Chez Serge

Second Life has recently generated a lot of buzz. It all starts with a post on their web site : restructuring. More precisely, 30% of employees are fired, and their Singapore office closes.

I have seen many analysis of this, talking about too much business orientation, or a step back to B2C.

It is a serious issue for all people who, like me, believe in the business of virtual worlds. We need to remind the history of Second Life. At first, Linden Lab was created by Philippe Rosedale with the aim to build a software that would be sold. Second Life was created as a showcase, a proof of concept of the capacity of the platform. However, Philippe is more and more interested in the social aspect of Second Life, to a point that he may have forgotten the business issue. Not a good thing to do when you have Mitch Kapor, Pierre Omidyar, and Jeff Bezos at your board. Well, facebook is in the same mood, but facebook is over 400M accounts, while Second Life is around 16M, which is low, even though the peer to peer transactions amount for 567MUS$ in 2009, a very nice result in my view. So Philippe was replaced by Mark Kingdon, who decided to go back to the basics, and create an enterprise solution. Well, he created two enterprise solutions : the first one is just owning islands in the main grid, with specific features such as more flexible avatar naming, the second one, “second life enterprise beta”, being beyond the firewall.

This is, in my view, the main issue: it is not possible for a company to have both a B2C offering, and a B2B one. Amongst all issues, two are huge. The first one is about culture; sales process are not the same, sales people are not the same, pitch are not the same, eventualy it means having two different teams, which is costly. Let us remember the excellent words from crossing the chasm: “better have three customers who talk together than ten who don’t”. The second issue is about product roadmap; obviously, the requierment from a social network are not the same as the one from companies, specially the corporate ones. This leads to tension on the product. Moreover, social networks are not that much the friends of CIOs… Consequences are dramatic: the residents are more and more disapointed by SL; the B2B inworld offer seems to offer a low level of reactivity (friends of mine talk about many weeks before having an answer), and Second Life enterprise beta business model, 55k$ per year for 16 islands but only 8 simultaneous, looks outrageous, specially compared to other collaborative tools.

What is emerging now looks like a willingness from resident to step back from SL. For those close to the original spirit of Rosedale, opensim seems a good candidate. However, remains a big problem: what about all assets inworld? Everybody has in mind the closure of metaplace, and the resulting loss of assets. Maybe this is why VenueGen has offered to migrate enterprise assets into its own world?

Shall we assist to a Second life crash ?

Whether it happens or not, I still believe in virtual worlds. This excellent post made after Thinkbalm decision to widen its scope (we must all thank Thinkbalm for their constant support to Virtual Worlds), shows many positive coming aspects. In the hype curve, Virtual Worlds are clearly at the Trough of Disillusionment. Time is now on for technology to show its real potential.

What are the positiv aspects?

• The most important is the vgov initiative, lauched by US administration, whose aim is to manage virtual worlds for all US administrations. The Internet exist mostly because of the important role played by US administration, the same story may happen with Virtual Worlds. Four platforms have been selected : Web.alive; Forterra (acquired by SAIC); teleplace; and Vastpark, a very innovative opensource platform. Second Life has not been slected.
• Education already makes a huge use of virtual worlds. The special issue of the journal of virtual worlds research shows interesting results.
• Healthcare sector is also a place where virtual worlds have had big success, in SL, but not only. A study performed by doctors showed the great advantage of virtual world as a training tool.
• The enterprises are more and more interested in serious games. I can’t imagine serious games without a virtual world. With this respect, the openness of the technology becomes a key success factor…
• Tradeshows, conferences, are more and more moving to virtual worlds, with big success. Not only cost of travel is lowered, but the quality of interaction is superior to 2D tools, as shows IBM case study. Robots have now their virtual conference; and even wine producers go for it !
• On the social side, Kids and teens are massively in virtual worlds. Almost 400 millions users in the range 8-15 years old are in virtual worlds, comparable to the number of facebook users, but who has a larger scope in terms of age.

