In participatory design, not only the designer, but other people such as end users participate in the process of designing to satisfy as many people as possible with the final product. The problem is, however, that interaction among people does not necessarily lead to good results. In our project we aim to make a design that reflects many people's intentions without interaction among them. We use rich text data from the web and computationally analyze the data to extract the intentions of people concerning the design. In this paper, we propose what data to use and how to analyze the data from a cognitive linguistic point of view. The points are that (i) we should use texts which are written spontaneously (such as those found in micro-blogs) and (ii) we should analyze the tense markers and epistemic modalities of the texts to extract intentions. We also introduce in the paper our workshop held at a library in an art university. In this workshop, we have visualized the intentions of students on digital maps.
