First off, you have to deal with the time line involved. It may take months or years before the right candidate in your price range comes along.

Furthermore, if the company is tied up in any kind of bankruptcy proceedings, it could take more months and years to settle that and you're not going to get your hands on the source code until it's settled.

Furthermore, you have no guarantee as to how good or bad the engine's source code actually is.

I, personally, don't understand the need to reinvent the wheel when there are some pretty good wheels already out there. I can think of at least two off the top of my head that could, with a small kick in the pants, easily be what you seem to be looking for. They are Delta3D and CrystalSpace.

I, personally, don't understand the need to reinvent the wheel when there are some pretty good wheels already out there. I can think of at least two off the top of my head that could, with a small kick in the pants, easily be what you seem to be looking for. They are Delta3D and CrystalSpace. As a manager of a development team in real life, it's my opinion that you could hire a couple of coders and "fix" an existing OS project in less time and less money than you were looking at spending on Ryzom.

As I understand, it isn't really just a matter of finding a good 3D engine. From what I gather, there are some very good ones out there. (In addition to the ones you mention, the OGRE engine is supposed to be very good -- Worldforge is using it as the basis for their client)

as I understand it it must be GPL compatible.

What would be GPL compatible?

I mean, what would be necessary for the media to be GPL compatible? How would we go about this?

BuildAWorld.net is moving forward with the creation of a game engine, and we have a little art with our graphics engine (Delta3D has an art pack) plus WorldForge is contributing their art into a common repository. We found a few other places with art, and we have recruited half a dozen or so artists so far.

BuildAWorld.net is moving forward with the creation of a game engine, and we have a little art with our graphics engine (Delta3D has an art pack) plus WorldForge is contributing their art into a common repository. We found a few other places with art, and we have recruited half a dozen or so artists so far.

While it may behoove us to buy art or a game engine, there are enough talented people and people who learn quickly that can make these things. Yes, a lot of money can just hire people and they will start writing code 40 hours per week. Money does talk, but other revenue systems may work better. I don't think buying time brings out passionate people well. I don't think it produces the best product. It also may not be the best value per dollar. In many cases, it has a longer time line.

If we do have $250k to throw down, I'd rather have contests with rewards. Let us say we start with a landscape. We want a forest scene, and the top 5 choices get paid and all submissions go into our library of usable art (creative commons required). We hold a similar contest for more landscapes, and buildings in different genre. Our top 5 rated buildings each earn a commission to make a set of buildings (10-30 different buildings).

Similar contests can be done for weapons, animations, different races, clothing, and definitely player models. Since we want to build many worlds, we will benefit from having many different kinds of art.

Actual server code is a little harder to run as a contest, but specific tasks like Il8N can be set up with a goal system. Whoever achieves the goal gets paid. the more fun work will probably be done without money, but there is a ton of work that few people would want to do. Richard Stallman ended up doing a lot of really grunt work to make GNU work (like the library linker) and this annoying, hard or unsexy work might be eligible for pay. If reasonable progress towards a goal isn't shown, then we would allow multiple people to sign up for a single task. Depending on overall % completed they might split the pay.

I think that might be a better use of money than buying a pile of content. Granted a pile of content should work with the other pieces. On the other hand, the pieces might not have full source ("Fred didn't check that into the file server so we can't edit buildings."), may not work with any engine changes we want, etc.

I'm concerned that we'll have chibi japanese anime meets Frankenstein's monster or something.

How do my AI contributions stack up against someone making jars for 10 hours?

I agree that people will aggregate. I do think that randomly contributed art will have serious gaps. I also think that there are certain base standards that need to apply for art in a single scene to look right. Shades, color tone, crispness, etc all immediately come to mind. Having a potpourri of art would look... interesting. I think it would be more useful to have themes, and ask for contributions that blend in a theme. Art is easy to get. Good art that blends well...

Thats always an issue with community art projects - you can end up with a mishmashing of clashing styles. Often from my expirence its better for small groups to work on an entire town or map together. This keeps each section much closer in style. Another option is an art kit with a bunch of house models, a few textures for each that can be switched, and thus once created can be reused over and over for new unique places with identical styles.

Often from my expirence its better for small groups to work on an entire town or map together. This keeps each section much closer in style. Another option is an art kit with a bunch of house models, a few textures for each that can be switched, and thus once created can be reused over and over for new unique places with identical styles.