Labor moves on Open Standards and Open Source Software for Government IT
And
Comments on Intellectual Property & the FTA
AUUG 2004 Conference
Melbourne
1 September 2004
The Labor Party has been vocal and active in promoting Open Source Software for use in the Federal Government Agencies and Departments. I have made many statements supporting Open Source and believe that Open Source Software has successfully challenged many assumptions government agencies have made about software vendors in the past.
How Governments manage their purchasing policies is directly relevant to the opportunity Open Source Software solutions have in winning Government contracts, and while I won’t be detailing Labor’s ICT government purchasing policy today, I do want to articulate Labor’s approach to encouraging the uptake of Open Source Software in Government.
Under Howard, Federal Agencies and Departments have discovered for themselves that the Howard’s Government’s mistaken approach of large scale outsourcing as a panacea for efficiency has not worked.
In contrast, Labor believes that there are real quality improvements and efficiencies to be gained through a strategic or selective approach to ICT sourcing. This type of sourcing has become more prevalent for Agencies and Departments not locked into long term contracts under Howard, particularly since the Howard Government was forced to abandon its IT Outsourcing Initiative.
Long term, vertically integrated, prescriptive IT contracts have stifled innovation and added to the cost of IT to the Commonwealth for the last eight years. Open Source Software represents one of the greatest opportunities to re-invigorate an innovative climate within the public sector. Many agencies and departments have already led the way in the absence of any policy direction from the Howard Government.
I note with interest that the third recent Minister for IT, Senator Coonan only this week finally issued a statement on Open Source in Government. This statement echoes many calls I have made about the need for agencies to pro-actively consider Open Source Software. It promises to promote successful solutions within the Commonwealth – something agencies have been doing word of mouth for a long time anyway in the absence of any leadership or policy direction.
I find it interesting that after many years of opportunity for the Federal Government to lead on Open Source, they wait until the first day of the formal election campaign to state the obvious: Open Source is a legitimate alternative to proprietary software, so check it out!
Labor will create an environment where the innovative use of Open Source Software is encouraged and explored. In particular, Labor will remove the many barriers that prevent smaller companies, open source vendors, from tendering and competing for Federal Government ICT contracts. I will be saying more about this in a few weeks time.
So let me share a few thoughts on the strengths of Open Source Software.
The freedom to access and modify OSS is its greatest strength. OSS allows users to build on, and refine software that has already been written.
The “open” nature of OSS tends to give it many positive attributes which are not necessarily present in proprietary software, in terms of reliability, security and cost efficiencies. It provides users with greater power in respect to software vendors, and has positive ramifications for local software development. Finally, OSS has specific relevance to the theory of democratic and open governance.
Faults in OSS are generally easy to identify and repair, but on the whole OSS has, over time, proven to be extremely reliable. In his paper Why Open Source/Free Software (OSS/FS)? Look at the Numbers!, David A. Wheeler notes several studies, undertaken from 1999 to the present day, which support this:
- GNU/Linux is more reliable than Windows NT, according to a 10-month ZDnet experiment.
ZDNet ran a 10-month test for reliability to compare Caldera Systems OpenLinux, Red Hat Linux, and Microsoft’s Windows NT Server 4.0 with Service Pack 3. All three used identical (single-CPU) hardware, and network requests were sent to each server in parallel for standard Internet, file, and print services. The result: NT crashed an average of once every six weeks, each taking about 30 minutes to fix. That’s not bad, but neither Linux server ever went down.
- GNU/Linux is more reliable than Windows NT, according to a one-year Bloor Research experiment.
Bloor Research had both operating systems running on relatively old Pentium machines. During the test, GNU/Linux crashed once due to a hardware fault (disk problems), which took 4 hours to fix, giving it a measured availability of 99.95 percent. Windows NT crashed 68 times, caused by hardware problems (disk), memory (26 times), file management (8 times), and various odd problems (33 times). All this took 65 hours to fix, giving an availability of 99.26 percent.
- A study by Reasoning found that the Linux kernel’s implementation of the TCP/IP Internet protocol stack had fewer defects than the equivalent stacks of several proprietary general-purpose operating systems, and equaled the best of the embedded operating systems.
