Digitally managing punctuation in the editions of medieval manuscripts is one of those issues that initially look like minor details, but later reveal themselves as a tangled web of problems spanning from computer science (how to represent punctuation signs?) to philology (what types of signs do exist?) through epistemology (is the processing of punctuation a mere technical transformation or a valuable part of the scholarship?). The aim of this paper is to address the theoretical aspects of these questions and their practical implications, providing a couple of solutions fitting the paradigms and the technologies of the TEI. The debate on how to deal with medieval punctuation is a long and still open one. Following Contini (1992), the interpretative edition of a manuscript is the translation of a historically attested system into another system. Accordingly, the philologist should recognize the punctuation system of the manuscript and convert it into a modern one. There are, however, no established universal methods for doing so; most of this work is left to the experience (and taste) of the scholar. In fact, editors often substitute the original punctuation with a modern one. This improves the text readability, but often leads to a loss of textual information. In this paper we describe how we dealt with the encoding and transformation of the punctuation signs in some German manuscripts of Marco Polo’s work. Technically, we implemented a set of general rules (as XSLT templates) and various ad-hoc exceptions (as descriptive instructions in XML attributes). In addition to this, we discuss the philological foundation of this method and, contextually, we address the topic of the transformation of a single original source into different transcriptions: from a “hyperdiplomatic” edition to an interpretative one, going through a spectrum of intermediate levels of normalization. We also reflect on the separation between transcription and analysis, as well as on the role of the editor when the edition is the output of a semi-automated process.