A Business Model for Sims.

[swestrup]

So, I've been thinking lately about makeing an open-source version of the Sims, tentatively called "Crystal Cities" because it will be based on the open source Crystal Space Engine. Now I posted here about the features that I felt such a system should have, so today I'm just gonna talk money.

If one does research on The Sims you quickly discover that it is a) the most popular game ever produced and b) lots of folks are unhappy with its limitations.  I think this adequately indicates that there is a market niche (if not an entire cavern) that one could fill. The secret to being both profitable and open source has been shown by such folks as ID with Doom and later with Quake. They freely give away the game engines that they use, but they charge for the combination of engine+levels. Clearly then, what they are selling is the artwork and level designs. Something very similar can be done with regards to the Sims engine. That is:

1. The game engine, design docs and API manuals are freely given away.
2. The system supports cryptographically signed template objects that represent bodies, clothing, furniture, walls, etc.
3. Signed templates can't be modified without breaking the signiture, so an entire level of overrides, modifiers and reparsers must be supported.
   
4. These crypto temlates can be used to display an object instance at low resolution for use on a Sim, or at high resolution for when doing walkthroughs or for modelling screens.
5. We use cryptographic money protocols for the installation of templates into the system, making them not only unforgeable, but unduplicatable as well. Once installed, a template can generate an unlimited number of in-game object instances.
   
6. We now have the basis for a real world economy based on the production, consumption and trading of object templates.
7. Producers of new templates may give them away anonymously, sign them and give them away so they are freely duplicable, or sign and seal them so that money changes hands in order to acquire a copy. You may also buy a copy of a template, wrap another layer around it, and sign, seal and sell the result. Every sale you make will require you to buy another copy of the base template...
8. The base game designers and artitsts will sign their work, but won't charge for it. They will be compensated for this by being issued shares pro-rated (by an as-of-yet undertermined method) to the importance or popularity of their contribution. These shares will pay dividents from the object brokerage service set up to help object buyers and object sellers find each other (and which will take a cut on all transactions.)
9. We will also approach real-world furniture and clothing designers and ask them to fund the development of templates that mimic the designer's products. These templates will be a form of advertising, since all game objects will support queries such as "Who designed this?" and even "Can I buy a real-world instance of this?"
10. The overall intent is for there to be a base game that one can download and play, complete with a large selection of free objects to include in the environment. The better the game play of the completely free version, the better. This is to get a prospective buyer hooked on the game and thinking about expansion. From their template catalog they will be able to download new free templates or shop for commercial templates.
11. The game must allow multiple games to connect and form distributed networks of neighborhoods. That way, you can have your Sim show off the fact that you paid for a gucci handbag for her, or bought him a Ferari. This will help cause the 'keeping up with the Joneses' sort of mass consumerism that will make swestrup a rich boy.

So, what do folks think, will it fly? Obvioulsy there are many details that need refining and protocols that will have to be carefully designed, but does anyone see any showstoppers that will cause it to fail as a plan for world economic domination?

Comments

[labyrinthman.livejournal.com]

Well, you don't know me, and I don't know you... but as long as you can put together a better distributed neighborhood than "Sims Online", you've got it made. Thats been a real thorn in EA/MAxis' side.

Frankly, I'll be happy if you can make the damn thing easy to install and patch.

[swestrup]

Well, making things easy to install and patch is a forte of mine. Massive distributed compute-environment designs is a forte of _sps_ who I'd hope to rope into the design of this puppy.

One thing I DO fear is that this thing will have a 10-gig footprint for a basic install by the time we're done. Then again, by the time its done, that might be considered reasonable.

[labyrinthman.livejournal.com]

10 gigs...as long as you aren't the /first/ to that point, you are laughing. (From a user perspective here.) From reading the game pages, whenever some game manages the first time tor break size-x, it gets panned for that.
Already getting close. My install of Final Fantasy XI takes up a solid 6 gigs, and more with every update.

As for the money...well, I really like the real-world advertising feature. Who wouldn't want their Sims to be able to drive to work in a Jaguar? Or have a "furniture assembly skill" that applies only to funky furniture from Ikea? (Swedish for: Where the fark did I put the last 2 screws?) Hell, hook up with SMED and have their furniture in offices, and have design as a career path.

May I link to this, by the way? Its a fascinating concept, and I tend to have a few gamers and open source advocates on my list that could prolly lend better quality remarks than I.

[swestrup]

Sure. I'm not one to post secret info in a public domain and expect no one to see it :-) I'm currently building a website which will eventually host a cleaned up version of this and a number of other proposals with, I hope, similar technical merit. In the meantime, feel free to point anyone interested to this place. Feel especially free to point it out to any investors who may want to get in on the ground floor of this!

[tjernobyl.livejournal.com]

The only problem that I can see is the possibility of decompiling and rendering copyable objects-- the distribution mechanisms are secure, but the template will need to be decrypted to be rendered/displayed. Knockoff Ferraris!

[swestrup]

As in all cryptography, we're gonna have to be DAMN sure we've dotted all of our i's and crossed our t's for it to be secure, especially since the encryption protocols and routines will all be open. All I can say is that preliminary analysis makes it look feasible, but we'll need a full design before we can be sure we haven't overlooked something crucial. At the moment it looks like someone wont be able to copy the contents of an object without installing it, and that will (if done right) create an electronic trail that can track them down later. If they're cracking templates to give them away, there may not be anything we can do except sue. If they're selling them, then the money transactions will go through us before being converted to hard currency, and we'll be able to nix the transaction.

So, yeah, you have a legit concern and we plan to address it by one means or another. I'll let you know if we actually come up with a design document for this system, and you can check it for errors yourself. In fact, I would appreciate it.

[sps.livejournal.com]

Yeah, the situation differs from digital cash in that (a) all useful fields of the data object are exposed in normal use, and (b) the average end user has no reason to prefer the authentic object over forgeries. The only way I can see around this is to have in-game challenges of some sort. But even then, all participants would have to want to.

[swestrup]

I've posted a better thought-out reply to your concern. It can be found here.

[thothmeister.livejournal.com]

An opensource Sims like game would be interesting. Obviously, you'd make it different enough that Maxis wouldn't come after you for copyright infringement...

I think the end product would have to take up less than 10 gigs of space though. Granted, games like FF come close but the original Sims takes up much less space, and that would be a potential 'selling' point (quotes since it wouldn't cost to get it).

I think the most complex thing for you to program will be the logic behind the characters actions.

[sps.livejournal.com]

I don't think you'd want it similar enough that Maxis could come after you. Would you?

I can't agree with your assessment of complexity, though. I think the hard parts will be the parametric skinning for good sizeable clothing, the ontology, and the secure mobile code mechanisms. Or ... maybe those are just the bits I find interesting enough to worry about.

Oh, wait, I forgot about generalisation. Yup, the hard part of the AI is implementing cognitive quantification, which is going to interface the extensible ontology with planning.