I would like to add a personal touch. Many times, I have met people very reluctant to the internet, people who did not understand the importance of social networks, forum, for whom the web was no more than an online content tool. I was always able to change their perception by showing them machinimas realized in corporate situations. Their reaction was always the same : “at last I understand what it is all about”. Well, let us be honest : facebook interface is amongst the most complicated we can imagine. At least, a 3D interface is easy to understand.

The immersive web is now entering a time of obvious concrete realizations. On the technological side, tools are diversified, some close to games or social networks, some close to corporate intranet tools. The biggest improvment is their availability through the browser. I It is now obvious thet the browser is the entry point to the Internet, and 3D plugins are the preferred solution to clients download. WebGL norm, when available, will be a huge step.

Virtual worlds are now entering their mature phase, and we shall all thank SL to have shown its potential value.

The conference was held 13 /14 may 2010. Link is here : http://www.ndu.edu/irmc/fcvw/fcvw10/live.html

Twitter : http://www.twitter.com/fcvw

Vastpark with player 1.3 vp://demo.vastpark.com:8278/fcvw

I had recently a very interesting conversation with Greg More, from the Spatial Information Architecture Lab at RMIT. Greg is a really innovative person, sees the potential for virtual worlds, and does a lot of interesting projects around spatial data visualisation. Greg and his students have worked with different virtual world platforms.

Greg @ RMIT

In the middle of the lunch, as I asked him about Vastpark which he uses a lot, he suddenly said “Second Life is like writing in html, Vastpark is like coding in php”.

I found his remark fascinating, especially in the context of corporate companies usages. Collaboration is, so far, the main application of corporate virtual worlds, and both companies have made such announcements. Linden Lab launched its collaborative portal, and Vastpark opened the beta of 3C, its collaborative environment.

So far, Linden Lab has not released a new tool. Their collaborative portal is a marketing campaign aimed at presenting interesting usages of Second Life by corporate companies.

The architecture of SL is quite straightforward : one island is one processor, and users do have a set of tools to create content, which they own. Authentication is performed via a centralized server outside of the corporate firewall, a little bit like a telecommunication operator OSS/BSS.  The client is open source, but not the server. In terms of API, apart from SAP which seems to connect to the reader and not the server, the only open APIs are registration, mapping, search, and some live data. Objects created are stored on the platform, and any duplication is done “manually”.

Well, this reminds me of the first e-commerce sites, where people had to manually transfer the catalogue from a database to full html web page. Pure hard coding. Then, came php, mysql, mashups, javasript, ajax, etc… And, suddenly, life became both much easier, and much richer.

Now, Vastpark. It is an open source product, made of 4 components, all downloadable (with source code as well): the reader, the creator, the publisher, the server. The reader, the creator, and the server are easy to understand. But what is the publisher?

The architecture of Vastpark is totally different: it is a distributed architecture based on components which can sit virtually anywhere. When someone creates an object, this object is published somewhere, and can be retrieve by any server, by any developer. The virtual world which is developed can simply access all the objects created,  in a mashup type logic. Therefore, no need to duplicate elements inworld : duplication of the access is enough, as long as the objects are published (many already are on Amazon S3).

Another interesting point: everything can be described using IMML, an XML type language developed specifically for this purpose. Therefore, objects can be created “on the fly” just like php which generates html on the fly. In fact, it can be done in three ways: through its IMML markup, using a scripting language (LUA is currently supported but more choices will come) and through plugins (plugins may be written in any .Net language: Iron Python, C#, VB, etc). All of these can be written as components of remote widgets that developers can easily embed into their worlds.

In Vastpark, almost everything is a plugin. Voice control is a plugin. Avatar control is a plugin. We could even say that Vastpark is just a network which does interconnect plugins. With this respect, connecting the server to existing content is easy. A connector to Twitter, Flickr, eBay and Skype are already available.

This architecture is much more distributed than SL (who runs three server farms, one in Seattle, one in San Francisco, one in Texas). The partnership with Badumna allows to interconnect in a very efficient manner users world wide distributed.

We can see now the difference in the architectural approach: the component based chosen by Vastpark allows a greater flexibility, and a smoother integration into brand new worlds. As the virtual world projects mature, as more and more virtual world will integrate coorporate companies, this forward thinking approach makes all the difference.