- Another survey by Netcraft, keeps track of the 50 often-requested sites with the longest uptimes at the 30 June 2003 uptime report, 92% (46/50) of the sites use a version of the open source web server software known as Apache. Apache is a popular and reliable server tool. Netcraft data on market share for top servers for the Internet shows that Apache has been dominant for several years.
Obviously, a key factor in Government software acquisition is cost. Governments are accountable to the taxpayer, and must spend money as efficiently as possible.
A key benefit of OSS is that the purchaser does not pay royalties or licensing costs to software developers. These are payments for the right to use a proprietary product, over and above the cost of the work required to develop and maintain it. While this difference in cost might not necessarily result in a cheaper product, it could mean a more efficient expenditure of taxpayers’ money, as every dollar spent goes toward developing a good solution rather than paying for the right to keep using the software.
As a result, OSS potentially has lower life-cycle costs than equivalent proprietary systems. This is increasingly the case when proprietary software companies are moving to licensing schemes that expand their revenue, without necessarily an increase in product quality. Conversely, because no single company controls the intellectual property of OSS, expenditure on such software goes toward either technical support or improvements to the software solution itself.
Labor believes that there is a genuine opportunity to achieve some cost reductions through the increased use of Open Source Software within the Federal Government. Open Source Victoria, the industry body that has been successfully articulating the benefits of Open Source in the Australian market identifies “long term expense reduction” as one of a number of positive outcomes.
However, experience has shown that it is extremely difficult to assess the overall expenditure by the Commonwealth in relation to ICT in general, let alone break the expenditure down to ICT operating costs.
An ABS survey published in July 2004 puts Commonwealth ICT spending on ICT operating expenses at about $1.4 billion. To this end, Labor believes that a modest goal for reducing costs around 2.5% (or $35 m) on average per annum, over the next four years, attached to a policy of assertive promotion of the strengths of Open Source Software within the Commonwealth is worth pursuing.
I stress this cost reduction goal is not one that can be applied in an arbitrary fashion across all Government agencies and departments. Rather, the lessons of the previous debacles show that different Agencies and Departments have different needs and business outcomes to achieve. The nature of any potential cost reduction would be that they were realised later rather than sooner, as more open source software was adopted.
For example, many small agencies are already using Open Source Software or have satisfactory contracts with proprietary software firms and would have little scope for any cost reductions at all. Conversely, many large agencies and departments are just beginning to explore the potential of Open Source Software within their suite of ICT needs, for which any cost efficiencies are yet to be realised. Labor believes that these will come, hence taking the opportunity to flag the potential for a saving. But there are other reasons as well.
‘Lock in’ occurs when an agency or department becomes dependent on a single proprietary software provider. The agency may often stay with one software provider even when alternative systems or software might be more suitable for their needs, as the cost of transition may be greater than the benefit of new software.
Conversely, a purchaser of OSS is not dependent on the original supplier to upgrade, modify or service the solution. If the original supplier is unable to continue to look after the agency’s needs, they could transfer their business needs to a new supplier.
OSS creates an open standard on which anyone can develop compatible products. This gives freedom to the user to select different vendors, which in turn adds competitiveness, increasing flexibility and innovation, responsiveness to the changing needs of Government as well as improving the cost-efficiencies of the solution.
Another topical issue is the security credentials of OSS. The argument from OSS advocates is that while open source and proprietary software are equally likely to contain security holes, if the source code is open, these holes are more likely to be discovered, and a repair developed, which users of the software can then apply.
A recent parliamentary Joint Public Accounts and Audit Committee Inquiry into the Management and Integrity of Electronic Information in the Commonwealth urged “agencies to consider the benefits or otherwise of using OSS or closed source software as a normal part of their IT risk management processes and their cost/benefit analysis of new resources.”[3]
In the United States the National Security Agency is using an in-house (but open-source) version of Linux for some of its work, and a report to the US Department of Defence has warned against banning OSS in that department.
OSS also fits well with the principle of open government. In the information age, citizens should be able to scrutinise both public information, and how it is processed.
As OSS code is open, interested citizens would be able to see how information systems perform vital pieces of information processing – such as tax collection, the provision of social services, and, in the future, electronic voting. For example, the Australian Capital Territory employed OSS to ensure that the computerised voting process in the 2001 Election was open.
Open source can also ensure that archived electronic files are still accessible into the future, by employing open file formats for electronic documents, which do not require a specific proprietary platform (which may be long disused) to access them. It is a credit to the Australian Archive that they are world leaders in this field with their OSS system.
