Information Management for
East and Central Africa Agricultural Networks
An ASARECA Workshop
Entebbe, October 9, 1998

by Jeffrey A. Cochrane, AfricaLink Advisor
Michael Hailu, AfricaLink East and Central Africa Coordinator
October 20, 1998
Updated January 8, 1999

Summary and Findings

• AfricaLink results and five-year plan were presented to a meeting of the directors and network coordinators of ASARECA in Entebbe. AfricaLink efforts to date have focused largely on connectivity to the Internet for ASARECA network members. The second stage of activities takes advantage of this foundation to build complete information management systems for the effective sharing of information.
  
  
• AfricaLink's second stage is closely aligned with the mission of the newly formulated (but not yet funded) Regional Agricultural Information Network (RAIN). This new ASARECA network brings together national agricultural information centers and research services. RAIN will maintain the stock of knowledge, locally generated as well as acquired internationally, that will be accessible to scientists in the research networks and to the world, via the information infrastructure established with the assistance of AfricaLink.
  
  
• Estimated costs of fulfilling needs identified during the workshop include $45,000 for professional electronic conferencing facilitation, $90,000 for critical site basic connectivity, $81,000 for enhancements to connectivity to facilitate conferencing, and $40,000 for East and Central Africa coordination, a total of $256,000.

AfricaLink and ASARECA

The Association for Strengthening Agricultural Research in East and Central Africa (ASARECA) is organized around agricultural research and support networks. These networks are focused on particular crops (e.g. beans, maize), or on ecosystems (e.g. African highlands), or on special support functions (e.g. information management). Depending on the nature of a network's activities, the professional membership might number as many as 200 scientists in the 10 countries of the region.

The meeting in Entebbe brought together the directors of national programs of agricultural research in the region, who together serve as the board of directors of ASARECA. In each country there is typically a national program of government-sponsored agricultural research. The national directors of these programs sit on the board of ASARECA to coordinate the regional participation of their national scientific offices and to approve joint research plans and budgets. It is to this board that AfricaLink results and its five-year plan were presented.

AfricaLink Results Summary

• Developed an action plan for AfricaLink implementation, discussed and approved it with ASARECA members.
• Provided basic connectivity for network members in most of the ASARECA member countries.
• Initiated a technical support mechanism through a regional help desk.
• Initiated a user training programme at national level.
• Developed a database of network members and their connectivity status.

AfricaLink and Network Coordinators

Also in attendance at the Entebbe meeting were the professional coordinators of each of the ASARECA networks. There is generally a paid coordinator for each network affiliated with ASARECA. Typically an international center for agricultural research (e.g. the International Centre for Research on Agroforestry - ICRAF) will receive funding from a major donor (e.g. the US Agency for International Development or the European Union) to manage a network in its subject-matter area of expertise (e.g. agro-forestry). This international center will usually assign a network coordinator to oversee this management.

Network coordinators are directly involved in the research and other activities of their networks' members, at times facilitating transfers of support funds, securing necessary consultative support, compiling data on research results, and submitting an annual report and resource request to the ASARECA board for approval. One of the purposes of the Entebbe ASARECA meeting was to hear and approve the annual reports and budgets of the coordinators in behalf of their networks.

In each network, it is the network coordinator who is uniquely positioned to oversee the management of regional information. It is thus to the network coordinators that the AfricaLink Advisor from Washington and the AfricaLink Regional Coordinator from ICRAF turned for guidance on information management support. Additional assistance was provided by a representative of the Southern Africa Non-Governmental Organizations Network (SANGONeT).

AfricaLink and RAIN

Information management constitutes the infrastructure upon which regional collaborative research is built. As an organization for the strengthening of agricultural research, clearly the facilitation of adequate information infrastructure is a key component of the ASARECA agenda. The AfricaLink program funded by USAID and coordinated by ICRAF in East and Central Africa exemplifies ASARECA's commitment to this facilitating objective. The newly approved (though not yet funded) Regional Agricultural Information Network (RAIN) is further evidence.

