Low prim, high fidelity architecture for Second Life - The Heretic Pixies store

This is the first in a series of posts about the design and contruction of our new Heretic Pixies store.

For most people starting out with a building in Second Life, prim limits (the number of building blocks you can permanently place on a piece of land) are probably one of the most limiting factors. With an unlimited number of prims it would be easy to create an attrctive building, but when you may only have 30 or so, your options are fairly limited. Contents is always a factor in this - you want to have enough prims left to be able to put something inside your building.

At Pale Heretic we use a process that allows us to produce high fidelity buildings by pre-rendering things like complex shadows into the textures themselves. One of our products is a single room that has less than 30 prims, but the interior conveys a sense of light and space you wouldn't normally find in a small environemnt (the box measures 10m x 20m). You can see another example of this that we are constructing at the moment - our new Heretic Pixies store.

The Heretic Pixies store is constructed from 16 prims (for the floor, ceiling and walls) and measures 20m x 15m. It's going to be our store for a new character based range of products and we want to make it look as plush as possible on a low prim budget. The first part of this is making the interior itself look as good as possible and to do that we are using baked interior textures. Doing this allows us to both get realistic graduated lighting and shadows on the interior walls, but also allows us to put murals on the walls and for them to look like they are on the wall, rather than floating in front of it.

To do this we build a replica prim-for-prim in an opensource 3D application called Blender (www.blender.org). We then pick the faces that we wish to bake textures for and they are individually textured. The textures we use for these surfaces are sourced from photographs. The key is to avoid obvious tiling of textures. Seeing a regular repeated pattern in a wall or floor will break the illusion of reality. For example, the concrete texture for the walls in the Heretic Pixies store is created from a number of unique photos, layered and mixed to create a 2048 x 1024 texture that can be used to span the longest single wall with out any repeats. Onto this we add logos and other graphic elements, then bake out the textures, so that the local lighting and shadows are precomputed into the textures and we do this so that we get a good match with the lighting that we add in Second Life.



Generally, we will create a single 1024 x 1024 pixel texture for each 10m x 10m face. This increases your texture upload costs because you need a unique texture for every face (16 for the interior of the store) but then, that's still less than one US dollar. Using high resolution textures like this, while having some impact on older graphics cards, does allow the use of text and graphics directly into the wall texture without the need for additional prims and textures for signage.

The design and build for the Heretic Pixies store has so-far taken about 5 hours, since we already had the images and photographs that were used to create the textures. Over the next few days we'll be adding more to the store interior and exterior. If you would like to take a look, you can get there from http://slurl.com/secondlife/Dangun/176/77/143/ or by searching for 'Heretic Pixies'.

Second Life ... the things we need to fix next

I'm an advocate of Second Life for a number of reasons: it's adaptability to a range of uses, the completeness of it's model and workable DRM, the friendliness and creativity of its population, and the richness of its experience.

Having said that, there are a couple of things about it that are becoming (in my opinion) a serious issue.

The first is the mature content. Now, I know everything has its place, but Second Life doesn't have enforced zoning to manage the spread of locations that include mature content or a content filtering solution implemented in the client viewer to allow you to filter what you see ... much life the safe search filtering you find in the image search tools on Google.

Second Life needs to learn some of the lessons that the internet as a whole, and the search engines especially learned a while back - people will keep on putting sex on the web, but you need to offer some workable tools that allow users to choose what they see before it arrives.

The next thing on my wishlist is some search integration outside the SL client - I'd like to see the content from within SL presented on a search tool that can be accessed directly from the SL website, and more than that, I'd like to see it done in a way that Google can index. My aim with this is that,for example, whenI search for 'IBM Training' on Google, included in the returned results are the locations of IBMs SL training facilities, and not because they took the time to add the SLURLs on their website, but because the small-ads associated with the locations are accurately indexed by the search engines.

My last wish? Proper web-page-on-a-prim. I want to effectively see the output of the browser rendered onto the surface of an object and that includes all the mime types my browser supports (flash, pdf,video etc). The one media / web page limit per parcel is too limiting. There is a huge volume of traditional web content and tools that would work very well if they could be displayed on an in-world object and have the functionality preserved ... scrollbars too.

But what I would like to maintain is that SL is very much a viable platform for us to take VWs forward - we all know it isn't perfect, but taken as a whole it is a viable, robust, well tooled and documented solution that has the buy-in of it's steadily expanding population.

Goodbye Lively...

Google Lively posted this today http://googleblog.blogspot.com/2008/11/lively-no-more.html

I have to say, despite my issues with what they chose to do and how they managed telling their user community about it, that I think it's a pity Google have chosen to close lively.