OSS can have a positive effect on industry development because it allows small-to-medium sized software firms to work on state-of-the-art technology as a starting point, rather than having to start from scratch. Hence, in response to a Government tender, they can provide a product of comparable quality to more established enterprises, subject to their own resources and capacity for innovation.
Improved access to Government ICT markets by Australian enterprises would benefit local industry development, jobs growth and contribute to curbing the rapid growth of Australia’s $14.4 ICT trade deficit.
Australia has a disproportionately high number of Open Source coders and innovators, providing an even greater opportunity for growth. It makes sense to build on this strength.
I hope this goes some way to clarifying Labor’s approach to Open Source and the opportunities that we see within wider adoption in the Commonwealth.
This issue of the potential industry development, is a useful segue into the second part of my presentation today which relates to some of the intellectual property provisions of the FTA concerning the software sector.
Comments on Intellectual Property and the FTA
The free trade agreement raised many issues about the management of Intellectual property and its impact on software developers in Australia. In particular, the open source community expressed concerns about the possibility of expansion of the US software and business process patents regime to Australia. Another area of concern was the adoption of provisions mirroring the US’s Digital Millennium Copyright Act.
For the remainder of my presentation I want to focus on these issues and the strategies that Labor has identified to ensure the stated purpose of the anti-piracy provisions of the FTA are what we enshrine in this country, not the plethora of unintended consequences that we now know are being exploited in the US.
Under the free trade agreement with the US, Australia has two years to amend our legislation to reflect the commitment given as part of the FTA. So it a timely exercise to begin discussing the implications of this.
The Senate Select Committee’s Report on the FTA gives Australia a good head start as to where the problems are. The intellectual property chapter of this Report has page after page of evidence about the concerns in the IT sector, the legal profession and academia. Labor Senators have used this evidence to inform the recommendations they prepared arising out of this chapter.
I turn first to the issue of anti-circumvention devices. The law that governs this in the US is the Digital Millennium Copyright Act, or DMCA, which was claimed to be about stopping piracy. The FTA requires us to adopt similar provisions, so it is important to assess whether it has met this objective.
I ask these questions:
- Have there been unintended consequences arising from the DMCA?
- Is the same thing likely to happen here and what can we do to ensure it doesn’t?
I will respond to these questions, but first want to turn to the issue of harmonisation per se. The Agreement states that harmonisation of IP law is to be an outcome.
This has been interpreted as Australia being required to harmonise our laws with those in the US. However, when challenged as to what this meant precisely, the formal response from the Government negotiator, Mr Stephen Deady in the Senate Inquiry was
“Under the terms of the Agreement we have commitments which, whilst they require some legislative changes, still provide Australia with sufficient flexibility in these areas to ensure the very high level of intellectual property protection that we already have and balance that against consumer interest and other things.”
In other words, the general goal of harmonisation does not necessarily require legislative change, nor does it dictate the details of any legislative change considered necessary.
So what does this mean? Does it mean that current Australian IP law has as much clout under the terms of the Agreement to build a case for legislative amendment in the US, in pursuit of harmonisation? This is easy to scoff at, but it may well be the key to taking a responsible approach and push for the legislative reform needed in the US to remove the more serious unintended consequences of the DMCA, as well as save Australia from the same fate.
This is a lofty goal, but hey, we punch above our weight in software innovation so why no in the laws that permit it to thrive?
In addition, there are already strong challenges to the DMCA US and the potential for legislative reform, so how ironic it would be if the US obligations under the FTA to Australia prevented this from proceeding?
The world is a rapidly changing place. Various national and corporate interests are lining up to best position themselves in the 21st century. It will be an era where those countries with cultural and economic strength will be able to dominate in cultural content like music and movies, and the essential tools that enable communication, commerce, administration and distribution: software.
So this lofty goal dictates that we must take what we know from the operation of intellectual property law in the US and let it inform what is good and bad for innovation. This lofty goal then dictates that we form a view as to what is good about US and Australian IP law and enhance it, and take was is bad and fix it.
A predictable political response would be to try to advantage Australia’s interests as much as possible through this process. However, I contend that these issues go beyond just the pre-occupations of Australians versus the US interests. The guiding principles should be those that determine a dynamic and strong economy: innovation and competition.
