Research in robust open-domain text processing has seen considerable progress in the last couple of decades. It is probably fair to say that language technology tools have reached satisfactory performance at the level of syntactic processing. Therefore, it is timelier than ever to consider deep semantic processing as a serious task in wide-coverage natural language processing. This is a step that requires the integration of syntactic parsing, named entity recognition, anaphora resolution, thematic role labelling, word sense disambiguation with fine-grained semantic analysis. Accurate automatic semantic interpretation of text will benefit newly emerging sub-areas such as affectivity and sentiment analysis of texts, textual entailment, and consistency checking, and applications such as automated question answering, summarisation, and machine translation. This volume addresses these ambitions by presenting a collection of papers presented at the first workshop on the Semantics in Text Processing (STEP 2008), held in Venice from 22 to 24 September 2008. It is divided into three parts: (1) regular papers describing new results and completed research; (2) reports and descriptions of state-of-the-art systems that participated in the shared task on comparing semantic representations; and (3) short papers addressing ongoing work, novel techniques, or project descriptions. This is the first volume in \textit{Research in Computational Semantics} series launched by College Publications. Computational semantics is a relatively new interdisciplinary area in natural language processing, focusing on developing techniques to automate the interpretation of spoken and written natural language. It is an exciting area combining linguistic insight, logical reasoning, and knowledge engineering using both symbolic and statistical techniques to achieve robust and scalable methods for processing human languages.