Even with all the exciting and innovative tools that MOOs offer, the "intelligent classroom" cannot intuit the kind of preparatory work that should take place with students prior to logging on to the MOO. Thus, the value of its features is directly proportional to the basic training necessary before bringing students online. Many teachers who use MOOs have learned the importance of introducing students to the rules of netiquette, the basic commands for talking and emoting, and the goals of the activities while online. Traci Gardner's helpsheet is a great resource for such preparation. In addition, it is important to provide good MOO clients (interfaces) to reduce the kind of text interruption that can be disconcerting for new users. For help on the various MOO clients available and information about their installation, see Nick Carbone's MOO client page.
While intelligent classrooms in MOOs are paving the way in the effort to reconceive conventional teaching and learning space, new problems arise in terms of conventional evaluation of student work. One of the most promising new models for evaluation of MOO activity is Peg Syverson's notion of the Online Learning Record. According to Syverson, the models for writing assessment are mired in evaluation of products rather than both products and activities. In her five-part schema, evaluation looks at growth according to confidence, skills, knowledge, use of experience, and reflectiveness. Teachers may record observations of students' interactions with the technology itself, the players, the MOO environment, and other programmed non-human features like 'bots.' In essence, teachers should prepare students (and themselves) before logging on, and they should create a history of students' online activities across a number of goals and criteria.