These two desirable outcomes of intellectual property law represent the test that we ought to subject the DMCA to.
These two desirable outcomes of innovation and competition lie at the heart of the concept of IP law which was to create an incentive for inventors, giving them some control over their work. The laws were designed to prevent monopolies by making public important technological innovations, but providing inventors with temporary exclusive rights over their creation as a trade -off.
The underlying assumption was that innovation best occurred with free-flowing ideas in the public domain. If ideas are prevented from entering the public domain in perpetuity, or there is a profound disincentive for research into new ideas, then something is very wrong.
What is at stake here is the very heart of innovation, and not just in software and information technology, but all areas of creative endeavour and innovative invention.
This sound very dramatic, but the problems identified with the DMCA in the US have been damaging to both innovation and competition. This means they are potentially damaging in Australia as well. Only in Australia the companies exploiting the unintended consequences are likely to be big US companies, at least to begin with, and the victims are likely to be small local companies.
Let me turn to the first question:
- Have there been unintended consequences arising from the DMCA?
The answer to this question has to be yes. There are many documented examples of how the DMCA has been used as an anti-competitive tool. Event he potential for the threat of litigation under the DMCA has had what has been described as a ‘chilling effect on research, development, innovation and competition.
This is despite the stated purpose of the DMCA as a specific anti-piracy bill. Congress believed it would stop copyright pirates from defeating anti-piracy protections added to copyrighted works. Congress also believed it would ban black box devices intended for that purpose.
Several of the higher profile cases of DMCA litigation saw corporations target students, researchers and academics. The arrest of Dimitry Sklyarov, Professor Felten versus the SDMI and Blackboard using the DMCA to issue cease and desist notices to students presenting at a conference. The inevitable effect is that these people are intimidated by the use of the law and are forced or choose to abandon their endeavour. This is the chilling effect. And there are dozens of other quotable cases and first hand testimonies. It is no wonder there is a real agenda designed to rectify some of the really insidious effects of the DMCA!
So it is a fair conclusion for us to come to is that were laws analogous to the DMCA enacted in Australia, then they would be equally exploited.
In addition, it is also important to assess the legislative context in Australia. For example, Australia does not have a fair use regime. This would mean that the effect of the DMCA, particularly in relation to consumers private use of copyright-protected material, would be that much more disempowering for consumers and therefore create a worse scenario than is the case in the US.
I would like to use an analogy to illustrate how the DMCA works in practice, and why it is problematic.
Say I buy a nice couch with stylish fabric covering the upholstery and frame. As the proud new owner of this couch I get it home only to have my three year old son smear his vegemite sandwich on it. My first response it to try and sponge it off, but the lovely fabric surely needs a professional job. So I call up my regular couch and carpet cleaner and he turns up at the door. He inspects the couch and determines it all needs to be washed or risk creating a patch if it were spot cleaned.
My cleaner then informs me that he is not able to take the cover off the couch because it is clipped together using a technological protection device. If he takes it off, he will be breaking the law, because only the original creator is authorised to take the cover off to clean it. I am not talking about losing the warranty, I am talking about committing a criminal offence.
It turns out my couch cleaner knows someone who was threatened to be sued by this particular manufacturer for taking a cover off. He had to pay a massive royalty for the privilege, which put him out of business. It was that or be sued and lose his house as well. My cleaner can’t afford the royalty, let alone lawyers to defend himself against litigation.
So he leaves. I call the manufacturer and get told that I can have the cleaning job done at the same price only if I have it done every 12 months. Otherwise I pay more for a one-off service. And I have to wait three months, because they have other clients that are more important than vegemite fingerprints.
Do you get the picture? This example illustrates how the Technological Protection Measure (TPM) provisions of the DMCA create a barrier for independent service providers that are unrelated to the manufacturer, to compete for business servicing a piece of equipment.
This is one application of the law in the US. It is the expanding use of the law into secarios like this that have set alarm bells ringing, including in the Senate inquiry.
I return to the question I posed earlier.
- Is the same thing likely to happen here and what can we do to ensure it doesn’t?
The Senate Report recommendations offer ways to address these concerns by outlining a series of exceptions that Australia should pursue in the legislation required to fulfil our obligations in relation to TPMs under the FTA. While the wording of the FTA prevents blanket exceptions, it permits specific ones.
