An open mind on Open Source
1 April 2005 | 0
Clichéd convention has it that when meeting a total stranger at a prearranged spot, one or other of you should be wearing a red carnation in their lapel to aid recognition. Hopefully there will not be another pair of strangers in the same place at the same time hoping to meet up by the same method or matters could get confusing.
Rendezvousing with Red Hat Software’s Michael Tiemann in a busy Dublin hotel presented no such problems. He could only be the gentleman with the red fedora!
Tiemann, Red Hat’s vice president in charge of Open Source affairs was in Dublin at the start of a busy day of talks with Irish politicians from several major parties. Usually, when the representative of a US software company meets with Irish politicians, the topic for discussion is: ‘Where are you building your factory/research lab/localisation centre? How many jobs will it create and how much help can we give you?’
But Tiemann’s purpose was different. He was here to make the case for Open Source software as a serious idea, worthy of government support and to gauge the attitude of Irish politicians to related issues. ‘I want to understand their perceptions about technology and competitiveness,’ he said.
Two matters of particular interest to him are governments’ attitudes to Open Source software and also the issue of software patents, which is currently the subject of debate at European level as a bill to harmonise patent regulation among EU member countries is proceeding through the legislative process.
In the first instance, the topic of Open Source software has moved gently into the political arena here, mainly as a result of comments made by the Minister with responsibility for the Information Society, Mary Hanafin, at the opening of the Irish Software Association’s annual conference this year, when she said that: ‘the long-term costs associated with Open Source may outweigh the short-term benefits.’ She added: ‘From a public procurement point of view,
the search for value is critical. The assessment of value must, of course, take full account of the totality of costs likely to arise over the appropriate time horizon for the project concerned. In that context, confidence in the performance of software may be a relevant consideration.’
Tiemann points out that the Open Source model has been adopted
enthusiastically in South East Asia, where it is perceived as ‘offering the countries there an opportunity to solve the competitiveness problems they have.’
He argues that researchers in Irish universities can make the best use of processing power and computer technology if they harness their available infrastructure to ‘a platform that is freely available’, namely Open Source software. ‘Open Source is an “equal opportunities” technology’, he points out.
It levels the playing field for researchers in universities who may not be able to afford the huge investments in expensive equipment that their rivals can make.
Tiemann’s experiences elsewhere have led him to conclude that European politicians are ‘moving in two different directions and at two different levels’ with regard to their attitude to patents. ‘Where they are responsible to their individual countries, they do not want to blindly accept US patent legislation,’ he said. ‘But at the European level, when they are talking to the US, it seems to be a case of “What can we do to help you?” But it’s quite clear that the individual countries do not want a repeat of the US patent problems.’
The patent problem
The ‘US patent problem’, which not everyone in the software industry would regard as a problem at all, refers to the ease with which software can be patented on the other side of the Atlantic. Currently, most European countries do not allow patents to be awarded for software, but the degree to which the new harmonised rules in the EU will allow software to be patented is a matter
of some controversy.
As things currently stand, the wording of the proposed legislation is broadly favoured by those who disapprove of patents being awarded for software.
However, there is some way to go before the legislation is finalised and matters could well swing back in favour of the supporters of intellectual-property rights.
Open Source software companies, like Red Hat, are among those who regard the state of affairs in the US as a ‘problem.’ Chief among software companies who hold the opposite view are Microsoft and SCO, the company that claims to own the core intellectual property of the Unix operating system.
It is a matter of some irony, then, that Microsoft has fallen foul of patent legislation in a number of court cases. ‘The company in most trouble in the courts right now with regard to patent problems is Microsoft,’ said Tiemann, referring to cases involving Microsoft’s battles with companies like Eolas, Stac Electronics and TypeWright Keyboards, all of whom claim that Microsoft has infringed their patents.
Tiemann is of the view that there may not be any malice in Microsoft’s actions. ‘Let’s give them the benefit of the doubt and say that they are not wilfully infringing other people’s patents,’ he said. ‘It’s the very broadness of the patent that is causing even Microsoft engineers to do something during development that inevitably infringes somebody else’s patents.’ If Microsoft, with its resources, can’t develop its code without stepping on somebody else’s toes, the argument goes, what chance has the rest of the industry?
Tiemann is also of the view that the manner in which software patents have been put to use in the US has been counter-productive, in practice, to the whole purpose for having patents in the first place. Patents are intended to reward innovation by giving inventors of products or methods deemed to be ‘useful, new and nonobvious’ a monopoly on the intellectual property underpinning their inventions for a limited period of time. The theory goes
that by rewarding such innovation, individuals and companies will be
incentivised to produce more innovative products, while technology will advance and propagate and be of more use to a greater number of people.
However, according to a report carried out this year by the Federal Reserve Bank of Philadelphia, there is evidence that the reverse is true in the case of software. It found that in general, the rate of patent issuance over the past 25 years in all industries had remained fairly constant, as had the overall rate of investment in R&D (research and development).
But in the case of software, the rate of patent issuance has accelerated over the past 20 years. The report estimated that in the early 1980s, about 1,000 software patents were granted each year on average. By 1990, this had increased to 5,000 per year and by 1996 the rate had doubled again. By 2002, the report said, nearly 25,000 software patents were granted. The report pointed out that overall, patent issuance had nearly doubled between 1981 and
2001—but this rate of growth is positively sluggish when compared with that of the software industry.
The report also suggested that instead of encouraging investment in R&D, an increased rate of patent issuance was associated with a reduction in the level or R&D. The report says: ‘During the late 1980s, firms that increased their focus on software patents tended to increase their R&D intensity, but the
relationship was weak…For the 1990s we found a much stronger relationship but it was negative—all else equal, increases in software patent share were associated with decreases in research intensity.’
The report admits that ‘the analysis does not identify the exact relationship that explains why an increase in software patent focus is associated with a decline in research intensity. In the language of statistics, this approach identifies a correlation but not causation.’ It also points out that the majority of software patents are awarded to companies outside the software industry itself, mainly to manufacturing companies.
Tiemann also highlights another danger, namely that in the US, universities are now being encouraged to concentrate on producing technology they can license rather than engaging in ‘what if?’ research. But he claims that for every dollar an American university generates in licence revenue, it spends $40 dollars in establishing its rights to a licence. Most of this goes on legal expenditure and other associated activities. This, argues Tiemann is a very inefficient means of revenue generation and is even less efficient than playing
a lottery. ‘Most state lotteries pay out more than one dollar for every forty they take in,’ he says.
Essentially, he says, the situation in the US is that ‘the oxygen is being pulled out of innovation because it is becoming impossible to do anything without talking to a lot of people to make sure that you are not infringing any body else’s intellectual property.’ His hope is that Europe does not adopt a similar model, if only for his belief that diversity produces a more dynamic environment for innovation and business than homogeneity.
‘Diversity is important,’ he said. ‘For example, all around the world there are different legal systems. Why don’t we all adopt the same one? The fact is that there are some situations where a different environment might be of benefit to a company in a specific situation. Likewise, there may be something somewhere that would benefit from a patent US style, whereas in other situations an invention would be of more use under the European system.’
At the end of the day, he says, no software company sells its products based on its patents. It does so to enable its customers derive benefits from it. As such the Open Source model is worthy of investigation and perhaps support by politicians. ‘It would be my hope,’ he said, ‘that before any politician took a stand against Open Source software that they would look at the successful projects that have used it.’
I’m sure he made his case well.