In exploring how to make programming easier for non-programmers, research into end-user environments has traditionally been concerned with designing better human-computer interaction. That traditional focus has left open the question of how end-user environments might support human-human interaction. Especially in situations in which end-user environments are enlisted to facilitate learning, we hypothesize that a key benefit may be their ability to mediate conversations about a domain of interest. In what ways might end-user environments support human communication, and what design features make them well-suited to do so? Drawing on ethnographic studies of an undergraduate algorithms course in which students constructed and presented algorithm visualizations, we develop a provisional framework of six communicative dimensions of end-user environments: programming salience, typeset fidelity, story content, modifiability, controllability, and referencability. To illustrate the design implications of these dimensions, we juxtapose conventional algorithm visualization technology with a prototype end-user environment specifically designed to facilitate communication about algorithms. By characterizing those aspects of end-user environments that impact social interaction, our framework provides an important extension to Green and Petre?s [1] cognitive dimensions.