X
    Categories: Uncategorized

Using GRINS to Improve Technology and Processes at Community Radio Stations

Radio Bundelkhand, one of the early community radio stations in India, started live transmission in October 2008. We visited the station in February 2009 as a part of Community Radio India Forum annual body meeting. During this visit we initiated talks of piloting the radio automation system being developed by us. We released the Gramin Radio Inter Networking System (or GRINS) in June, and setup GRINS at Radio Bundelkhand during our week-long visit in mid-July. This report describes (a) the operational setup at Radio Bundelkhand before GRINS was deployed, (b) the changes in the setup made by deployment of GRINS, © value added by GRINS to the setup, (d) challenges faced by us during the installation, and (e) future plans with the pilot setup. In brief, we describe below some important contributions made by GRINS. Please read the report for further details.

 
GRINS setup at Radio Bundelkhand

  • Playlist Management: GRINS now allows the staff at Radio Bundelkhand to not only prepare and save a playlist, but to also schedule a playlist for automatic transmission on air at a specified time in the future. As a result, there is no need for a person to be present near the computer to start the transmission. Other features like display of start time for each item in the playlist and the ability to preview items in the playlist while another item is playing on air have provided convenience to the staff.
  • Content Management: One of the primary concerns at Radio Bundelkhand was the need for a mechanism to manage content created by the station. GRINS enables content management by allowing users to save metadata regarding programs present in the system and to search for them at a later date.

    • Users can store a variety of information like language of the program, details of creators of the program, description of the program, trivia about the program, and so on. Different categories can also be assigned to programs based on its content. For example, a program regarding an interview with a doctor on polio vaccination could be assigned categories of “interview” and “health”. Once metadata information has been fed for programs, searching for them becomes extremely easy with our “facet” based search engine.
    • GRINS library search engine allows one to search for programs based on several independent aspects or “facets”. One could search for programs based on length of the program, categories assigned to the programs, trivia or description of the program, and so on, each of them reducing the search results independent of the other. Effectively the results thus obtained are a logical AND of all the constraints specified by the user. The ability of a user to search programs using constraints of so many types allows him/her to specify very powerful search queries. This is particularly useful when the content database grows very large. Once programs of desired kind have been searched for, GRINS allows users to obtain group statistics of the searched programs. These group statistics include total number of hours the searched programs have been on air, cumulative frequency of transmission of the searched programs, etc.

  • Diagnostics: One of our main concerns with deployment of GRINS at Radio Bundelkhand was the unavailability of a local expert with knowledge of GRINS. This is expected to be the case at most radio stations even for the future. This meant that in case of any kind of failure of the system, the station staff would be helpless. To avoid this kind of a scenario, GRINS includes a “Diagnostics Widget” that allows a user to diagnose the reachability of all “services” and connectivity of audio cables. A diagnostic test like this could be run by the station staff before a transmission to ensure that everything is in order. We believe this feature would significantly offset the unavailability of local manpower with knowledge of GRINS. The true value of this feature will become known as and when more deployments are done and the system is used for longer periods.
  • Robustness: In order to ensure that local staff could rely on GRINS, it was imperative that GRINS be robust in failure detection, resilience to network failures, and ensure recoverability with the least disruption. Significant effort was put in to ensure online detection of failure of many components of the system, resilience to temporary network failures, and recoverability in case a service encountered an unexpected error.

    Anecdotally, as can be seen in the picture below, the “magic box” and the switch connecting the “magic box” with the computer running the user interface was placed near the feet of the staff during the first week when the system for being deployed and tested! Given the situation it was quite conceivable that a slight kick to the network cables could cause temporary loss of connectivity.

 
Network hub and GRINS under the table, installed dangerously close to feet

  • Transmission Logging: GRINS logs programs that are played out on air. The staff therefore need not maintain a large archive of all audio that is played out on air. A “Log View Widget” allows them to simply look at all the programs played out between any two dates. This ensures that Government regulations can be met with the least effort.
  • We will keep working very closely with the staff at Radio Bundelkhand to ensure that there are no technical problems faced by the station. Our next visit to the station will happen very soon when we will deliver several minor features requested by the station staff. We will also collect logs of operation of GRINS at the station for further analysis.

    Our next major upgrade will add telephony capabilities to GRINS. With these capabilities, the staff at Radio Bundelkhand will be able to accept phone calls and record conversations. Currently Radio Bundelkhand accepts song requests for only one hour during the day, where a staff member receives the call manually and records the conversation on a recorder by keeping the call on a speaker phone. With telephony capabilities, the station will be able to accept song requests throughout the day in an automated manner without the staff having to manually receive calls and talk to the caller.

    We also intend to collect logs from the radio station periodically over the next one year to study if the system experiences any performance or robustness issues. In the long run, this would provide us with information regarding stability of the system.

    Zahir Koradia :

    Comments are closed.