AfricaLink has focused to date largely on the connectivity requirements of the members of the ASARECA networks. Funding has been provided generally for hardware, software, and startup fees to establish Internet links at the offices where network members are employed. The offices themselves have agreed to assume the monthly running costs for telephone lines, subscription fees, and equipment maintenance.

Connectivity is merely a first step. The second step takes advantage of this physical foundation by establishing information management systems. Such systems comprise the skills required by network members to share information effectively via the Internet, consultative and facilitation services to coordinate and encourage smooth and timely transfers of information, and links for the routine acquisition and dissemination of important programmatic information.

This second step in the AfricaLink program is closely aligned with RAIN. This new network of ASARECA brings together the leadership of national agricultural information and documentation centers and research services. These national agricultural information focal points (NAIFPs) seek to unite under RAIN to integrate their various holdings, and to make them available regionally. They also seek to share costs and increase their collective holdings of expensive commercial databases.

RAIN has been formulated at an opportune moment. AfricaLink is presently considering a pilot proposal from the proposed focal point for Uganda, the Agricultural Research Information Service (ARIS), to develop stocks of research information and make them available to network members. The resources requested by ARIS are quite expensive, and beyond the capacity of AfricaLink to support. RAIN will make it possible for services such as ARIS to acquire these resources on a shared basis at a reduced cost.

As scientists implement joint regional research programs closely coordinated via electronic mail, aided by technical support and facilitation services implemented under AfricaLink, it is critical that RAIN establish procedures and the reference resources required to respond in a timely manner to the specific scientific requests of network members. As new knowledge becomes available as a product of network research, RAIN appears also to be the appropriate coordinator of any centralized repository in the region. RAIN holdings can then be disseminated to the extension services and farmers where they can be put to best use.

Requirements for Regional Information Management:
Networking Facilitation

The coordinators of nine of ASARECA's networks convened for a one-day workshop to discuss strategy and support for more effective information management. The coordinators were polled as to their perceptions of greatest needs.

Network Coordinator Survey

• 1380 scientists in the represented networks
• 250 with access to electronic mail
• 121 with access as a direct result of AfricaLink assistance
• 44 per cent (of 250 with electronic mail) sending/receiving scientific information weekly
• 88 per cent (of coordinators) rate regular (weekly) scientific exchange as very important
• 56 per cent (of coordinators) rate regular scientific exchange as the most important benefit of Internet access
• Asked to list two key areas requiring future support, 44 per cent of 18 responses were for further assistance with basic (minimal email) connectivity, while a further 22 per cent were for improvements to existing basic connectivity.

Note: Coordinators were asked to estimate the numbers above without access to their files. They request that a more formal survey be conducted if great reliance is to be placed on these numbers.

Coordinators were first asked about strategies for collaboration. In the past, networks have met once or twice a year to share research results and to agree on a program of research for the year to come. The implementation of electronic mail makes it possible to "meet" electronically much more often. In addition to sharing results, it is now possible for scientists to implement a joint research project with even weekly coordination. Just as a scientist in a university might meet weekly with the scientists across the hallway or in the next building, so too might the scientists in a regional research network communicate regularly for joint project implementation.

All but one of the coordinators agreed that closer intraregional collaboration in the implementation of research programs is a critical requirement. The kind of close collaboration for joint project implementation described above does not appear to be taking place in the region. They estimated that only some 40 per cent of the professional members of their networks with access to electronic mail were presently sharing scientific information on a routine (i.e. weekly or better) basis with their colleagues within East and Central Africa. (Note: The coordinators requested that a more formal survey of network members be undertaken if this statistic is to be used to determine AfricaLink strategy.)

A possible solution to this problem is networking facilitation. One facilitator would be designated within each network. A professional Internetworking cofacilitator would then be employed to assist. The professional would teach facilitation skills, including strategies for encouraging productive debate, electronic activity management, and online research.

