The making of Schome Park Beta

In April 2007 the Schome project opened an island on the Teen Grid of Second Life exploring the role a virtual world could play in forwarding the projects larger aims of designing a new kind of educational system.  By the end of 2007 two phases of the project had been completed, with the island being redesigned between each phase.  A new island was purchased to expand Schome’s Second Life presence, and this page offers an illustrated overview of how that island, named ‘Schome Park Beta’ by the community, moved from concept in late 2007 to completion in January 2008.

Goals

There were a number of very specific goals for the new island build.  These were to:

  • Assure staff wishing to prepare and/or lead an event, subject strand or series of sessions they would have space in which to do so without dealing with the community planning permissions and regulations associated with more permanent developments
  • Design and develop a space with high design values and a more cohesive aesthetic than staff had previously delivered
  • Ensure the environment provided enough variety to appeal to different groups of users and their respective needs, goals and interests *
  • Deliver a higher level of in-world orientation and SL specific tutoring, of particular value to newcomers (a facility the team intended to expand and improve over time)
  • Finally it was always clear that the development would not involve building on every inch of the new island, nor utilizing the prim count in its entirety.  The island must have room to respond to the needs of the entire community, and thus evolve over time.

* one example of designing to provide a broader appeal to the student user group started in a Schome forum thread concerning the Bartle Test.  This test revealed that many Schomers were classified in terms of gamer psychology as ‘Explorers’.  As such several island features were designed into the new environment that were not immediately visible upon entering the island, but would only be discovered through exploration.  Of course one difficulty is that on what is a relatively small land mass it is the case that all corners and facets can be explored in a relatively short space of time.

Concept

The new island was approached with the aim of creating an environment with a more cohesive look and feel than had sometimes been seen on previous phases of project, phases in which a random assortment of builds and buildings had often developed in a fairly organic way.  This would be achieved through designing an ambitious sim-level ‘total concept’. 

The concept decided upon was that of a sealed crater.  A ‘bio dome’ would be created from the natural rock floors and walls of the crater, with an artificial man-made roof completing the sealed environment.  Conceptually the scenario was that the bio dome exists as a refuge from a harmful external environment.  The nature of the external environment was never categorically decided upon, but suggestions during the design phase placed the bio dome as a safe haven from post-apocalyptic or global eco-disaster scenarios.

The concept was fleshed out bit by bit with 3d renders of crater landscapes, overhead plans and notes detailing purpose and key features for each area and design element.  Much of this work was developed on a private staff wiki.

early crater concept renderearly crater concept render - elevatedpreliminary layout plan

Design Aesthetic

With the basic concept and features in place more specific ideas concerning the design started to develop.  The crater type environment was obviously the key influence and as such elements concerning rockiness, lava flows and terraces situated at different heights appeared.  With a desire to keep the Japanese Garden, the terrace idea also allows different locations a certain amount of independence.  The garden was placed very early in the design at the lowest level with access to the water.  Whilst the Japanese Garden has become a much loved feature since the first phase of Schome in Second Life, to me it fits the new island concept perfectly, representing the hope, freedom and leanings towards nature that are (intentionally) slightly more repressed in other elements of the build.

The other key early decision was to include underground sections, an element that had not been delivered during any of the previous phases, and to do so in a way that felt tangibly different to previous staff builds.  Thus the Crystal Caverns were conceived, using sculpted prims to create tunnels and surfaces with a more organic look and feel, and using colour-changing crystalline structures not only as decoration but also as a light source.

The crystal caverns then had an important knock on effect on other builds.  Expanding on the notion that ‘colonists’ of the crater would look to make good use of natural materials, the crystal became the key to many design details.  As a building material the crystal became a recurring theme not only as a material but also as an influence on shape as can be seen throughout the orientation area.  Schome Central embraced the crystal metaphor to its natural conclusion, with the entire building resembling a giant angular crystal.

Schome Central

The ways in which colonists would innovatively use their crystal surroundings was deemed to include the creation of power.  Power in SL is an entirely abstract concept as none is needed, but it became part of the design in order to create a richer, more cohesive environment.  As such crystals can be seen at the centre of the wind turbines where electricity leaps between the rotating blades and the crystal.

Crystal Turbine

The Build

Previous experiences terraforming directly inside SL had shown it to be a very time consuming process when dealing with large projects.    Whilst a number of options exist for terraforming outside SL the decision was taken to use a program called Backhoe.  The accompanying images show the finished terrain file as it appears in Backhoe, and how it appeared when uploaded to SL.  The time spent on large scale terraforming in Backhoe was ultimately a very efficient way to work, with approximately 90% of the island land mass as it appeared on opening day remaining exactly as imported.

