Skip to Main Content
The UK government, for example, is firmly committed to making XML the basis for electronic transactions through their e-Government Interoperability Framework, e-Government Metadata Framework, and GovTalk program. Indeed, many UK government agencies are investigating the adoption of XML as the protocol for e-services, such as prescriptions, contracts, and personal health records. But although many have touted XML as a true e-business enabler, rarely does a technology suit in every IT scenario. In XML's case, the drawback is performance. XML is an uncompressed textual syntax that remains in human-readable form from creation to deletion. This characteristic can degrade performance because it takes time to construct and deconstruct the syntax, and the translation increases data-stream size, which in turn increases data transfer time. Motivated by this concern, we devised a series of tests to compare XML messages with equivalent messages written in Abstract Syntax Notation One with Basic Encoding Rules (ASN.1/BER). ASN.1 is a protocol specification language, first standardized in 1984. The encoding rules, of which BER is only one variety, are used to condense the ASN.1 textual representation into a binary data stream. The goal of our test was to gather data to inform the UK Department of Health, which has requested the use of XML for encoding electronic prescriptions. In short, we wanted to see if XML or ASN.1 was the more efficient mechanism. Our tests cover the creation, transmission, and retrieval performances of the two languages. The test bed was a trial electronic prescription system already in place that uses an application certificate to transmit a prescription.