XML’s popularity has made it the prime standard for exchanging data on the Web. Its self-describing nature makes it more amenable for use in loosely coupled data-exchange systems, and the flexible semistructured data model behind it makes it natural as a format for integrating data from various sources. Currently hundreds of XML dialects exist designed for particular industries, applications and needs. As examples, it is worth to mention Genomic Sequence Annotations in XML (http://www.fruitfly.org/flyannot/format.html) and HL-7 ver. 3: electronic interchange of clinical, financial and administrative data in XML format (Health Level Seven, Version 3 Draft, http://www.hl7.org/) In the course of ongoing collaborative effort with New York University School of Medicine Sleep Disorders Center, Pro-Tech Services, Inc (www.pro-tech.com) has developed an XML dialect for describing PSG Data, tentatively named SleepXML. The results of that effort have been presented at the 18th Annual Meeting of the Associated Professional Sleep Societies, Philadelphia, June 9th, 2004 [1] SleepXML aims to provide a flexible, extensible, non-proprietary, open electronic format capable of describing polysomnographic data and present tools that could be used for the data transformation, archiving and exchange. Although several standards exist for PSG data interchange that deal adequately with raw data (most well known is EDF - European Data Format [2]), the processed PSG data, such as sleep stage scores, respiratory scores and other events, as well as reports are handled less well. All these data are necessary for a complete description of a polysomnogram, but no common language existed for such data exchange. There is variety of PSG event, score and report definitions, all of that being in some proprietary vendor formats. SleepXML intends to change this situation.