Method
Analysed websites
In order to illustrate a model of analysis of digital genres that allows us to propose pedagogical criteria based upon the features of such genres, we will use two websites – one in English and one in French. These websites were selected according to the following criteria: type of discourse, multimodality, interactivity and hypertextuality, and adequacy for our students’ level of proficiency and interests. Regarding the type of discourse, the selected websites belong to popular science genres, that is, informative and explicative texts aimed at a non-specialist audience. We selected this type of discourse because the ultimate aim of the project reported here is to design materials based on hypermedia texts in the FL context. The students for whom these materials are intended are Engineering students and Philology students. We need therefore to focus on non-specialist texts whose content can be accessible and interesting for Philology students. The choice of the topics and the control of the level of difficulty of the text also answer to the need to provide appropriate materials for our students. Additionally, since we are interested in analysing “real digital genres”, i.e. digital genres that take full advantage of the multimodal and hypertextual affordances of the Internet, we selected websites that do not only include multisemiotic traits but also display a specific design of contents that responds to a rhizomatic-like structure. This will allow us to explore how the multimodal, interactive and hypertextual nature of the Internet can be harnessed to train students in multiliteracies. The selected websites are Automates Intelligents in French and Robot Pals in English.
Methodology and instruments
As pointed out in the introduction, in order to analyse digital genres it is necessary to extend the traditional text-based model of genre analysis and rely on a semiotic theory that can account for multimodal elements and that enables to study web texts as interactive media, where users perform actions and construct meaning throughout their trajectories across hypertext.
Therefore, in order to analyse and describe our websites, we followed three complementary and interrelated steps:
The heuristic description allowed us to formulate some tentative hypotheses to be contrasted with the results obtained in steps (ii) and (iii).
The analysis of the architecture of the website (i.e., the second step) was carried out with the help of a program that combined two instruments: Web Sphinx, which reveals which pages in the site are grouped together and more or less closely related through hyperlinks, and Yed Java Graph editor, which, taking as input the results from Web Sphinx, builds up the visual representation of the structure of the site, in terms of clusters of pages and links between the pages. The analysis of the hypertextuality of the websites was reflected by means of the IHMC Cmap tool, which helps to construct concept maps that represent hypertextual website models. In order to distinguish the different types of links in the resulting Cmaps, we employed a colour code: internal links were highlighted in light yellow, generic and semigeneric links in dark yellow, external links in blue and, finally, action and interaction links in pink.