Backhoe island Terraformed island

With the landmass in place the first builds to begin were the underground sections, literally with a view to working ‘from the ground up’.  These comprised the stone brick canalway (linking the water outside the crater to the lake in the crater interior), the stone brick tunnel (linking the canalway to the Japanese garden and the crystal caverns), followed by the sculpted prim-based crystal caverns section (including several vertical shafts to provide rapid exit/entry to and from above ground areas).  For the crystal caverns in particular this phase involved the creation of a shell, and not the full finished build.

TunnelingTunnel SectionCrystal Cavern Development

With the main cavern shell in place Schome Central began to be built.  It was developed from the inside out, with the staff presence boards being the first element to be created primarily due to it sitting lowest in the building, literally just centimetres above the crystal cavern.  Getting the ‘feel’ of this element right, particularly with respect to time-consuming texturing, also helped inform and secure the style that would appear in various aspects of the project.  The external faces of Schome Central are essentially a separate build entirely, a shell or skin enclosing the inner build.  Though prim heavy this provides a flexibility to both shape and texture that would be almost impossible to equal with a more conventional and lightweight approach.

Schome Central Development 1Schome Central Development 2Schome Central Development 3

As Schome Central progressed a number of developments took shape in parallel with the Schome Central development.  The first of the orientation elements to appear was the maze built by Anna Peachey.  The river bed and waterfall were designed to fulfill dual purposes, not just adding to the general aesthetic (particularly from the Japanese Garden) but also to hide the canalway underneath.  The presentation area and lava flow areas also began to come together at this time, providing a sense of continuity through the re-use of the colour-changing crystals, but also offering a stark contrast to the organic formations of the crystal caverns by being built in amongst dark, angular volcanic rocks.

Debating/Presentation Area 1Debating/Presentation Area 2

These separate elements however only really began to hang together in the penultimate week of development.  At this time the split-level flooring beneath the orientation area was put in place a process which included implementing the first boundaries of separate tutorial areas.  In doing so the focus on ‘cohesion’ was retained through re-use of the crystal texture, using a different colour to delimit and sign each orientation area.  Other texturing ideas from Schome Central were also re-used in the orientation area boundaries.

The final stage involved one of the more enjoyable elements, adding trees, shrubs and grass.  This process always seems to breathe life into the environment, soften up the hard edges, and make it feel inhabitable.  At this time Anna added the orientation elements related to building in Second Life, and Olly Butters contributed the scripts for each of the door systems.  One of these, the main ‘security’ entrance proved particularly troublesome and had Olly and myself working very late into the night just hours before the island was due to open.

The Good …

Ultimately the results of the build were well received.  Especially when using the Windlight client (which offers a few additional features compared to the current release client) the island has some genuinely pleasing elements.  My personal favourites would be the rotating wind turbines with their animated lightning textures, the viewing gallery area that you walk into having coming through the main security entrance, the texturing of the area of Schome Central containing the staff presence boards and without a doubt the more ‘hidden’ sections like the canalway and caverns.  The latter in particular really give the island a different feel from previous builds.

With the island having only been open a couple of weeks it is too early to definitively answer whether we achieved everything we set out to, but I’m confident by the end of Phase 3 we’ll all be pleased with what was accomplished.

The Bad…

From a reliability point of view SL has good days and bad days.  Unfortunately we experienced both during the development window and many hours were lost unexpectedly due to unplanned downtime.  This remains one of the big issues for institutions and companies who often host their own systems and services, or who would operate with an appropriate service level agreement when using third party providers.

Sculpted prims are a welcome addition to SL, but need to be used with consideration.  They occupy an area often much larger than they appear to.  For instance a sculpted prim apple is (approximately) a sphere.  However that sphere is contained within a bounding box, a torus-shaped volume invisible to the user, but a volume which the user nevertheless cannot penetrate.  Aesthetically this is a non-issue, but ergonomically it can prove problematic.  Try to walk though a narrow tunnel lined by such prims and you may find it literally impossible to get near let alone proceed down said tunnel.  The solution chosen in the use of sculpted prims on this build was to make them phantom, allowing the avatar to walk ‘through’ them.  This isn’t ideal however, necessitating the building of a non-phantom shell in which the tunnel and cavern sections sit.

