How Useful Is Second Life For Architects? "Linden Prize" Winner Compares Pros And Cons With OpenSimulator

Landmark "Rietveld Schröder House" recreated in OpenSim by Jon Brouchard using an imported SketchUp file(source)

As I noted last week, one of the first "Linden Prize" innovation awards deservedly went to Wikitecture, the innovative building platform that uses Second Life to design architecture spaces in a collaborative, "wiki" style. Co-created by real life architect Jon Brouchard (known as Keystone Bouchard in SL), it is in my opinion one of the most powerful real world applications of Second Life. However, that's not to say this makes SL an ideal virtual world platform for architects, at least not yet. In recent months, as it happens, Brouchard has started doing more architecture work in realXtend, the OpenSimulator virtual world, using Visibuild, a new service that enables the importation of 3D StudioMax, SketchUp, and other industry standard 3D authoring files.  (Here's his own company site, a great reference for architects interested in virtual world technology for architecture.)

Why the shift to OpenSim?  Last week I emailed Jon congrats, and then asked him. 

"I realize Second Life was never intended to be an architectural software," Keystone tells me, "and perhaps it never will be, even though it already offers a rich combination of features that are very attractive to architects that can't be found elsewhere (realtime, multi-user, collaborative, etc.) But the missing link is 3D model interoperability. I appreciate what a technically challenging feature that is, but the time and skill it takes to replicate professional models with 10m prims just doesn't always outweigh the benefits. My clients have their own reasons for working in other platforms, but for me, the only carrot is the ability to import models."

In other words, Second Life still has value as an architect's resource, just in specific contexts:

"For Wikitecture," he continues, "Second Life makes sense because that kind of community-wide collaboration requires simple modeling tools, prim-like granularity for collaboration, and access to a large and diverse community. But that's when the ability to export prims into professional software becomes as critical as importing models. At some point, the designs intended for the physical world needs to get real, and working drawings for construction need to be generated. Until building contractors start accepting prims instead of Construction Documents, we're going to continue searching for model interoperability! In the meantime, I intend to continue enjoying Second Life for what it is, and expanding my purely virtual design practice, where interoperability isn't as much of an issue."

Source

As that suggests, Keystone still does SL-based projects. In fact, he recently completed the new auditorium space for the Metanomics show (pictured above) which Dusan writes about enthusiastically here. For that matter, OpenSim has its own considerations:

"It's definitely ready for early adopters who are OK with a little experimenting and troubleshooting here and there, but then again, Second Life is like that too-- for professional architects anyway. However, I am becoming increasingly confident that this is the right vector-- the right combination of tools architects will need to take advantage of this medium. Model imports, in-world modeling tools, permission systems to enable collaboration, and more. All of the pieces are falling into place.

"However," he adds, "I still think Second Life will play an increasingly important role in all of this, as a community hub, and the 'town commons' of the metaverse. It is still by far the most stable environment (wow, never thought I'd say that), and the most robust/diverse community of users. That will continue to have a lot of value into the foreseeable future, in my humble opinion."

Comments

[radar]

Linden Lab REALLY needs to take a look at what others are doing to allow importing of meshes and 3d models that's currently not possible with the limited sculpty model. I don't honestly know how well some of these other options scale, when you're importing models into realXtend, etc, but I do know that it opens up a whole new world of possibilities and ease of quick modeling to implementation in world for builders.

[Graham Mills]

I agree with radar. At the Federal Consortium of Virtual Worlds meeting a couple of weeks ago, this topic also appeared to be quite high up the agenda.

[Mo Hax]

Although used synonymously in this post, OpenSim and RealXtend are different platforms born (once upon a time) from the same OpenSim code base. RealXtend is a significant fork from OpenSim with its own goals, governance, and functionality, including 3d mesh import which OpenSim does not support (last I heard). I'm not suggesting one is better than the other, just that they need to be considered separately when making a platform selection.

[Mo Hax]

That's OpenSim and RealXtend. Apologies for broken URLS above.

[Mo Hax]

The modrex project is one attempt to bring OpenSim and realXtend into some level of compatibility. Again, the two are not compatible today and are certainly not the same thing.

[John Phoenix]

The headlines in the paper should read: Linden Labs commits suicide! That is basically what they did by giving away the source code for the SecondLife viewer. OpenSim and all related projects like RealXtend is only possible because open source community developers have back engineered the viewer to figure how the SecondLife server side software works. The Open Source community can build the tools the people want without the restrictions Linden Labs puts on it's SecondLife users. OpenSim itself although still in it's Alpha testing phase, is being designed to be more multi-platform and application versatile. This means technologies for wider ranges of applications can and will be and are being developed for OpenSim making it a truly 'Open" platform. Because of this, OpenSim will one day surpass anything Linden Labs tries to do with Secondlife. The wants and needs of Sim community are simply being addressed by OpenSim developers, who for the large part are fed up the restrictions of the Secondlife platform and they have the means, to make wonderful things happen. I predict that within the next 3 to 4 years or sooner we will see OpenSim grow into maturity and be such a strong competitor with SecondLife, that SecondLife will be seriously damaged and will become ' The old way to sim who's time is at hand. ' Yes, I believe the one day OpenSim could actually shut Linden Labs down because they won't be able to keep up with the progress and innovation of OpenSim and it's realated technologies.

[Mary Elizabeth]

Basically, LL gave $10,000 to a group of architects who don't need the money and are really sort of moving away from SL as a platform anyway.

Was the award an attempt to get them to invest more heavily in SL?

[FlipperPA Peregrine]

@John Phoenix: I'm afraid you are incorrect, and this is a common misconception. Open sourcing the viewer had nothing to do with OpenSim development. The protocol was reverse engineered prior to the libsecondlife project, which in many ways led to the start of OpenSim, BEFORE the viewer was even open sourced. OpenSim developers are explicitly discouraged from looking at the SL Viewer source, due to conflicts in the GPL license the viewer is released under versus the BSD license OpenSim is released under.

OpenSim would have still happened (and, in fact, it existed before) whether or not Linden Lab released the source code to the viewer.

[Hamlet Au]

@Mo: That's a fair point, thanks, I had considered adding an explanation on the relation between OpenSim and realXtend.

[John Phoenix]

Ah A common misconception.. I see.. Thank you for clearing that up for me. It seems I learned from that school that was incorrect. Now I know better. I have only been reading up on these issues for a month. however, I do still think OpenSim is going to be poised one day to surpass SecondLife and Linden Labs. Yeah Hamlet you are right. A lot of people still think of OpenSim as a sim itself like SecondLife. They fail to realize the OpenSim is the server side platform that many different grids and technologies like RealXtend are either run on or developed from. Thus RealXtend is one of many 'clones' of SecondLife if you will. RealXtend has even branched from the regular OpenSim type sim and is developing it's own new wonderful technologies and is an excellent example of how this community is growing it's technology to suit the users needs. There will no be one SecondLife type sim, but many and already are now. This is going to give Linden Labs tons of competition once Opensim and it's technologies mature.