The work presented in this document is part of the LIRICS project and performed in Work Package 4 in close collaboration with two related initiatives. First, the area of semantic content annotation has been recognized as important by the International Organization for Standardization (ISO), which has formed the Task Domain Group TDG 3, Semantic Content, as an international expert group devoted to work in this area, within Technical Committee 37, Subcommittee 4 (Language Resource Management). In addition, an independent scientific peer group was formed within the ACL Special Interest Group on Computational Semantics (ACL/SIGSEM), the Working Group on the Representation of Multimodal Semantic Information (MMSemR). The work in WP 4 of the LIRICS project is carried out in close collaboration with this Working Group and with ISO TC 37/SC 4/TDG 3, as witnessed by joint workshops that took place in January 2005 in Tilburg (The Netherlands), in April 2006 in Marina del Rey (California), and in January 2007 in Tilburg (The Netherlands). (See also http://let.uvt.nl/research/ti/iso-tdg3 and http://let.uvt.nl/research/ti/sigsem/wg.) Following on from the discussion of methodological factors in the approach to isolating semantic descriptors for representation as outlined in the LIRICS deliverable D4.1 (Bunt and Schiffrin, 2006), this document will deal with: (Section 2) Issues in identifying commonalities in alternative approaches to the annotation and representation of various types of semantic information, and how these issues may be addressed. (Section 3) A proposal for a set of data categories in each of the four areas of consideration for semantic annotation; namely: temporal, dialogue act, reference and semantic role annotation. (Section 4) Concluding remarks on the status of the proposed data categories, their application in the constructing of multilingual test suites (described in LIRICS Deliverable D4.4), and the role of these data categories in the fields of computational linguistics and language engineering.