I happened to hear about Atmosphir from a Facebook comment a few weeks ago; quickly running over the site I thought it was a cute, quirky program and happily put my information into the beta signup section. Atmosphir, as the website explains, is a "video game" creation tool geared towards amateurs: levels are designed on a grid by placing blocks and tiles where you want them, so a square section of grass gives way to a tall cliff that needs to be scaled to find the monster you have to kill for the gem to complete the mission. Since there's no programming knowledge needed, anyone can pick it up and begin creating levels right away.
Well, at least in theory. While I've played with similar incarnations of programs like these before and am familiar with the concept, I found Atmosphir's editor to be less than intuitive. For example, the front page of the site bullet-points "Build customer adventures in minutes by simply stacking blocks on a grid"– simple, except that I couldn't for the life of me figure out how to do the stacking bit. Playing other levels it's obviously possible so I just wasn't getting it, but it seemed to shoot down the "in minutes" part of the equation. Another issue was that the editor had a menu bar with a lot of unexplained buttons; some of them gave a title when moused-over, but not all of them did (and generally those that didn't were the ones I was most curious about because they caused no quantifiable effect when fiddled with).
Despite only being in private beta, Atmosphir already has well-defined real-money trading (popular in many Asian MMOGs and newer Western ones like Free Realms) built-in. You buy a chunk of game currency– called atmos– on their website, which are then used to buy customizations for your avatar. The initial free options are pretty limiting and even with spending the 200 credits I was given I had a difficult time "customizing" myself into anything other than variations of a pouty teenage lesbian; I did buy a nice lightsaber though.
I can't find any mention as to whether users will be able to create their own items (like clothing or game editor tiles) and publish them online or even sell them to other users; while I don't expect this to be the case, it does make the difference between this being sort of a poor man's Little Big Planet or something like Second Life, where user generated content has literally made the platform. Considering the lackluster graphics in the beta right now, an economy driven by graphics designers pushing the system could easily improve the look of the platform. Until then, the only things players can create and publish are their levels, which are made available on a browser when you click the Play link from the main menu. You can sort them by the name, designer, the overall rating of the level and the number of times it's been played in case you want to find the levels everyone else likes the best. Running through a few of them quickly they ran the gambit from simple monster hack-and-slashes, mazes and puzzle levels and even pretty well thought out races. However, I did feel that the game environment suffers from the same unintuitiveness as the editor and many of the mechanics of the puzzle ones (even the tutorial level created by the developer) had little explanation on what the player should do, resulting in a lot of trial and error by comical avatar death.
In the end this is still only a private beta and for a cross-platform (Windows, Mac and Linux) tool it's already working respectively well. I did find that the Windows version was buggier than the Mac one, usually crashing any time I set it full screen for example. Hopefully with time the chunky bugginess of the player/editor can be improved. I think the editor is too stripped down to be anything more than a gimmick, however, and as long as content updates only come directly from the developers, players will essentially just be using the same pieces as everyone else. Innovation eventually hits a wall when you realize there's only so many ways to reuse the same brick tile.