Kind of following on the tracks of those who are interested in the marriage between the material and the digital, I want to talk about the ways in which we can use web standards to create the best digital variorum edition of a given work. I’m working with two sets of problems: The first is how to represent manuscripts and print editions as close as possible to the originals. I’m thinking of creating a tighter connection between CSS and TEI (ex. linking @rend to HTML representations), using XSLT to generate the stylesheet. The goal is to try to capture as much of the original bibliographical codes (non-textual elements like page layout, font, etc), whether these are determinant (as in experimental modernist texts) or not. The second goal is to visualize change over time in a compelling way. We have several tools that do this already, but I want to do this using well-represented texts. Perhaps do a Juxta-like comparison that takes into consideration the bibliographical codes. Originally Juxta was conceived as an analytical tool, but recent developments are transforming it into a representational-tool. Perhaps one of the conversations we can have is how DIFF software can use the TEI-CSS relationships I’m trying to build to make comparisons resemble having two paper versions side by side.