Skip to main content

Front Lines is a section of globetechnology.com reserved for guest viewpoints offering perspectives on current issues and events from people working on the front lines of Canada's information technology industry. The author is a writer and corporate communications consultant who has worked in the Canadian high-tech industry for more than 12 years. He is the founder of The Pangea Partnership, which assists environmentally distressed communities in the developing world.

If you watch the tech news closely, as I do, you may have noticed over the past couple of years that governments around the world are moving away from proprietary software platforms and toward 'open-source' solutions such as Linux.

Germany, Brazil, China, the European Union, Mexico and recently, India, are among the many jurisdictions that have introduced initiatives to promote 'free software' solutions in government IT. Many countries are developing their own versions of Linux and other open source software in an effort to wrest control over their computing environments away from American-owned monopolies. At the same time, they want to make better use of limited resources and encourage home-grown software industries.

But there are other, more profound reasons for this shift involving issues at the very core of a free political system. Many governments have come to see proprietary software as a threat to their democratic institutions and have started crafting legislation to meet the challenge.

Notably, in 2001 the government of Peru introduced Bill 1609, Free Software in Public Administration. Bill 1609 proposes that where a choice exists between open and closed source software to perform the same function in government, that open source be used. In open debate with Microsoft and other vendors, the legislation's chief author, Senator Edgar David Villanueva Nunez, has intelligently articulated the bill's reasoning and dismantled, examined and discarded Microsoft-sponsored fear, uncertainty and doubt about the legislation. He brilliantly argues that the core concern of the bill is the distinction between open and closed source software and its implications for control over data handled by governments.

Programmers write software by typing instructions known as 'source-code' into sophisticated text-editors, and the code they write is saved in simple text files that can be opened in any standard editor. But this text format is designed to be easy for a trained human to understand, not a computer. Computer processors, by comparison, are manipulated using instructions in a format called 'machine language' which turn individual circuits on and off in the processor to manipulate a user's data. To be useful, source code must be translated into machine language, and the resulting executable files are completely unreadable by humans. Typically, only the machine language version of a commercial program is shipped to customers.

Such 'closed source' software products are often called black-boxes because the user can't 'see inside' the program to review how it is manipulating their data. When we use the vast majority of proprietary software that follow this formula, we put data in one end of the box and hope that we get the result that we wanted from the other end. We have no way of knowing what happened inside.

Governments like that of Peru see this loss of control over data as a fundamental threat to the integrity of a free society. Every day governments handle huge amounts of information that do not belong to them. Held in trust on behalf of its citizens, the State manages and manipulates everything from tax and social benefits data to driving records. While this information is essential to the functioning and regulation of society, it is entrusted to the State as part of a fundamental set of covenants between it and its citizens.

In return for the measure of sovereignty surrendered by the citizen with this data, the State undertakes to make information freely available to those it describes, while ensuring its protection from those who have no right to it. For instance, we all have the right to examine our own tax data, but others do not.

The State also undertakes to ensure the permanence of public data. There is a multitude of practical reasons for this. Land title records, for instance, go back hundreds of years, and questions about the completeness and accuracy of this information would have an immediate and detrimental impact on the confidence of real estate markets. But there are also basic reasons that the permanence of the public record is important in a free society. As George Orwell famously pointed out "He who controls the past controls the future. Who controls the present controls the past." Computerized data is part of the public record, and the transparency, completeness and permanence of that record must satisfy citizens and historians that the past in not being corrupted to conveniently justify present policy.

Another basic guarantee made by any democratic state to its citizens is that of security, which includes not only the physical security of the person, but necessarily extends to the secure handling of information held in trust about the citizen. Mishandling of, say, the current address of an abused wife in the protection of government services can lead directly to dire physical consequences. We all have a right to these guarantees that our personal information will not be exposed to other individuals or groups. The State has a further obligation to ensure the security of information held for the common good, even, in certain circumstances, by withholding it from individual citizens for the collective security of the nation.

Villanueva argues that closed source software, which is to say the vast majority of commercial products, cannot meet these tests of openness, permanence and security.

Take the example of an electronic voting system. It must allow the voter to freely examine how his or her vote is processed, thus assuring accuracy and fairness in the process. In countries where elections are questionable dealings, international observers often provide exactly this scrutiny, and in our own mature democracies, it is a right we should always hold in reserve. Further, the freedom to vote one's conscience derives from the anonymity provided by the system, electronic or otherwise, and security of the voter against intimidation is essential. Finally, electronic votes must remain perpetually and permanently available to the citizenry for inspection, just as physical ballots are stored against possible recounts.

Only by allowing the implementers and users to inspect the source-code of such an electronic voting system could citizens be assured that their franchise is being protected. In the absence of such assurances, we may need to wait years, and hope that historians are able to uncover an election with rigged or erroneous results from permanent records.

Of course there are many other concerns that governments share with any user of computing systems. Closed source software vendors are often able to sell product based on marketing rather than technical merit. Service is expensive, closed source software is reputed to have more bugs generally than open software which take longer to fix, and proprietary data formats lock customers into relationships against their will. But into this mix are thrown a set of core issues that are central to the principles of a free society, and as such we are all stakeholders in this debate.

To date the Government of Canada has not stated a comprehensive position on open source software in its own IT strategy. I urge policy makers and all Canadians to consider these issues, and in particular, to read Senator Villanueva's letter to Microsoft dated April 8, 2002. After years of back-and-forth debate over the power of the software industry, Villanueva cuts through the hype to expose the intrinsic dangers posed to the trusts on which any free society is based.

Interact with The Globe