Information and communication technologies (ICTs) are powerful tools when it comes to spreading and sharing information and knowledge contributing to poverty alleviation, empowerment and income generation (Bridges 2001; Kenny 2002). Under the right conditions ICTs have proved to be able of inducing social and economic development in terms of improved education and health care, agriculture and trade, employment and of promoting local culture (Harris 2004). Concerning, ICTs like FOSS, several authors see FOSS as offering particular advantages to DCs because it has the potential to curtail some obstacles towards full use of ICTs and to contribute to bridge the digital divide (Rajani 2003; Uimonem 2004; Van Reijswound and Topi 2004; Wong and Sayo 2004; Gauguier and Douine 2005; Krakowski 2006). The openness of the FOSS model promotes the concepts of access (more transparency of process and product), community (leveraging local and global resources), collaboration (sharing ideas with peers) and choice (more competition) (Dravis 2003). Nonetheless, it is clear that ICTs cannot solve every problem, and there are significant downsides and limits to ICTs from a DC standpoint (Kenny 2002; Harris 2004). Introducing FOSS in Dcs is not about adopting software free of charge, but rather about acquiring software that can be modified and developed to meet local needs and participate in the innovation process (Weber 2003; Uimonen 2004; Cook and Horobin 2006).
There are several prerequisites and challenges that influence the use of ICTs as tools to alleviate poverty, and they are particularly strong in DCs. This section explores how the adoption of FOSS in DCs can act as enabler for ICTs for development goals and overcome some of the challenges presented in the previous section. To structure the review and discussion, the key elements – technological, socio-cultural, legal and economic – from environmental or contextual frameworks proposed or used in previous literature (Mbarika and Okoli 2003; Bajaj and Leonard 2004) to evaluate ICT challenges in DCs are adopted.
ADDRESSING ICTS CHALLENGES WITH FOSS
One of the key factors that limit access to ICTs in DCs is the lack of adequate and affordable hardware and software (Guaguier and Douine 2005). In fact DCs rely on major external IT vendors based in developed countries for supply, support, future upgrades or re-development of technology (Dravis 2003). This technological dependence for the provision of software and hardware prevents DCs from shaping technology according to their needs and interests, consumes precious financial resources and may have negative consequences on national security and data protection, as systems can be controlled from a distance by external suppliers (Rajani 2003; Weber 2003; UNCTAD 2003; Weerawarana and Weeratunga 2004). Moreover, systems produced in developed countries embody relations, practices and views that do not generally fit the context of DCs (Westrup 2000; Heeks 2002).
To address these technological challenges, FOSS offers to DCs a set of high quality and adaptable systems and applications: in fact the most mature and popular FOSS applications have proved their higher security, performance, stability and reliability comparing to most proprietary software (Wheeler 2003; Grimshaw 2004; Wong and Sayo 2004). FOSS’ development model makes patching bugs easier and more likely, and motivates software developers to spend more effort on the quality of their code (UNCTAD 2003; Hopeman and Jacobs 2007). FOSS’ availability of source code increases software security as it allows independent assessment of the exposure to the risk and vulnerabilities to be identified and resolved by third parties (Wheeler 2003). Systems with open-source code allow the State and citizens from a DC to inspect the code themselves and check for back doors and spyware, and thus to be independent from providers and external control (UNCTAD 2003; Weber 2003; Weerawarana and Weeratunga 2004). The use of open standards and formats not only ensures application interoperability and minimizes proprietary technology lock-in, but also increases free access to government information by citizens (Dravis 2003; Wong and Sayo 2004). The freedoms associated with FOSS make it flexible and adaptable to local contexts and information needs allowing DCs to bridge the gap between a foreign technology and its full application to local conditions (Camara and Fonseca 2007).
Concerning socio-cultural challenges, the lack of local language, locally relevant and adapted contents, applications and interfaces is a major barrier to increased access and use of ICTs in DCs (Hall 2000; Guaguier and Douine 2005; Harris 2005). IS can only work effectively when they are in the mother tongue of their users and take into account local cultural factors (Hall 2000). Localisation of software can boost e-literacy and reach those categories of people like illiterate, rural poor or women in DCs, who are generally outside the élite business and government circles in which software often circulates and are less likely to know to the official and/or national languages (Keniston 1997; Kelkar and Nathan 2003).