And the Ugly…

The lighting model in SL continues to frustrate.  When using local lights SL can use up to six light sources.  It is relatively easy to design appropriate lighting in enclosed spaces (like the crystal caverns) to these parameters, but it becomes more limiting when working in large open spaces.  This can result in a kind of ‘popping’ .ie. when you approach a light source it ‘pops’ into life as you leave the threshold within which the client was previously prioritising another light source.  This isn’t so much of an issue if your client is only displaying a handful of metres at a time, but if the client is set to display a whole sim or more it does weaken the visual appeal, and you have to think carefully about light density and placement.  Additionally the lighting engine doesn’t perform the kind of real-time shadow-casting that some user may have witnessed in modern game engines.  Visuals certainly aren’t everything but some of the things the SL lighting engine does are a long way from delivering a believable world.

The basics of Linden Scripting Language (LSL) are simple to grasp for anyone with even limited programming experience.  However eliciting anything more than basic behaviours becomes an exponentially complex process, never more so than when scripting linked sets of objects.  Far too often negative experiences with LSL have led to design ideas and approaches being scaled back to make them work in what appears to be a rather poor object and scripting architecture.  Ultimately this remains one of the biggest barriers to developing more sophisticated ‘living’ worlds in SL.

Some More Pictures From the Finished Build

Avatars arrive onto the island through an entrance high above sea level.  Upon stepping through the doors they walk into this suspended gallery which provides a view of the entire crater interior.

View from the entrance galleryAnother view from the entrance gallery

At the lowest level of the island a canal way joins the water surrounding the crater to the lake in the centre of the Japanese garden.  A secluded entrance at the back of the garden links the open and enclosed spaces, providing one of the entrance points to the Crystal Caverns.

The underground canal wayThe Japanese Garden at nightSecluded entrance to the underground areas from the Japanese Garden

Underground tunnelsCrystal cavern tunnel sectionThe main crystal cavern

Schome Central is situated directly above the Crystal Cavern build.  A number of trapdoors around the area provide access to vertical shafts which can be used for rapid movement between the above and below ground areas.  The interior of Schome Central shown below and to the right, shows the staff presence boards and emergency help button.  It will also become the home for a range of community ‘freebies’ including objects, scripts and textures.

Trapdoor linking the surface and underground sectionsSchome Central interior

Advertisements
2 Responses “The making of Schome Park Beta” →

  1. Mark Gaved

    11 February, 2008

    The island is looking gorgeous Dan! Congratulations to the build team. Should be getting awards from the Architectural Association!

    Are students in it live? How many students? If live, what have their reactions been, have they asked to engage in the fabric-level build process? How are you managing their requests to influence design in an on-going manner?

    I really like the ‘glass dome’ – was this purely aethetic or partly as a means of controlling the vertical aspect of the island (given that the ability to fly means students could theoretically travel/build very high up).

    I’ll drop you a mail and catch up with you offline.

    Reply

  2. danseamans

    11 February, 2008

    Cheers Mark. Students have been back in Schome Park (both islands) since January 22nd. With groups from New York and the South East Grid for Learning now in-world the Schome community continues to expand.

    The students have influenced the Schome Park Beta build in an indirect way. Attention has been paid to what worked in the past and what didn’t, what we haven’t done but would like to do, and certainly responds to some very specific ideas that we’ve seen mentioned by students in the forum. What we didn’t do was ask “what would you like us to build?” Given that the original Schome Park (now Schome Park Alpha) is completely open to student development in this phase, they can already build whatever they and the community would like, and we’d much prefer they had that opportunity to work as a team and to contribute to the community space in that way, than us simply to do it for them.

    As such this build consciously delivers a concept the community didn’t choose, but nevertheless delivers on the objectives set out above, welcomes the addition and evolution of the space by the entire community, and may inspire something crazy and challenging on Schome Park Alpha.

    The response has been very pleasing thankfully. The most contentious issue has actually been the glass dome which you like. For me the inclusion of this was hanging in the balance for a long time. On the one hand it completes and indeed is essential to the island concept, but on the other it has a direct impact on the usability of the space. That said it does create a tangibly different feel to the environment, and from that perspective certainly delivers the feeling I was after.

    Reply

Leave a Reply

Fill in your details below or click an icon to log in:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out / Change )

Twitter picture

You are commenting using your Twitter account. Log Out / Change )

Facebook photo

You are commenting using your Facebook account. Log Out / Change )

Google+ photo

You are commenting using your Google+ account. Log Out / Change )

Connecting to %s

%d bloggers like this: