Advocacy for Open Source in Government

Author: Con Zymaris, <conz@cyber.com.au>
Date: 2002-10-27

The thoughts expressed here have had a multi-year gestation period, reaching a degree of cogency only in recent months, when a multitude of events, including the statement of intent to consider widespread deployments of open source technologies by close to 40 countries around the world, including the whole European Union, as well as an attempt (led my Microsoft no less) to stymie this brewing stampede, have caused me to consider this more completely.

This piece was written partly as a response to a column by Andrew Parsons in ZDNet, who wrote there recently about the misdirected push by sections of the our industry to push for Linux and Open Source adoption by government due to their zero licence cost. While much of Andrew's thesis can be quickly dismissed and countered, which I will do below, he is right on a couple of points. In his piece (Free source or free beer? source: http://www.zdnet.com.au/newstech/os/story/0,2000024997,20267851,00.htm) Andrew rhetorically asks if it's a good idea to opt for the cheapest technology (i.e., the one with no licence fee costs) to which he doesn't render a complete response. I'll provide an answer in lieu of his, with the following rumination: the freedom from price of any software should not be the most significant aspect for the selection of software for government. What is of absolute importance is a combination of open and inter-operable formats, APIs and standards, along with transparency of software. What's at stake is not merely or primarily a discontinuity of commercial interests. It's far more fundamental.

The core concept here is that government functions differently to commerce, and rightly so. Open government is good government. The code which runs under the machinery of government should also be open; open to inspection; open to review; open to discussion, somewhat mirroring the Freedom of Information legislation which permeates through most advanced democracies, and much like open contracts and publically reviewable government processes, which are a pre-requisite for the viable governance of democratic society. We must always remember that governments are, and should continue to be, different to profit-driven businesses. Democratically elected governments have a much higher calling and should therefore be measured against a much more stringent gnomon. What might suffice as acceptable practice and probity in business, is never enough for the leaders and purveyors of citizenry.

Let's expand on these basic tenets of open government computing. Firstly, open standards for intercommunication and file formats should be an absolute must for any system implemented for government. Document standards should be fully disclosed, and preferably implemented by a range of developers. This is for reasons of longevity and permanence of the government record. We must not have documents which have been created by technologies that only one supplier can provide. This means that while Microsoft Word should never be used to create or store documents, open and clearly defined standards such as PDF and OpenOffice.org are fine. Further, systems technology which only runs on platforms from any single vendor (for example Microsoft's .Net) should likewise never be used. Once again, this has to do with ensuring the longevity and permanence of applications which can manipulate the data formats used, as well as encapsulate the machinery of government. As an alternative, standards-based Web-services, which are clearly defined and ratified by a non-vendor body such as the W3C, and implemented by many suppliers, are fine. Another example: VB.Net should not be used as it's only available from one supplier. Java is fine, as once again, it's available from many suppliers, including compliant and free open source versions.

Secondly, there should be strong and viable competition amongst suppliers of platform and application software technology. Construing that the government have one, and only one supplier of desktop platform and productivity software, as it does now, is just not on. As indicated by the Sincere Choice project (http://www.sincerechoice.org/) software suppliers should compete fairly on the merit of their products, rather than by attempting to lock each other's products out of the market. At the risk of belabouring this point, open formats and platforms, particularly with Open Source reference implementations, assist greatly here.

Lastly, as a democratic society, we need to view the increasing adoption of the technologies which run counter to open government computing (i.e. closed file formats, platform specific technologies and the spread of technologies which have only one single, monopolistic supplier,) as alarming. Whilst this wasn't as big a problem 20 years ago with the minimal pervasiveness of computer technology, in 20 years from now, when e-government will be a fundamental component of many a democratic society, we will have serious and irreparable fractures unless the technology which deploys e-government in all its facets, is also open.

