Date: Mon, 1 Dec 2003 09:16:17 +0530 From: Subbiah Arunachalam Friends: Here is an article from The Hindu, one of India's leading English language dailies, that must be of interest to all of us working in the area of ICT enabled development. Best wishes. Arun Reddy, C. Rammanohar. "The free software option." The Hindu (22 Nov 2003). http://www.hindu.com/2003/11/22/stories/2003112200721000.htm The UNCTAD report suggests that there are many advantages to be had if the developing countries seriously explore the use of free or open source software as against the proprietary licence-to-use software. INDIA AS the back office capital of the world is the new mantra that is being chanted for the country's entry into the league of developed countries. The rapid growth of Business Process Outsourcing (BPO) contracts to India is behind this mantra. The more perceptive observers point out that while there are many high-profile BPO activities that have moved to India, only a very small proportion of the businesses that can be out-sourced have been contracted out, and, of that, an even smaller fraction has come to India. The larger question, however, is if India can really ignore manufacturing altogether and pin its future hopes so much on servicing the rest of the world. The answer is "no"; services like BPO can make an important contribution to India's GDP but they cannot replace the potential of manufacturing. The obsession with software exports in general and increasingly with the BPO segment, has had an unfortunate side-effect. "Information and Communication Technologies (ICT) for development," which was much talked about in the late 1990s, has now been reduced to an empty slogan. There are no doubt a number of e-governance initiatives in the States, but if we are talking about a serious attempt at using IT for development, then that effort is largely absent. Of course, the fundamental problem is that the basic infrastructure necessary for this purpose is not yet in place. In spite of a rapid increase in tele-density over the past 18 months, the telecom infrastructure is woefully inadequate. And while internet use keeps growing rapidly, the population of internet users (estimated at 7 million in 2001) in India is far too small to speak about using phones, computers and the Net for development. Any serious attempt at harnessing the power of the new technologies must begin with expanding the infrastructure. Here, while there is at least an awareness of the gaps in telecom and the cost of computer hardware, there is very little acknowledgement of the demands that software in its various forms can make on users. A new United Nations publication, "E-commerce and Development Report 2003" of the United Nations Conference on Trade and Development offers important insights into the kind of software that developing countries can use if they are serious about bridging the "digital divide". The UNCTAD report suggests that there are many advantages to be had if the developing countries seriously explore the use of free or open source software (FOSS, also referred to as FLOSS, or Free Libre or Open Source Software) as against the proprietary licence-to-use software that now dominates the IT sector in the developing countries. Free software, as a popular slogan of the FOSS community describes it, is "`free' as in `free' speech and not as in `free' beer". What marks out FOSS is that because the code - the hundreds of thousands of lines of instructions that make up the software - is available to the larger computer programmer community, it is constantly improved upon, modified, tested and the problems removed. This is not possible in the use of proprietary software, where users are dependent on the licensors of the product. Users have no choice but to turn to the software "proprietors" for correcting the smallest flaws. Worldwide, the growth of open source software has been remarkable in recent years. But in spite of the fact that India is home to perhaps thousands of programmers who have contributed to the development of FOSS, whose best-known product is the Linux/GNU operating system, it is yet to be taken very seriously by either the Government or the corporates. Universities do use FOSS on a fairly widespread basis and the odd State Government has explored the possibility of using open source software, but this software is still not considered a reliable mainstream product. But reliable and mainstream FOSS already is. UNCTAD reports some statistics, which establish how deeply entrenched FOSS products have become in many areas. One, Apache, the open source software which runs on web servers, now has a market share of 63 per cent, more than twice that of Microsoft products. Two, GNU/Linux has a 30 per cent share in the operating system used on web servers, while the proprietary Windows has a 50 per cent market share. Three, 40 per cent of large U.S. corporations and 65 per cent of comparable Japanese firms are reported to use GNU/Linux in some form or the other. FOSS products have also proved themselves in the market. In a survey last August, the 20 web servers which had the longest uninterrupted uptime (without a breakdown) in the world were, in all but one case, running on FOSS operating systems. Since FOSS is non-proprietary software, the initial cost of acquisition is considerably less than proprietary software. However, while the saving on upfront costs can be a major incentive in developing countries, this is not the only cost to be taken into account. What is relevant is the total cost of ownership (TCO), which consists of the cost of acquisition, training, deployment, maintenance, repair and upgradation. UNCTAD's cautious assessment is that the "jury is out" on where FOSS stands on the total cost of ownership for, while the upfront costs are considerably less the training and deployment outlays can be as high as in proprietary software. Yet with FOSS suffering from fewer bugs and less downtime, the savings on maintenance costs will be considerable. For governments in developing countries with limited IT budgets, savings on costs - even just on upfront costs - can be a significant factor in aiding the spread of ICT. However, as the UNCTAD report points out there are a number of other advantages of special relevance in developing countries. The availability of the source code of FOSS facilitates involvement of a large domestic programmer community in repair and maintenance activity. This contributes to skill development at home. Overall, to quote the UNCTAD report, "The availability of source code makes it possible to use in-house expertise to fix bugs or change configurations, as well as to hire external support from a competitive market that anyone can enter. What seems clear is that FOSS can help a business or a public institution avoid getting into a vicious circle of hardware and software upgrades, and changes in data formats that require investing in new licence fees and significant retraining and can provoke major down time." And "the use of free software means that installation, training, support, and maintenance can be flexibly contracted out to a range of local suppliers competing on quality and price." Governments of a number of developing countries have already decided in greater or lesser degree to turn to FOSS. China, Brazil and South Africa are the most well-known examples. In addition to the advantages of cost and reliability, the increased security provided by the use of FOSS is a factor that has prompted a growing number of developing countries to turn to this software. No Government has as yet decided to make a 100 per cent shift, but the explicit policy directions given in national and regional governments to use open source software is a pointer to which way the wind is blowing. Governments in the advanced countries - including the U.S. - are also increasingly using FOSS because of its advantages in costs, reliability and security. In the developed countries, FOSS is close to replacing proprietary software in the "back offices", but on the desktop it is still the latter which has almost a complete dominance. The non-availability of a wide range of applications for the individual user and a perceived difficulty in initial installation of the FOSS operating system have held back the spread of open source software in homes. However, it is just a question of time before the spreading use in offices leads to adoption at home. A new problem that has emerged is that a few U.S.-based companies selling FOSS operating system software have been slapped with suits for alleged infringement of intellectual property rights. This has created some uncertainty about the future of the core operating system. With commercial interests having a strong stake in arresting the development of FOSS, it is inevitable that every possible attempt will be made to halt the spreading use of open source software. This, however, need not come in the way of developing country governments switching to FOSS. Next month, the U.N.-sponsored World Summit on the Information Society will discuss the many options possible to use ICT for development. The use of FOSS as an option to save costs, develop IT skills and facilitate the adoption of open standards will hopefully figure prominently on the WSIS agenda.