Open Source Simple Computer for Agriculture in Rural Areas
(OSCAR)

Field Trip

Field Visit in Nepal:

A field trip to assess the perception of OSCAR tool in Nepal was made by Dr. Rico Lie (CIS, WU) and D. Balasubramanian (IFP) from the 20th September to 25th September. The focus was to find out the possible end users and the possible ways of implementation of the OSCAR tool. The possible end users were thought of as farmers in the Indo-Gangetic plains, extension workers, students and scientists. Meetings were held with farmers and farmer groups, Krishi Vigyan Kendra (Farmers Science Centre) staff, Government extension workers, research station staff, NGO’s and universities and information was gathered through observations, formal presentations, interactions and dialogues. The outcome of the field trip can be succinctly summarised as follows:

1. There is need for information on weed management and control, not identification alone;
2. The OSCAR tool cannot be supplied isolated in a simputer but needs to be bundled with other relevant information for rural areas to make it a viable tool at the rural community level;
3. The effectiveness of OSCAR as an educational tool for students and teachers;
4. The design of the software was found to be plausible for all audiences but the numbers of character states were insufficient. This is understandable given the fact that it is only a beta version of the forthcoming comprehensive tool.

Field Visit in India:

Following the second workshop at Karnal, field trips were organised in the four states of Punjab, Haryana, Uttar Pradesh and West Bengal in India which belong to the Indo-Gangetic Plains. The team was split into two, the first team to work in Punjab and Haryana and the second team to work in Eastern UP and West Bengal.

The workshop was immediately followed by testing of the application with farmers and extension workers at Directorate of Wheat Research, Extension offices of State Governments and Universities in and around Karnal and Kurukshetra. This was organised from 30th September to 3rd October, 2004.

Then the two teams parted in different ways. One team proceeded to collect photographs of the weed species in Punjab and Haryana, was coordinated by the local RWC staff.

The second team proceeded to work in areas around Varanasi, Eastern UP and New Jalpaiguri in West Bengal. It was supported by the local RWC partners at Benaras Hindu University, who also are running active extension programmes in the areas around Varanasi. The software was also tested with farmers in the region who were involved in the e-chaupal programme, a private sector initiative in IT&C and agriculture in the region and was very well appreciated for its easiness of use. They also inferred that the availability of such an application on portals like e-chaupal would be greatly beneficial for them, since the access to extension services is very limited in the region. In New Jalpaiguri district of West Bengal, the photographing of weed species was continued and the farmers here too concurred with the above stated views, although there is hardly any presence of IT&C initiatives in the area.

Field Visit in Bangladesh:

The field trip and workshop in Bangladesh was organised between 28th November to 10th December, 2005 with the help of the partner Rice and Wheat Consortium (RWC) for Indo-Gangetic Plains. RWC had arranged for field trips in the districts of Dinajpur, Rangpur, Gazipur and the workshop was to be held at Dinajpur.

The field trip in Bangladesh was mainly organised to assess the appropriateness of OSCAR with the three target groups - farmers, extension workers and students. The OSCAR tool had been updated with new character states and weed description pages and we also had a beta release for the simputer version. It was decided to conduct systematic tests with the target groups with both the desktop and simputer versions.

Tests were conducted with farmers from the Dinajpur district who were associated with the activities of Wheat Research Centre, Dinajpur. This included demonstration of OSCAR with farmer groups in organised meetings and also direct tests with individual farmers on their fields.

Tests with extension workers comprised of concise introduction of OSCAR and letting them to use the software individually or in groups of two or three. They were asked to identify a few weed species on their own. Feedback was collected after the individual/group tests were conducted. The Department of Agricultural Extension was approached in the districts of Dinajpur, Rangpur and Bogra to organise the tests. Most of the tests were conducted parallely to the training sessions organised by the DAE to the Assistant Agricultural officers and the sub-assistant agricultural officers.

Tests with students were conducted at the Universities in Dinajpur and Gazipur districts. They were the Haji Mohammed Danesh Science and Technology University, Dinajpur and Bangobandhu Sheikh Mujibhur Rahman Agricultural University, Gazipur. The tests involved faculties and Masters and Ph. D students of Botany, Ecology and Agronomy. A short presentation of the project and demonstration of the OSCAR software was made followed by hands-on testing of the software by the students and faculties either in individual or groups of two.

The outcomes of the testing with all the target groups can be summarised as:

Visual interface:

Provision of a visual interface to the identification process is a major advantage in the tool. This eliminates any prerequisite botanical knowledge about weed species. All the end users felt comfortable with the interface design and found it easy to identify weed species. Though there were issues with their first exercises, subsequently every one was able to navigate through the software with ease.

Nature of information:

Information that is provided as part of the species description pages was found very useful to the students. For the extension workers and the farmers information on weed control measures needs to be updated. The scope for an ICT based tool like OSCAR is well felt in the field areas provided that it addresses the real needs of the target groups, here, the information on weed management and control. This needs to be immediately addressed in our future tasks.

Character and Character States:

It was felt in many places that not all character states were present for the Leaf character. Also the possibility of adding the characters of fruits and flowers needs to be explored. It might prove useful to add these characters as it may result in discriminate identification of species.

Local Language issue:

It was found necessary that providing weed control measures in local language, here Bengali is important for farmers. But for the other target groups such as extension officers and students it was not felt as important as the medium of instruction is English in all universities.

Other related issues:

It was a common question among the farmers and extension workers on how to continue using such useful software unless they have constant access to a computer. Though this is a typical question in the rural areas of the third world which is not covered by the scope of the project, we should also recognise the importance of the question. Deploying simputers with all its limitations of high price, long term support, availability of a number of relevant customised applications can only be judged over a long period. A follow-up of OSCAR is a possibility where we can try to work with a smaller project area with a specific target group and evolve a comprehensive ICT approach to rural areas.

The fact that OSCAR is free and open source needs to be emphasised and it seems to catch the imagination of the end-users, right from the farmers to the faculty of universities. They felt this could be a major advantage of OSCAR so that they can customise and enrich the software with the weed species of their region.