Networking Facilitation Examples

AgNet is a Leland Initiative supported electronic conference of agribusiness professionals in Southern Africa. A professional facilitator in Washington posts information of interest to agribusiness professionals in Southern Africa who previously participated in a regional workshop.

The facilitator of AgNet works 10 to 20 hours per week searching the Web and monitoring related electronic discussion groups to glean items of interest for reposting to the Ag-Net forum. She also responds to questions raised by members of the forum, or refers questions to others for response. Based on standard rates for consultants in Washington, such facilitation might cost $15,000 for six months for one network for 10 hours per week of support.

Afrik-IPM is an electronic conference that receives partial support from AfricaLink. A research associate at an American university is working closely with a member of the Integrated Pest Management (IPM) network to facilitate discussion among forum members and to disseminate relevant news items gleaned from Web sites and related electronic conferences. The principal difference from Ag-Net is the co-facilitation furnished by an African and American scientist.

The Canadian International Development Research Centre (IDRC) has published a booklet with guidelines for facilitation services. Discussions with IDRC staff suggest that a single professional facilitator might productively assist up to five networks, focusing on one each day. There is at least one Internet service provider (ISP) in the region that offers such facilitation services. While subject to negotiation, the approximate cost for a single network for six months might be $5000, or $45,000 for nine ASARECA networks.

Professional cofacilitation services would be provided to each network for the introductory period, though networks could of course choose to continue facilitation services for longer periods if the members found it helpful. A single electronic conference could be established for each network for general discussions of network business. However, it would be a simple matter to establish additional special electronic "meetings" from time to time for specific projects.

For example, a meeting might be established for a working group of scientists devoted to a particular task, such as a jointly implemented field trial across several countries. Another meeting might be established for the drafting of a joint document, or for the preparation of a workshop. The professional cofacilitator could assist the network's own member facilitator with each of these, suggesting options for meeting formats, moderation, posting of summaries on a Web site, etc. Network coordinators for the PRAPACE and ECABRN networks indicated they already have projects for which such regional task-oriented electronic meetings would be quite appropriate. One ISP in the region charges only $60 per six month period for hosting an electronic meeting on their server.

The professional cofacilitator might also be asked to assist with the development of a network Web site or online database. As suggested above, a Web site might be a place to post summaries or even real-time transcripts of comments in electronic conferences or meetings. A Web site could also host online databases of information. Such information might be established to assist network scientists in their research, or they might be used to publicize the outputs of their research. One ISP in the region charges a nominal $50 to provide space for a Web site, plus an additional small amount each month to sustain it, depending on volume.

Many of the network proposals at the Entebbe meeting of ASARECA included line-items for Web sites. When asked about these, the network coordinators did not have any specific plans for implementation. It may be helpful to have a single Web site managed under RAIN and hosted on an ISP server in the region. ASARECA might then register the http://www.asareca.org domain name, and create subsites for each network.

Requirements for Regional Information Management:
Basic Connectivity

While nearly all network coordinators agreed that more intensive intraregional networking was a critical requirement, they were also quick to point out the physical barriers to such networking. Connectivity remains a serious problem in the region. When asked to name the two most important support requirements for intraregional networking, 8 of the 18 responses were for basic connectivity to be facilitated at sites where there is presently no Internet access at all. The requests were quite specific:

• Mulungu in the DRC
• Uyole in Tanzania
• Ilonga in Tanzania
• Mkwapa in Tanzania
• SWRC in Sudan
• FRC in Sudan
• Areka in Ethiopia
• Kifu in Uganda
• Cala in Madagascar

In many cases these are remote sites with quite substantial infrastructural problems. Expensive solutions may be required, such as the radio-email solution now being tested at Namulonge in Uganda. A capital investment of $8000 was required at Namulonge, in addition to monthly usage charges that are generally much higher than standard telephone line access to the Internet. Some $90,000 would be required to implement similar connectivity at all the sites listed above.