However the primary drivers for the localisation effort of the large IT publishers and software vendors are market expansion and higher revenue, and certainly not DCs’ cultural diversity preservation or access issues (Schäler 2003). Software vendors generally do not always consider adaptation and translation of their products commercially viable in DCs: the markets are too small and few of the people who speak these local languages can afford expensive software (Uimonen 2004; Noronha 2005). So they tend to translate software into the languages spoken by the affluent people of the northern hemisphere and limit localisation for DCs to “the lowest common denominator and global symbolism” (Schäler 2003: 7). In fact the localisation strategies of multinational vendors aim to bring little change and use and spread globally accepted standards, symbols, icons, conventions and references in order to reduce costs and create global and homogeneous markets to their products (Schäler 2003).
FOSS instead offers DCs a bottom-up localisation approach that respects their cultural and linguistic diversity and let them shape technology according to their development needs. FOSS enables software adaptation and localisation by single developers or communities that wish to do so without recourse to negotiations with the owner (Uimonen 2004; Weerawarana and Weeratunga 2004; Wong and Sayo 2004; May 2006). As the code is immediately available for adaptation, language and culture-based modifications can be introduced in the software locally to produce local solutions that suit the specific demands of particular groups of users or of cultural regions in DCs, regardless of their economic weight (Uimonen 2004; Weerawarana and Weeratunga 2004; Wong and Sayo 2004; May 2006). Support can also be localised and is not dependent on authorisation from foreign providers (May 2006).
Legal and Economic Issues
Another barrier to the adoption and use of ICTs in DCs is represented by the cost of software. Software is especially expensive for them not only because it is an imported good but also because it is subject to Intellectual Property Rights (IPRs). Software piracy, defined as illegal distribution and/or reproduction of software, has been a way to drastically reduce the cost of acquiring software. It represents a problem in both developed and developing countries. However, piracy rates in DCs, where low incomes make software far more expensive, are steadily high and can reach up to 90 percent of the total software used (IDC 2007). A culture of software piracy hurts local software industry development, as there is less incentives for local developers to innovate and start-up a business (Wong and Sayo 2004; IDC 2007). While DCs with poor protection for IPRs are not as attractive to foreign investors (Wong and Sayo 2004).
ICTs have enormous potential to increase productivity of economic sectors, to overcome problems of inefficient administration, and to expand and improve the quality of services (Avgerou 1998). Moreover the creation of a national IT sector can contribute to the growth of a country. However lack of financial and skilled human resources, high cost of IT initiatives and dependence on foreign providers prevent DCs from boosting the development of local IT and IT-based sectors and from gaining most of the benefits that ICTs promise to offer (Rajani 2003; Bridges 2004; May 2006).
In addressing these legal and economic challenges, a reduction in software piracy results in clear economic gains, including greater fiscal contributions and employment opportunities. FOSS undoubtedly provides an alternative to dealing with proprietary IP regimes in contrast with software piracy, and is thus a means of reducing IT investment costs within a framework of compliance with IPRs and standards (Rajani 2003; UNCTAD 2003; Weber 2003; Weerawarana and Weeratunga 2004; Wong and Sayo 2004; Krakowski 2006; May 2006).
FOSS is generally considered a cost-effective solution because the price for licence and up-grades is generally much lower than the price of proprietary applications. Most common software applications, operative system and their up-dates are available on CD-ROMs or to download from the Internet without any cost (Tong 2004). Moreover most common FOSS solutions generally can run on cheaper or more basic hardware platforms than proprietary alternatives (Weerawarana and Weeratunga 2004). However the cost of a given software option is not limited to licensing fees alone: the total cost of ownership (TCO) of a technology assesses both human and technological costs of the software across its life cycle to calculate the benefits gained. TCO includes direct costs, such licence fees, installation, training and support expenses, and indirect costs associated to the use of an application, like maintenance and hardware and software upgrades (Pogue and Day 2004). In developed countries, where labour costs are high, the cost of labour-intensive components of the TCO, such as software support, customisation, maintenance, integration and training, is high relative to the licence fee for software. Therefore when TCO is calculated, the cost of the licence fee is not a crucial component and shifting from proprietary software to FOSS would make small cost saving (Ghosh 2003). Instead in DCs, where labour costs are low, the cost of the software licence becomes a relatively more important cost component. The reduction or absence of licence costs make a deep difference to the TCO in DCs, where labour costs, income and purchasing power are considerably lower (Ghosh 2003). Nevertheless, as with proprietary software, the implementation, maintenance and support costs of a FOSS solution can only be effectively assessed across economic, social and environmental dimensions over the short, medium and long terms (Pogue and Day 2004; Uimonen 2004).
