number of griefers has not increased as much as the number of the overall population). Sure, there is still innovation — at the resident level. But… overall, SL seemed to be going nowhere except to become “more stable”, step by step.
The same, ironically, happened to Twitter. We still remember the days when Twitter had the same number of users as Facebook. Twitter suffered badly from stability, so for at least two years, they did nothing else but work on stability — allowing Facebook to get ten times as more users, a whole ad network, and gazillions of applications able to squeeze money and information out of their users. It’s tough, but in this very harsh world, staying still and not doing anything worth blogging about (”after six months of performance testing, we managed to optimise three hundred MySQL queries, which will allow overall performance to increase by 0.3%”).
2010 is showing to be way more promising. There is a lot that is happening simultaneously on all areas. There is even a roadmap! And LL is sticking to it! Some things might fail, some might not — but trying and failing is far better than hiding under the hood working at the engine without anybody knowing about it. The key point is that SL is alive and kicking because LL is alive and kicking too — and kicking in all directions at the same time. That’s good. That’s important.
So in the next few months — and specially if meshes do get introduced shortly after the Big Anouncement about the whole SL Enterprise launch next month — there is no way all this is going to be ignored. I’m not surprised that all of a sudden we saw some players in the market giving up and dropping out of business: they feel that the giant is awakening again after almost three years of deep sleep, and nobody has a successful business model to show off — just cool technical gadgetery. At some point, even the most skeptical media will realise this, and some have suggested that 2010 might become the Second Hype year (after the 2006/7 rush). I might very well believe that, if LL plays their cards well — and so far, that’s what seems to be happening!
Personally, all it takes right now is for the financial crisis (or pseudo-crisis) to be abandoned by the media once and for all, and for organisations to finally make their decisions. I’m aware of insanely huge projects to be deployed on SL — at scales far, far beyond what was done in 2006/7. The era of let’s-set-up-a-virtual-presence-in-a-month-for-US$-1-million is gone. The next generation of projects in SL will take 2-5 years to complete, but they will be true gigantic in scope. They’re all waiting, very eagerly, for the hints that Linden Lab gives them about SL’s future, and for the economy to pick up again (at least the stock markets seem to be crawling again to the same levels as before the crisis). So, the timing is perfect for the ‘Lab.
I hope that in 2015 I’ll be blogging from within the shared desktop inside SL and write: “Facebook? It was a nice idea, but they totally failed to realise that the only thing people want to do on social websites is play games and get dates”.
@Daniel, yes and no. If T Linden was talking about Linux, or Mitch Kapor was talking about FireFox, using the exact same words, would you have the same reaction?
Don’t forget that the biggest drawback that you’ve quoted — Search inside SL 2.0 — is not even built-in into the viewer itself, but it’s just a webpage. A webpage that can be changed by LL’s many web designers at any moment — without even requiring a new release or a new download. That’s one of the most powerful things about SL 2.0: several areas (Search being probably the more well-developed right now) are just plain old HTML, served from LL’s servers. This gives LL a huge leverage on what is displayed there. Remember also that in 1-2 years, almost all HUDs and interactive devices in SL will be served from web servers too.
What does that mean? The “SL experience” will be much less a release after a release of new viewers, every time something needs to be tweaked; but the “viewer” will become more and more a framework for plugins. Some — like Search — might come from LL itself. Others will be provided by the community. They will be constantly updated without the need of further downloads. So something brand new might just get a simple prototype to show functionality — that’s all what the current Search tool is — but quickly finished for being ready for production. That’s just because it’s easy to find Web designers and programmers, but it’s insanely expensive to assemble teams of hard-core developers of 3D rendering engines.
Granted, I’d love to have the ability to replace the “official” search page on SL 2.0 by any other — just like we do on web browsers every day
@Gwyneth: I don’t know, maybe for such a complex thing as SL the best idea might even be to develop different viewers for different user types? All viewers would do the basic things the same, but one type might be targeted at the social crowd, another one at content creators? They could even all have the same functions, just organized differently, like “skins”.
@Daniel, yes, possibly you might be right. The slight disappointment I had with SL 2.0 is that it was sort of announced as “a dumbed-down version of SL” for the casual user who just logged in and had never been in a virtual world before. While I can see from the design elements that the notion of “familiarity” (it looks liek a browser; the chat is like Twitter or GTalk; the notifications are at the bottom right, just like in Facebook or MySpace or Netlog), I don’t think it’s “enough dumbed-downed”. Look at Blue Mars’ interface as a contrast (and yes, I know, we cannot compare the infinite complexity of SL with Blue Mars; but at least we can look at each UI and see how the approach is totally different).