Whilst it wasn't the revolution that we were all hoping for, in my opinion it did have it's niche.

What I would hope is that this isn't the end of Google's interest in virtual worlds. We need another big player to step in here and create some diversity with a decent budget behind it, and no, in my opinion that doesn't mean we need another mirror world. What we need is something that can realistically compete with Second Life. Something that can encompass user created content (from within the VW engine), with rights management and a workable economy, and better media and web content integration.

In my opinion there are only a few companies capable of doing this. Google, Apple, Microsoft, and the larger games companies.

Google: please don't let one failed project keep you out of this sector for long, and if you want some advice on how to position your next offering ...

Google Lively ... gone a little quiet?

Over a month ago now, Google sprung Lively onto the virtual worlds scene to a mixed reception, but certainly a lot of discussion.

Now, things don't seem to have moved on very much.

I was expecting Lively's launch to be rapidly followed up but some direction and commitment from Google explaining their intended direction, time scales, goals. This just doesn't seem to be the case.

We know that content creation tools are somewhere in closed beta, but have no idea when that will end. I do know that Google hasn't yet finished deciding on either rights management and copyright ownership for content creators, or plans about monetizing that content. Conversations with people at Google lead me to believe that those issues are being discussed at the moment.

The "Lively is only 20% complete" is an interesting stat that I'm not sure has ever been coroborated. What I'm more curious to know however is: how long will we need to wait for the next 20%?

The one thing Google Lively has over Second Life...

The number of blog posts about Lively seems to be increasing at a healthy rate, and some common themes are emerging.

A significant number (and you can count me in this group) are disappointed with Lively and what it offers. Some are glad to see Google enter this sector of internet communications and others are picking up on distinct aspects of Lively and it's emerging user base.

One of the themes that I've seen picked up in a number of places is the rate with which the sex/porn industry and it's fan base has jumped into Lively. There's no doubt that within the first 24 hours the first Lively cybersex rooms had been set up.

It's a bit of a sorry state of affairs, but lets face some facts about the history of the internet here: without pornography, games and file-sharing the internet that we use today wouldn't be anywhere near as developed and mature as it now is.

The same is true for virtual worlds as a whole, but one key factor is the relative youth of the virtual worlds market compared to the internet / world wide web as a whole. I think of virtual worlds in the same way I remember the internet about 7 years ago - more chaotic, more randomly mixed (with pornography popping up in the most unlikely of places and seeming to be 50% of the results from any web search). Second Life is a good example of this. Wonderfully creative content designed by all sorts of people and leading to some amazing experiences. Scattered in between is the virtual world equivalent of all those random sites I used to bump into on the web.

Something I noticed recently was that they seem to largely have vanished. The web seems like a much cleaner, nicer place to explore now. This is a change I'm waiting to see happen with the virtual worlds. Just to be clear, I'm not saying that there isn't a place for sex in virtual environments if that's your thing, but what I don't want to have to worry about is who's going to buy the plot of land next to my Second Life office and if they are going to build a brothel on it.

A suggestion for Linden Labs: wouldn't it be nice if anyone running an adult site in Second Life (was required to) set a flag on their land specifying that their site fell into that category. Then in my client application, I can set an option that says "don't show me adult content" and then my client simply doesn't render any avatars or objects that are within that piece of land? Seems like a simple process and would certainly make it much easier to promote Second Life as a business and educational tool. The only problem I can see with this approach is that it lessens that need to buy an island in order to get some control over what is around you, thereby reducing Linden Labs land revenues.

This is the one thing that Lively has over Second Life. No neighbors. My Lively room is a component on the homepage of my website. If I want to direct people to it, I can send them straight to my website, straight into the room. This is a branded experience and I have control over all the elements of it.

Second Life, until something changes, will always have the risk of someone erecting a 50m tall photo-textured penis right next to my office. Oh, and that isn't a challenge by the way. My current preference is still definitely for Second Life - just or the amazing scope of what you can do and (all things considered) how well it all works. But if Lively's artist tools are as good as they *could* be, and the rumor that they are they are only 20% through their development turns out to be true...

A little more Lively?

You may have seen my previous (and somewhat disappointed) post about Google's new web-based virtual world, Lively.

I was bemoaning the lack of an apparent strategy or any solid information about the future of Lively or the kind of service into which it may grow.

Not being one to give up on new technologies straight away, I emailed Google about becoming one of the content developers for Lively and tagged a couple of additional questions on to the end of the email.