FOSS philosophy and development model encourage the development of computer programming and software developmental skills within the local user community (Dravis 2003; Wong and Sayo 2004; May 2006). FOSS allows developers to easy obtain, use, learn from, modify and build on existing knowledge and pre-built components, creating a new form of technological transfer. The skills that participation in FOSS communities offers are not only related to programming but to all the software development process, such as teamwork and team management, license and copyright law (Ghosh 2003). The open and collaborative nature of FOSS offers to student and junior developers a high level informal training and apprenticeship system based on learning by doing (Ghosh 2003; Wong and Sayo 2004; Gauguier and Douine 2005). The development of such skills could stimulate a nascent local software industry (Dravis 2003; Gauguier and Douine 2005; May 2006). Because generally most of the software in DCs is imported, the purchase of software licenses and up-dates absorbs important financial resources that could be rather spent within the economy of the country or on developmental projects (Dravis 2003; UNCTAD 2003; Wong and Sayo 2004). FOSS offers the opportunity to limit the transfer of capitals to procure solutions which can be produced locally by national companies thus reducing money and knowledge drains in DCs (Rajani 2003; UNCTAD 2003; Uimonen 2004).
LIMITS OF FOSS IN DCS
Although the adoption of FOSS in DCs seems so promising, FOSS is not suitable for every situation. The majority of FOSS development is still performed in the developed world, and, although FOSS is an international phenomenon, it is not a global issue (Feller and Fitzgerald 2002). Several challenges have limited extensive diffusion and adoption of FOSS in DCs and prevented them from enjoying the opportunities that FOSS offers.
Poor infrastructure is a major barrier to the development of FOSS and ICTs (Rajani 2003). FOSS development needs a good Internet infrastructure as reliable broadband access is fundamental to download, develop and disseminate FOSS applications and documentations and to create FOSS communities (Uimonen 2004). Prohibitive cost of computers and connections is one the greatest impediments to the work of the local programmers in DCs (Bridges 2004). The dearth of appropriate local assistance in implementing FOSS is even more dramatic in DCs: here the small dimensions of the FOSS communities who also serve as a support centre, and communication problems with FOSS mailing lists based in developed countries increase FOSS adoption problems (Van Reijswoud and Mulo 2006). Other factors include the lack of manuals and up-to-date documentation in local languages and of trained IT professionals aggravated by the brain drain (Rajani 2003; Bridges 2004). Learning and supporting FOSS may require a greater training input than equivalent proprietary software: in fact the impact upon the workforce of several years’ exposure to and use of commercial software (predominantly Microsoft) has been so strong that most IT workers in DCs have a solid rounding in the use of that software and are not pre-disposed to learning how to manage, support and use anything else (Rajani 2003; Morton 2005).
Another challenge to the migration from proprietary systems and applications to FOSS systems is users’ resistance to changing and learning how to deal with a new technology (Uimonen 2004; Van Reijswoud and Mulo 2006). This resistance to the acceptance of FOSS is also motivated by the lack of awareness of the FOSS option and the stigma against local talent and locally produced solutions (Bridges 2004). Products and applications that have been developed in USA or Europe are perceived as better and more reliable to the detriment of the domestic ones (ibid.). The tendency to mistrust or devalue anything that does not appear to have a cost and the perception that software with a retail price equivalent to a year’s income (or greater) “must be superior” to FOSS fuel the prejudice against FOSS (Bridges 2004; Morton 2005). FOSS developers in DCs tend to be isolated because of the lack of infrastructure and of IT physical spaces (Bridges 2004).
DCs’ hierarchical and lethargic bureaucratic structures, characterised by power distance, strong vertical authority dynamic and based on familial relationship, inhibit the introduction and use of FOSS for new information systems in public sector organisations (Rajani 2003; Morton 2005). Many decision-makers’ are risk-averse and afraid to believe that FOSS solutions, built on a voluntary base, are really working. In countries where there is little money and few job opportunities, it is hard to accept that people have the time and motivation to work on something for free in their own spare time (Rajani 2003). Structural corruption and lack of political freedoms adversely affect the spread of FOSS in many countries: foreign companies can easily convince decision-makers to choose their solutions with bribes; official censorship, despotic control and lack of civil rights prevent the creation of hacker culture and online communities (Rajani 2003).
Legal and Economic Limits
Other important factors to consider are the lack of adequate IP and IT policies. Even if countries have IP laws, the legal systems are not always enforcing intellectual property rights and the relative impunity with which people are able to pirate software or violate FOSS licences terms or copyleft for commercial interests, block FOSS investments (Bridges 2004).