Still, it’s true that maintaining 2 or 3 different viewers is a lot of effort, costs a lot of money, and takes a lot of resources…
I’m really relieved about T’s comments, and I think the Lab and the community in general supports Daniel’s point about different viewers for different needs. I won’t get into a discussion here about the registration of third-party viewers, but clearly while this will limit choice, I think it will do so (mostly) to the benefit of the community – there will still be options, although innovation may slow down because someone hacking together new functionality may feel a chill at the registration process. But overall, there will still be what I think of as ‘channel choices’.
What’s even more intriguing to me is all that ‘real estate’ on the side of the viewer, and Esbee skated around the question of whether it could be used for anything else: what I’m talking about are the tabs on the right side of the screen, and I can’t help wondering why one of those tabs couldn’t be a “HUD tab” or an “educator tab” and the tabs could be coded by residents and because they’re Web-based we could have a whole new layer of application development where we’re building apps for the viewer ITSELF without having to drill down into the source code.
Imagine being able to code a tab for event management and it has all kinds of functionality in it for changing a sim, bringing in external media feeds, sending out polls, or whatever – and the code for that tab, being pulled from a Web site with in-world data calls and scripts, could be updated from the Web itself, which then allows calls to the world, etc.
So the answer might end up with a spectrum from different viewers for different use cases, PLUS an application layer inside the viewer itself, based on allowing people to “code the tabs”.
(Which might, like Shared Media, include stuff driven or attached to enterprise or on-line applications, whether Sharepoint or Facebook).
So, I’m really relieved to hear T’s comments – there was something about the messaging around Snowglobe that left me totally confused and I’m glad for the clarity as well.
Have to agree with Daniel, trying to shoehorn everything into one interface is the main problem with the beta and every previous client. New users login expecting iPhoto and get presented with Photoshop, it’s crazy.
Seems to me, since we’re being driven (against my better judgment) into the web, that there’s a logical and natural division that could be drawn. Give simple stripped-down access into SL for social or casual use via a web-browser (achieved through plug-ins or clever future use of HTML5 if that’s possible) and keep the (free!) full-featured standalone client for the creatives and more immersed user.
Apart from a few glaring errors of judgment in the beta client I’m generally really positive about it – but this one-client-fits-all approach is never going to please everyone, let’s not waste any more time even trying to achieve it.
[...] So the profiles as we see them today, we wanted to surface as much information as we could at once, … [...]
There are a number of great comments here regarding the new 2.0 viewer/browser platform. The fact is; the new viewer is now a richer platform like SL itself. I believe the Lab intended it to be that way. They will not need to build everything. For example Dusan’s ideas about making the slide-out side-panel usable by outside developers with plugin apps from HUDs to whole applications is great. Gwyne certainly does “get it” when she says.
“What does that mean? The “SL experience” will be much less a release after a release of new viewers, every time something needs to be tweaked; but the “viewer” will become more and more a framework for plugins”
The major issue I still have is just how slow rezzing things has become and the 50 avatar limit for a sim will need to disappear if concurrency grows. The complaints about rez lag are not going away. SL is stable but slow and waiting to rez is the killer to fun and “shopping”. The current “user experience” is akin to playing football with your legs tied together. Hopefully new HTML vendors and asset stores outside LL will solve the problem.
There’s a lot of disappointing information in this post, though as always, I thank you for writing it, and for asking the hard questions.
I only have one minor comment for now–since Ordinal Malaprop invented TweetBox, many of us have BEEN using Twitter from Second Life. Why do they laud THAT as such an amazing technological advance?
I like the new viewer, I really do, it’s just more “Heavier” on my computer, and shared media is an amazing add on but brings my SL experience to a crawl. Running the old viewer and the same flash/video in a firefox window wouldn’t cripple my computer like that so I hope Linden is working on that issue.
Miss Orr: while I don’t think the new Twitter functionality qualifies as an amazing technological advance, it does differ from Miss Malaprop’s device in that it doesn’t require you to share your password with anybody except Twitter, or to set up a relay script on your own server. (I trusted Miss Malaprop with mine in the serene knowledge that she wouldn’t misuse it, but I believe she would be the first to grant that this was not an optimal practice.)
Gwyneth: Sorry, I don’t agree that it’s “immature and infantile” to take issue with the fact that Linden Lab has, after choosing one path at a fork in the road, decided to change its mind, hop the curb, and drive pell-mell through the woods to try to get to a different path. The first path — staying out of governance as much as possible, open-sourcing all the infrastructure, and becoming one provider of sims in a much larger marketplace — was and is preferable to what they’re now attempting, which is to retroactively batten down the hatches and try to control all aspects of their virtual world.
I’m all for opposing viewpoints, but a bit of respect might be nice. (No, adding “with due respect” to your characterization of the opposition as immature and infantile doesn’t actually make it any more respectful.)