p>I recently had the opportunity to interview Dr. Stephen Jenkins, a founding member of the Triple-S Group and, in his day job, managing director of Snap Surveys. He brought me up to date on the Triple-S Group’s work.
Q: What is the Triple-S standard?
A: The goal of the Triple-S standard is to define a means of transferring the key elements of entire surveys between different survey software packages across various hardware and software platforms.
Triple-S essentially describes the structure and format of two text files. One, the Metadata File, contains version and general information about the survey together with definitions of the survey variables. This is used to interpret the contents of the Data File.
Survey software and system authors then write functions to write and/or read files to the Triple-S specification to enable their users to transfer surveys to software from other vendors. Some, typically analysis systems, may only read Triple-S; some, typically data collection/interviewing tools may only write Triple-S; some systems may both read and write.
Q: If you are licensing survey software or subscribing to a SaaS survey application, why should you look for Triple-S support?
A: Triple-S was devised as a viable solution to problems inherent in the alternatives strategies:
- Every software vendor having their own specification of a survey, this complicating the export process for the user and for the developer (one export / one import per target system).
- A vendor having no way of importing / exporting a survey so the user being effectively locked in.
- There being a de facto commercial standard that everyone else has to follow.
A supplier providing Triple-S support has therefore recognized the need in the user community to provide a simple way of transferring surveys between systems.
To date there are over 100 implementations, in applications from 86 vendors.
Triple-S is based on XML and there is active support and advice available from the Triple-S Group.
Q: What pitfalls should you be aware of when evaluating a system’s Triple-S support?
A: Version compatibility: Great care is taken to ensure that wherever possible, new versions are supersets of old versions. Not all implementations support all versions of Triple-S, although there are free tools available from the Triple-S website (http://www.triple-s.org) to convert between versions.
Be aware also of the choice of features and options used by an exporter. For example if an exporter generates a hierarchical Triple-S then it will limit the number of importers that can process the data.
Q: For developers, how long does it take to add Triple-S support to their applications?
A: That depends on the complexity of the application and the synergy with the standard. Variable types, data representations, etc. may all need some work or could just fall through from the internal model.
Subject to the above, generating Triple-S is usually fairly straightforward because you only have to implement the options or features that you require. But processing Triple-S requires more thought because decisions have to be made on how to handle options and features that do not translate into the application.
Q: What does the future hold for the Triple-S standard?
A: We are currently working on a new version, which will see the light of day in 2015. It addresses some current problems by introducing formal support for Unicode datasets, and provides solutions to difficult problems such as the use of formatting in text strings and labels.
The difficulty from the design point of view is the usual one with any software development: coming up with something that is easy to implement for those with simple requirements but rich enough to represent objects and entities in those applications that need the extra scope and detail.