DCs suffer from poor financial resources and other problematic issues, such as poverty, illiteracy, drought, conflict, etc. may be given a priority respect to ICTs and FOSS. FOSS is not perceived as the most cost-effective solution as Microsoft and other commercial software vendors have ensured their hegemony by granting very low cost (or even free) licences to DCs government (Rajani 2003; Morton 2005). Because in many DCs IT is not yet a part of the economic process, markets for software development are still small. Programmers are often underutilised by people who hire them, and find a business environment that is hostile to the incubation and development of their technical skills (Bridges 2004).
These factors coupled with the lack of appropriate FOSS policies to support national development goals have negatively influenced the up-take of FOSS in DCs. FOSS is on the agendas of donor organisations and international NGOs but not on the agenda of decision makers of least developed countries (Van Reijswoud and Mulo 2006). Although there is a growing number of initiatives to promote FOSS in DCs, few organisations are implementing FOSS and there is a lack of empirical ground-level research and best practices on the use of FOSS in DCs.
FOSS ADAPTATION IN DCS
Despite the shortcomings and the obstacles to its adoption, FOSS is able to meet specific requirements not met by either packaged or customized proprietary software (Bessen 2006), offering DCs not only the opportunity to foster the production of local software, but to adapt and shape software according to their own needs and context. The relative ease with which FOSS applications can be modified for specific language or cultural groups combined with local IT skills and expertise in the nuances of cross-cultural translation, can bring to successful localisation and therefore, help DCs comprehend and utilise the benefits of ICTs (Souphavanh and Karoonboonyan 2005). The localisation process consist of adapting, translating and customizing a software application or package in order to make it culturally and linguistically appropriate to a specific audience (ibid.). Local adaptation embraces not only linguistic issues but also the underpinning technology, economic and socio-cultural issues (Hall 2002).
Although adaptation and localisation of FOSS seems an issue of paramount importance in current literature, almost no research has been done on how to introduce and indigenise FOSS in DCs (Van Reijswoud and Mulo 2006). Also literature on software localisation and internationalisation does not consider the particular needs of the developing world. Their approach is generally technically oriented and ignores the social embeddedness of information systems (IS) in organisation and, most importantly, in a developing context. FOSS, as every IS, is a socio-technical system because it combines both the social – people and organisations – and the technical (Avison and Fitzgerald 2006). Understanding which factors should be considered in selecting, adapting and implementing FOSS systems and applications in DCs is fundamental not only to the diffusion of FOSS but also to realise the benefits that it promises. This research would go beyond technical requirements for adaptation, and draw knowledge from general IS literature to consider the socio-cultural requirements for adaptation. The process of localisation is much broader than simply translation, and awareness of the specific and unique linguistic and cultural context of a specific audience is necessary to the success of this process. FOSS localisation must address three main issues:
- Linguistic and Technical Issues: Technology localisation involves the acquisition and adaptation of new software or new features of existing software or the adaptation of existing features in existing software to suit local needs. It comprises the translation and adaptation of user interfaces, string resources and documentation. Different language characters and bi-directional texts imply specific requirements, for example Arabic and Hebrew languages are written and read from right to left, Far Eastern languages require more disk space as is required for English (Pustakalaya 2006).
- Socio-cultural Issues: Much of localisation, as practised today, is restricted to displaying all messages and interactions in the local language. Changing the icons, interfaces and cultural-dependent components to suit cultural variations is the next challenge. Issues such as adapting or designing specific software per the locally accepted norms, cultural conventions and regulations, called “locale” should be considered as well (Hall 2002; Jaffry and Kayani 2005).
- Policy and Legal Issues: Compliance to international IP laws and national IT policies must be respected and taken into account when adapting FOSS. The product resulting from the localisation process must respect the FOSS licence of the original application.
This second piece of our four parts series on FOSS has revealed FOSS to be a philosophical stance as well as a type of software and a means by which to share intellectual property. It has also shown that the adoption of FOSS offers several advantages to DCs, providing similar or improved functions to those offered by proprietary software.
However FOSS philosophy and development model present some limits that must be considered while making a choice between different options. At present the development potential of FOSS for DCs is still a theoretical potential (Reijswoud and Mulo 2006). If FOSS shows to be the most cost-effective solution for a country or an organisation, the localisation process of a FOSS application should provide the benefits offered by FOSS, reducing the downsides of this technology in the developmental context in which it operates. Moreover this process should address not just linguistic and technical issues but also socio-cultural adhering to FOSS licence and international and national legal obligations such as IP laws and IT policy.
The third piece applies these suggestions by proposing a FOSS localisation/adaptation framework that defines the different steps to follow with respect to an established model for IS development.
All references will be provided in the concluding piece of this four part series.
By Richard Boateng and Silvia Aimasso, PearlRichards Foundation, Ghana (www.pearlrichards.org)