Distributed data processing for public health surveillance

Lazarus, Ross; Yih, Katherine; Platt, Richard

All data is transferred to the datacenter in the form of eXtensible Markup Language (XML) since this is a flexible machine-readable representation and is easy to integrate with PHINMS. We used the Python [9] language for the development of the distributed software package. This choice was partially motivated by the fact that Python is an open-source language and thus freely distributable, partly by our very positive experience with Python as a general purpose application development language, and partially because in our experience, Python can be installed, and applications reliably run without any change to source code, on all common operating systems (including Linux, Unix, Macintosh and Windows), making it easy for the datacenter to provide support for systems other than Windows PC's. It is also a language with extensive support for standards such as XML, and securely encrypted internet connections. In addition, our existing web infrastructure has been built with the open-source Zope [10] web application framework, which is written mostly in Python.