Monday, September 16, 2013

turnkey enhancements, lessons from sept Kuru-Batti prototyping

Gordon, Chandana, and I drove across Sri Lanka (nearly 820km) mainly testing the FrontlineSMS software. At the Wayamba University Workshop, we had the opportunity to test the Ushahid mapping and FreedomFone interactive voice software.

From those prototyping and workshop lessons, the points below should be considered for future improvements, in relation to the turnkey solution:
  1. localized mobile phones - we noticed many farmers to carry phones with keypad and screens in Sinhala and Tamil. It was evident that SMS text from these mobile phones in Sinhala or Tamil would not trigger keyword enabled FrontlineSMS message forwards. Hence, FrontlineSMS would need to be improved to handle those utf-8 characters. Perhaps GUIs may need to be localized as well with features to transliterate with any non-latin scripting language.
  2. Freedom Fone - must be properly tested in; especially, with the dongles to make it cost-effective. The content upload feature must be simplified by introducing a well developed menu for instructional information.This is a high prioroty because both the Radio stations and the Department of Export Agriculture are keen in using it for localized (Sinhala & Tamil) large vocabulary continuous speech (descriptive) messaging to campaign with farmers.
  3. Bundle Ushahidi - would mean to include Ushahidi with FreedomFone and FrontlineSMS boxes. Ushahidi instance should function as a local-host with fixed maps. Ushahidi and Freedom Fone (exceptional to FrontlineSMS 2.0) should be accessible through the Mini-PC wifi with the Mini-PC serving as a localhost.
  4. Solar charger - farmers in Vakarai live on in the farm lands that are 10 - 20 KM away from their homes. They carry two batteries and charge the dead batteries when they go to town or visit home. We need to design a solar powered battery that can power the 20 Watt Mini-PC for daily use over 12 hours and an additional 5 Watts that can be used to charge phones when that capacity is not used to power the Monitor or other add-ons.
  5. Grab-n-Go Box - Although all the equipment can be easily fit in a tiny back-pack; it is ideal to fit them all in to a tool box type plastic container for easy-of-use and rapid operationalization.
While  improvement no. 1 with localization would be more of a subsequent phase longer term activity, the remaining improvements 2 - 5 can be achieved during this phase of the project.


  1. Thanks Nuwan. How much time do you think is needed to get FF ready for pilot testing?

    Please explain for the others what you mean by "fixed maps" with Ushahidi.

    Does Vakarai have a local community representative that could be assigned a turnkey for a pilot?

  2. Interesting summary Nuwan. The FrontlineSMS language issue is an important one. For the interface, 1.6 had a 'translation manager' feature [ ], but this disappeared in 2.2.

    With radio stations we assumed the mini-pc setups to be redundant, as all stations had at least one computer. However, I's like to know if you think a dedicated system is needed? Despite having computers, a simple set of recommendations on available, least-problematic dongles and some info on the best SMS rates from local telecomms would probably still be useful for radio stations.

    During our station visits it seemed that there was usually at least one tech savvy person who would have no problem handling something like FLSMS. The bigger question, for me, is whether the stations can effectively integrate these platforms into their programming and garner any real participation from the audience apart from song requests.