A company in Kampala, Bushnet, is technically capable of serving the entire region (500 to 1000 kilometers) via a high-frequency (HF) radio email system, though special operating licenses might have to be arranged for some countries. Bushnet has been working closely with the Director of the Namulonge agricultural research station to develop special software scripts to filter unwanted mail, to divert unusually large messages, and to attribute costs to specific accounts. These are all techniques to control the high costs associated with this technology. Bushnet is also now developing capacity to support microwave communications to deliver full Internet access at high speeds and lower costs over shorter distances (up to 50 kilometers, line of sight).

Requirements for Regional Information Management:
Enhanced Connectivity

Even at locations where electronic mail is now feasible, often as a result of AfricaLink assistance, there are still significant barriers to intraregional networking. In the survey of network coordinators, a further 4 of the 18 responses to the request for the two most important support requirements were for further assistance to upgrade the quality of connectivity at particular research stations. The coordinators explained that it is difficult for a scientist to participate in an electronic conference if access to electronic mail is not of high quality.

The highest quality access to electronic mail generally available in the region is through a desktop computer with an instantaneous link to the Internet. Messages posted in one country are available to such computers within seconds. An entire conversation can occur in the space of an hour.

There are many lower qualities of electronic mail access. The link to the Internet might not be instantaneous, for example, because mail exchanges might be scheduled to occur only two or three times per day to save on telecommunications charges. This is often the case where research stations cannot afford the cost of a dedicated connection to the Internet, which is in fact almost always the case with members of ASARECA networks. Intermittent access to the Internet delays response time for participants in electronic conferences.

Another lower quality option entails access to electronic mail through a computer that is not on a researcher's desktop. Instead, access might be through a computer in a common computer lab, or worse, through a computer operated by a secretary or receptionist. Electronic mail that passes through an intermediary like a secretary is particularly problematic, especially when the secretary provides this service to multiple users. Again, the motivation is to economize on the high costs of computers and telephone lines to individual scientist offices.

If some members of a network have high quality access, while others have low quality access, networking is significantly impaired. Those with higher quality access can easily dominate discussion because they see and can respond to messages more quickly. Expert facilitation can help the network to overcome this problem, but only to a limited degree, by counseling the better equipped network members about how to be more considerate of their less equipped colleagues.

Unfortunately, the highest quality Internet access solutions are also the most expensive. High quality access often requires the installation of additional telephone lines to individual workstation modems, or a combination of dedicated leased telephone lines and local area networks. Where the typical AfricaLink assistance packages cost $800 and impose a subscription charge of perhaps $50 per month on the host research center, an enhanced solution could easily cost $20,000 initially, and the monthly fees could easily be far beyond the financial capacity of the research center. Only in exceptional cases for the highest priority stations were telephone lines or extra computers purchased under the AfricaLink program.

AfricaLink thus assured basic electronic mail communications for a broad number of researchers to satisfy routine, low volume exchanges of messages. While network coordinators expressed great appreciation for this level of access for their members, they were equally quick to note that high quality intraregional networking with potentially high volumes of mail exchange will require higher quality access to electronic mail for all their networks' members.

A typical "upgrade" would require establishment of a computer room with a dedicated communications computer and an extra telephone line. Assuming the room was provided by the research center, a dedicated computer would cost $1500. The telephone line is highly problematic, and the cost is difficult to estimate, since some installations may actually require laying cable long distances. In other cases an upgrade at the telephone exchange itself might be required. For those sites that can be upgraded, a quite rough figure of $3000 may be used for the cost of line installation. Upgrades could be required at 50 per cent of all network sites, perhaps two of four research stations on average in each of the 9 countries of the region, a total of 18 sites. $81,000 is therefore the estimated total requirement.

[Note: Since this report was originally published, ASARECA has been awarded a grant of approximately $270,000 for support of the kinds of activities noted above. JC, Jan 8, 1999.]