Home

Project outline

Final report

Press coverage

related research

people

 

Papers

Tension between OSS-divergence and user needs for interoperability [1]
Tineke M. Egyedi, Delft University of Technology
T.M.Egyedi@tbm.tudelft.nl
September 2002

The current core-problem of IT is incompatibility. The future of OSS should therefore be assessed in this light. That is, research should address the tension between the in-built incentive for OSS-divergence and user needs for convergence (i.e. IT- interoperability).

These past years the term Open Source Software (OSS) incidentally crops up in standards literature (e.g. West, 2001). Usually the Linux operating system is mentioned, as is the question whether Linux will become a de facto standard next to and/or in symbiosis with Microsoft Windows. It is tempting to draw a comparison between the former, a potential open source de facto standard, and the latter, a closed source de facto industry standard. From a compatibility standpoint, however, there is an important difference. For a closed source de facto industry standard implies (internal) compatibility; whereas, to pursue the example, if Linux were to become a de facto standard this would not automatically be the case. Indeed, although the market has its favourite, different Linux distributions have developed.

The relationship between OSS and interoperability is a dual one and is tied to OSS-transparency. On the one hand, the OSS-approach offers a certain degree of technical transparency (i.e. by means of open source code), which eases the development of compatible products. On the other hand, however, the OSS-approach makes it easier to introduce changes, and in this sense provides incentives for diversity (incompatibility). (NB: The early IPR-licenses used by the Free Software/OSS community, that is, the GNU public license and the BSD license, were specifically designed to facilitate improvements.)

At the moment, standardisation of Linux (i.e. convergence) is taking place under the auspices of the Linux Standards Base. The aim of this group is "to develop and promote a set of standards that will increase compatibility among Linux distributions and enable software applications to run on any compliant Linux system". A group of Linux- providers has ‘promised’ to take the outcome of this standards effort, i.e. Linux Standards Base 1.1, as a starting point for developing a common operating system called UnitedLinux (Weiss, 2002). In addition, the Free Standards Group, a non-profit group which pleas for standards in open source technologies has asked the Open Group, an organisation specialised in interoperability certification to develop a certification programme for the Linux standard (Broersma, 2002).

Table 1: Expected shifts in the Open Source Software- approach (extracted from Egyedi, 2002, Table 11)

OSS initiatives towards convergence are a matter of general interest. The survival of OSS-products – and the possibility for consumers to choose among software - depends on whether or not interoperability will be achieved (see Table 1). A main research issue regarding OSS developments in the near-future will therefore be:

How are the main actors (e.g. Open Source Software (OSS) community, IT users, and the closed source IT-providers) going to address the tension between one of the main assets of the OSS-approach – the possibility to adapt and update source code to specific requirements (divergence) - and the user’s need for compatibility (convergence)?

[1] This contribution largely draws on a report written for the Dutch ministry of Public Works and Water Management called “Standaardisatie Trendrapport: Oplossingsrichtingen voor problemen van IT-interoperabiliteit” (i.e. transl. ‘Standardisation Trends report: Directions for solving problems of IT-interoperability’) T.M. Egyedi, Delft, September 2002