Note, no mention has been made of any financial basis for the adoption of Open Source software. This is a whole additional level and should form, as Andrew points out in his piece, a sideline issue. Having said that, Open Source is of great importance from a national economic perspective. By considering Open Source software systems (by way of import replacement for other expensive software systems) governments locally will not only be in line to save hundreds of millions of dollars each year in licence costs and help stem the current major trade deficit in IT, but also be at the forefront of this exciting next-wave now sweeping through our industry.  Further, by establishing this country as a center of excellence for the development of this type of software, we can play a more prominent part and gain financially from the services and support revenue this realm
fosters.

This combination of benefits haven't been lost on many functionaries of governments worldwide. Siegmar Mosdorf, German Secretary of State, voiced increasing support by the German Federal Government of Open Source and Linux software. Among others, he indicated the following reasons as to why governments worldwide should adopt Open Source software:

(Source: http://www.internetnews.com/intl-news/article/0,2171,6_408271,00.html)

Now, let's reach for the scalpel and bone-saw for some belated  deconstruction of Andrew Parson's arguments in his recent missive on this subject. On the proponents of Open Source for governments, Andrew states:

"They attempt to put forward the claim that to allow governments to use commercially supplied software somehow curtails their choice in the matter?"

Incorrect. The fact of the matter is that there should be a viable set of options for technologies which meet the open and inter-operable standards requirements, as discussed above, and from a variety of suppliers. This isn't happening at present. For example, if you seek information about the list of platform and technology suppliers shortlisted to offer desktop
operating system and office productivity software to governments locally, how many suppliers do you think will be on this list?

Andrew continues:

"Are these protagonists trying to state that these companies are not entitled to charge for their product? Are they not entitled to recoup some of their costs, and more importantly, as a BUSINESS are they not entitled to seek to make a profit so they can continue to BE a business?"

Incorrect. No one with a reasonable perspective could possibly expect that these suppliers should give their software away or to Open Source their code. That's never been on the agenda. What proponents of open government computing state is that government is a different realm, and unlike the commercial world, should mandate open and inter-operable document standards, platforms and communications technologies. If the closed-source vendors are able to meet these criteria with their products, then they too should be shortlisted for supply of technologies. It's then up to the individual purchasing officers to decide if the value proffered by these products is commensurate with the money asked for by these vendors, which must presumable therefore be above and beyond the quality or capabilities provided by the Open Source contenders to justify the extra cost. If so, fine. If not, default to the Open Source option. This is open and fair competition.

Finally we have:

"I challenge YOU, the one who says they're over-priced to produce a product yourself that rivals the commercially available products. I challenge YOU to then provide the infrastructure and support levels that are normally available with these so- called more expensive programs."

I think Andrew has not been following events in our industry over these past few years. In response, I would point to suppliers like IBM or SGI, both of whom implement open and inter-operable (in this case Linux and Samba) technologies on high-end, enterprise-class computing hardware platforms, and charge a lot of money to very satisfied government customers for the supply and support of these products. These are just a few examples of hundreds of similar cases. Now, while these are products from commercial vendors, let's see if they meet our open government computing requirements. Are the products inter-operable and open, and offered from many other suppliers in a similar configuration? You bet. Are the core technologies open source? You bet. Are the customers happy? You bet? Are we, the taxpayers saving money? You bet. Are there any issues with this? I don't know. Ask Andrew.

With our eyes on the horizon, it is becoming obvious, with the increasing importance of computer software in all manner of government function, that open and inter-operable computing, often best implemented as Open Source software libre, is the best software model which allows the IT-functioning of modern government to progress in a manner commensurate with the aims and motives of open and democratic society. This is an issue that you, the citizens of society, can help raise, discuss and bring to bear. Assuming that someone else will take up the challenge is a myopic stance. As with many other aspects of the body-politik, vigilance is the price for eternal freedom.

This piece was originally published on ZDNet.

Con Zymaris (conz@cyber.com.au) is the CEO of Cybersource Pty. Ltd. a
long-standing IT & Internet Professional Services company