An Empirical Study of Open Standards

Abstract

This study employs empirical data to provide insights into the impact of open standards. This work moves beyond the existing literature's reliance on hand-picked case studies by analyzing a very large number of open standards. The results of this research are timely as governments are advocating, and sometimes mandating, the use of open standards. Our study reveals inequalities in the impact of open standards that suggest a power-law relationship. We have also found that the duration of the development process in creating an open standard does not affect the ultimate impact of the standard, but that the length of the document describing the standard affects its impact.

There are two datasets used in our analysis. The analysis focuses on standards developed by the Internet Engineering Task Force (IETF) because it is held as the exemplar for developing open standards. The first set encompasses 3221 IETF publications from August 1987 to January 2006. The second set consists of 32 standards, selected randomly from 2000 to 2003. The variables of interest include information about the standard document, the authors, the development process, and the impact of the open standard.

The first part of the analysis found a power-law relationship in the impact of open standards. A few open standards have a very high disproportionally large impact, while there are many other standards that have a slight impact. This relationship should not be surprising. Standards are analogous to other decisions that fit the power-law relationship, because people can freely choose between many standards. This result has implications for improving the development process within standards organizations.

The second part of the analysis considers significant variables that affect the impact of open standards. Our results show that the length of the standard document (as measured by the number of words) is a crucial factor affecting the impact of a standard. The length of a standard often reflects the fact that there are multiple participants, divisiveness in opinions, and overall technical complexity, all of which suggest greater interest during the development process of the standard, likely resulting in a higher impact once the standard is promulgated. We also find that the duration of the development process does not affect the impact of an open standard. This finding carries significant policy implications as reforms are underway to shorten the IETF development process.