Accenture Chief Scientist: Virtual Worlds Need Standards

Kishore Swaminathan,  IT consulting firm Accenture's chief scientist, spoke at the company's annual CIO conference on cloud computing, but he also discussed virtual worlds, mostly, it seems, to say they don't have much value yet. "I don't think, two years from now, there will be a Second Life...," said Swaminathan. "These worlds are 'closed' and have proprietary currencies. Perhaps open standards will make [virtual worlds] more attractive to companies."

Swaminathan's been beating the interoperability drum for some time now, predicting the death of Second Life for the same reasons in December and expanding on the point for an Outlook article in September.

Interoperability would be nice, certainly, but more important than interoperability between virtual worlds is the ability to settle on standards for assets, whether between virtual worlds or simply between virtual worlds and standard development tools.

From conversations I've been having with people in the enterprise space, it seems like there are actually plenty of interested companies--in April Justin Bovington said that 85% of Rivers Run Red's business inquiries were then focused on enterprise--and that ease of use, access to existing tools and Web services, security, and reliability are more important.

That's not to say that interoperability between virtual worlds on a level approaching the Web wouldn't make for an ideal catalyst, but it seems like slow and steady in other areas might prompt the tipping point first.

There's one question I've been asking most people I talk to, but I'll go ahead and put it out for everyone (I'd love to hear as many answers as possible): If you're doing business in virtual worlds, what do you think is the most urgent problem to be solved?

Comments

[nic mitham]

i guess it depends on what type of business you're seeking to do.

if its marketing then the size and type of audience is a primary consideration.

the degree of brand management and control is also a major factor for companies - explaining why (for some) controlled worlds like there, vmtv and stardoll are more attractive.

if its colloboration (mainly internal stuff) then server and asset control/mgmt are important.

etc etc

[Kim Smith / Rissa Maidstone]

In my opinion, the most urgent problem(s) to be solved is ease of use for the user interface for those new to Second Life AND the firewall issues caused by the portals SL uses. Many companies cannot access Second Life because of this.

However, given those basic issues, the next would definitely be business collaboration tools. The ability to use a "whiteboard" and share documents in an easy, coherent fashion would be a tremendous boost.

[Tony]

I would tend to agree with Kishore on this one. If open standards are held up between virtual worlds, and (even better) open platforms are prioritized. Everything else will follow.
Apache runs the majority of webservers and is an opensource platform. Because of the open source nature of the platform many specialized modules are available to meet individual's needs. Also, if some custom feature doesn't exist you can create (or hire anyone to create) it for you.

Extending that to 3D platforms would mean that people's efforts should be focused on making the open source projects (opensimulator, RealXtend, Project Darkstar...) the standard along with contributing as much effort to make them as stable and functional as possible.

[Peter Quirk]

Identity management and integration with enterprise directories is key for large companies. No one wants to maintain parallel directories to map real names to virtual names in SL or Instant Messaging systems. The problem will be compounded when we have PBX bridges into virtual worlds.

Next is application integration, which for all practical purposes requires virtual worlds inside the firewall, much as IBM has done. Everyone is waiting for Linden Lab and the realXtend teams to deliver clickable web pages on prims or meshes. I wonder if Citrix also has a play here. (Are they even thinking about it?)

Cross-world markets for assets are important, as are standards for scripting assets and avatars.

Tools for compliance, security, watermarking, encryption and auditing will be required for serious enterprise applications that interface with public worlds. Tools for backup, recovery and performance management will be required by any IT organization delivering virtual world services.

Also some standard licensing system will be needed for expensive assets or services that are required for a limited time or by a restricted number of people (e.g. subscriptions to information services.)

[len]

A realization that without attention to language-based standards, asset portability, identity portability and long-life cycle maintenance are fantasies.

How do you think you got into the mess you have now? It isn't all Second Life's problem or their fault. It is the market itself that is fighting standards. You should ask yourselves why there are no standards you accept when there certainly are standards.

[Ron Edwards]

For enterprise applications, interoperability with corporate data and systems (including telecoms, LMS and CRM) are as important as 3D asset interoperability.

Authentication and identity are very important for organisations so people know who they are talking to and trust that it is them. Having real names associated with avatars with authenticated login from LDAP/Active Directory and photo based avatars are now meeting that requirement.

The use of 3D mapping technologies to visualise data is rapidly growing with organisations adopting Virtual Earth, Google Earth and other map based solutions to integrate with their own data. Enterprise virtual world platforms should standardize on GIS data interoperability and become geospatially accurate. Of note, Forterra have offered an open standard for this as a starting point. -- disclosure - we are Forterra's European partners --

The urgent problem on the demand side to be solved is to build awareness that there are enterprise grade solutions available and that the above mentioned aspects of reliability, security, authentication and interoperablity are important and don't have to be compromised. On the supply side, asset interoperability would help fuel growth and increase client platform choice.

[Damon]

I would agree with Len on "A realization that without attention to language-based standards, asset portability, identity portability and long-life cycle maintenance are fantasies."

I often wonder why I hear participants in virtual worlds talk of inter-operability and standards when those as Len pointed out exist. Now I do run into those that date themselves by saying, "you mean VRML?" The burns last long I can tell, but when I hear the need of a real-time communication of 3D data across all applications and network applications, isn that X3D? Its straight off of the Web3d Consortium page. They already have relationships with the OGC, W3C, and Khronos. True there is much needed to help this spec get ready for the upcoming Open Metaverse, but what platforms or STANDARDS are closer. Why cant we all work together to get this done? Is it bad blood, hurt feeling, or the need to re-invent that keeps the current standard out of these talks?

I dont know, but what I do is that for the true 3d web to succeed, we have to work together and open is the only way to go.

[len]

It's money, Damon. Open IP systems based on code even open source are still effectively closed to the tier of artists who don't create at that level, to the barriers are still in place. Go up a level to a language base (eg, X3D) and the barriers are lower but the complexity of using the language is still there. Go up the next level to the editors and it is easier to remove the language layer, but as in Second Life, the manacles are just as strong as on the lower levels.

So it will always come down to how much interoperability and data portability for how much cost/profit versus market opportunities. It really isn't difficult to build a metaverse of independent worlds. That's just YouTube with 3D. It is difficult to move an avatar across worlds on different platforms.

Sharing identity is straightforward. Sharing the affordances is not.