The reply that I received from Danny on the Lively team at Google did provide a little more detail (I'm omitting Danny's surname so that he doesn't get a surge of emails from people guessing his email address).

On the subject of content development for Lively Danny replied "Content creation for Lively is in closed beta right now, but I put your name on the list of future developers. We are planning to release the artist tools soon and we'll let you know when we do."

Now this is promising. Comprehensive content development tools for Lively would make it much more interesting as a virtual worlds tool.

I also asked about plans for a Linux version of the Lively client. Danny responded with "We're trying to bring Lively to as many different operating systems and browsers as we can, but I don't have any details on when they'll be available."

That sounds even more interesting. Following a less-than-harmonious experience with my previous laptop, I'm very happy with my EeePC 900. This of course is running Linux (OpenSUSE 11 to be exact) and I'd much prefer to be using Lively on this, rather than resorting to a PC.

I hope that this information is the start of more detail being provided by Google about lively and it's place in their online strategy. I hope to continue my dialogue with Danny and will let you know any information as I get it.

How do I look in this?

One feature common to many Virtual Worlds is the ability to change your virtual self (your avatar's) appearance at will. But your options probably go further than that: you have the scope to change your face, shape, gender, even species, and to do so with relatively little fuss. I can't think of another environment where re-inveting yourself has quite such broad possibilities.

So, the questions that spring to my mind are: what's wrong with experimenting? what will everyone else think?

Firstly, there's nothing wrong with experimenting. Try everything out. It's an interesting experience.

More issues surround the second question, and the answers will differ depending on the environment and your relationship to the people with whom you are interacting.

I'm going to use Second Life as an example in this, but (to varying degrees) I have seen parallels in a number of online game environments.

One of the simplest ways that people will identify you as new to Second Life is your appearance. Looking like one of the default characters usually means that you simply haven't yet discovered the options to change your appearance or the ability to change your clothes. Looking new in SL seems to have a couple of effects - people (especially other new residents) will be more likely to talk to you, where as more established residents may be less likely to spontaneously engage you in conversation.

The converse, is also true. If you look like you are an experienced long-term resident, you will seem to be one of the elite to a newly registered user and they will be more likely to address a question to one of their peers. Ironically, appearing to be non-human (if your chosen appearance isn't scary and doesn't have negative associations) may actually make you seem more approachable. Wearing a cartoon chicken suit certainly stops anyone seeming threatening.

When choosing your appearance, whilst the freedom to express one's self is great, it is worth remembering that the medium is jut as important as the message ... and in this context - that of a dialogue taking place, your avatar IS the medium. Sounds like stating the obvious, but given the number of people I have recently seen pitching Second Life, in-world, to a group of very new users whilst dressed as something very obscure / disturbing / inappropriate I think it's well worth pointing out. And the best bit? This is Second Life. You can change your whole appearance as easily as you would change your clothes. Choose a suitable appearance for the people you need to address. You can always change back into the avatar of your choice as soon as the meeting is over.

Google's Lively ... a missed opportunity?

A couple of days ago I started spotting the press-releases about Google's new virtual world offering called Lively.

I have to point out that this is a Beta and so we can expect some fairly major changes before the final version.

Lively, however, seems to fall somewhat short of the expectation put upon it by headlines like "Google's challenge to Second Life" which appeared on the BBC news website. Linden Labs really don't have much to fear from what is very much inline with my expectations of Virtual Worlds a few years ago and reminds me a lot of Adobe Atmosphere.

Coming from the company that brought us Google Earth, this seems miss-targeted and limited in it's conception. The concept of a 'virtual island on your webpage' is quite interesting as a nice extension to your website or blog, but the execution of it will I think type-cast it into a younger / teen market.

The other omission is the lack of any apparent developer or content creator support, or a roadmap for the plans associated with this. Given that Second Life has somewhat set the bar in this I would have expected Google to have attempted to address the developer community about what they can expect to get from Lively over the next year.

One final point is platforms. C'mon Google. Asus launch the EeePC and cannot make them fast enough to satisfy demand. Acer, HP and MSI (and possibly others) are all producing similar platforms. What do these all have in common? They all run on a Linux based OS (although a few do offer more expensive / less well spec'ed MS Windows versions). Launching a plugin like Lively and not thinking about a Linux version (the Lively site only mentions a Mac version in the future) seems to be cutting off a significant part of your target market. To make the example - I can easily run Linux Second Life on this EeePC 900. Lively? No.

So, Lively is on my things to watch list, but is something I hope will grow into something that I can use, because it falls short of my expectations at the moment.