The Senate inquiry addressed the use of TPMs in relation to DVDs and the potential to create fair use provisions in Australia. The Senator recommended that the Commonwealth ensure an exception to TPMs to provide for the sale and distribution of legitimate audio, video, DVD and software items, regardless of the place of acquisition.
This aims to prevent the use of regional coding for the purposes of anti-competitive behavior. Departmental officials confirmed that they did not believe that the sale of region-free DVD players would contravene the Agreement.
But the Senators were also aware of the other broader implications of TPMs and recommended specific exceptions that provide for the fair dealing or ‘fair use’ of copyright protected material to ensure that TPMs do not encroach into areas that would otherwise be legal, such as temporary copies, research and study and other legitimate private use of copyright protected material.
In acknowledging the specific software-related concerns, the Senators recommended exceptions to permit the manufacture of interoperable software.
In other words, the use of a range of exceptions are a way in which some of the more insidious effects may be tempered.
Another opportunity that present itself here is to define ‘circumvention’ so that it only applies to activities that would not otherwise be legal were a TPM not in place. This would mean that TPM’s would be able to fulfil their stated purpose of strengthening protection to prevent piracy, and at the same time prevent the measures from being used in a surreptitious way to extend the rights of copyright holders at the expense of consumers. Especially where there is an existing and established right under Australian law for use.
Yet another idea is to explore the concept of registering TPMs as a way to inhibit the retrospective claim of the presence of a TPM, which in the US has given rise to all sorts of litigation. Again, there have been a number of well documented cases of the way the DMCA has been used to this effect in the US, including Chamberlain Garages successful suing of rival remote control door opener manufacturer Skylink. Chamberlain claimed Skylink’s interoperable clicker violated the DMCA by bypassing an authentication regime between the Chamberlain remote opener and the mounted garage door receiver unit.
So perhaps registering TPMs would force companies to nominate them in advance and prevent companies from evoking the provisions only when confronted by a competitor. Registration could also present an opportunity to test a TPM’s legitimacy.
These ideas are not my own, but a sample of the lateral thinking that may help prevent the exploitation of law in its implementation in Australia, thus reducing the collateral damage. I encourage anyone with ideas to speak up.
In response to a specific recommendation of the Senate Inquiry into the FTA, Labor has committed to the establishment of anther Senate Select Committee in Intellectual Property to consider how we implement our obligations under the FTA into law within the next two years. This will present everyone with an opportunity to submit their views.
I would like to conclude this part of my presentation by revisiting the issue of harmonisation. Earlier I mentioned that the general goal of harmonisation contained in the FTA does not necessarily require legislative change, nor does it dictate the details of any legislative change considered necessary.
I should add that in relation to software patents, the Minister, in response to questions I asked in the Senate chamber during the debate, confirmed that no further amendments to the patents Act are required.
Harmonisation hence presents an opportunity for Australia to argue for the merit and validity of aspects of Australian IP law. It also presents an opportunity to argue for the unintended consequences of the DMCA to be specifically rejected.
In the meantime, Labor has supported the FTA and we hope to see its implemented in a way that best represents the public interest, interests which coincide with having a strong economy driven by innovation and competition.
In addition to the thirteen recommendation prepared by Labor Senators in the Inquiry, Labor has committed to the following:
- Monitor costs to Libraries and public research institutions as a result of the copyright term extension and consider exempting them;
- Expand fair dealing provision to cover ‘fair use’ including time-shifting and space-shifting;
- Identified specific exceptions in relation to TPMs;
- Review the standard of originality of copyright with a view to raising it;
- Establish an IP Select Committee to consider changes needed to implement the FTA.
Thank you for your attention. There are many other aspects of the FTA that I have not touched upon today in the interests of time. Needlessly to say it is about time that intellectual property is attracting the attention it deserves in a country that aspires to be competitive in the 21st century.
I look forward to working with you to develop a constructive agenda to ensure that the public interest is served through the implementation of the remaining intellectual property obligations Australia has under the FTA.
David A. Wheeler in his paper Why Open Source/Free Software (OSS/FS)? Look at the Numbers! <http://www.dwheeler.com/oss_fs_why.html#reliability> accessed 01/07/03
Netcraft Web site<http://uptime.netcraft.com> accessed 01/07/03
Page 85, Report of the Inquiry into the Management and Integrity of Electronic Information in the Commonwealth.
