(In what follows I use the term Mukta Software for FOSS.
This is to distinguish this from Mufta Software which
is only free of cost.)
There are a number of different activities that could
be undertaken independently:
1. Development of Mukta Software.
2. Deployment of Mukta Software.
3. Increasing Usage of Mukta Software.
There need to be some significant differences from the way
these aspects are dealt with in a "closed" industry like the
auto industry or the proprietary software industry. In particular,
there is no sharp division between the users and system
administrators or between the system administrators and the
software developers.
Let us take the above activities one by one.
1. There are a number of major Mukta Software projects and each
   of these has a "To-Do" or "Wishlist" of features that need
   to be worked on. There is also a publicly accessible version
   controlled repository of developmental code. There is usually
   also a mailing list for developers.
   
   Prospective developers need to join the mailing lists, download the
   developmental code and start working on the "To-Do" list.
   New projects can also be started but the failure rate of
   such projects is high unless there is significant involvement
   of the larger community.
   In this context co-ordinating with groups like FSF-India, TeX User's
   Group India, the Indic Language Localisation Group and so on would
   prove beneficial.
2. The deployment of Mukta Software is made easier by the ability of
   these solutions to be shaped to the requirements of individual
   organisations. However, this shaping is not "factory-built" and
   must be carried out by the system administrators. An additional
   complication is that senior management has already been exposed
   to the proprietary model (usually through home/appliance computing+)
   and is often unwilling to adapt to the usage model for
   Mukta software.
   In spite of this a large number of system administrators across the
   country are already deploying Mukta Software. These efforts can be
   co-ordinated by the creation of software repositories (or "mirror"
   sites) and support groups (mailing lists, IRC or Wiki) which also
   function as information repositories.
3. The user of a Mukta Software system contributes significantly to its
   Development and Deployment. A user not only learns to operate the
   system to perform tasks but also learns to accurately report problems
   (file bug-reports) when they arise. In addition, feedback in the nature
   of compliments or thanks is also appreciated in the Mukta Software community
   where this is often the only form of "payment" that can be transmitted
   from user to developer.
--------------------------------------------------------------------------------
+ In the proprietary model it is natural to speak of appliance
  computing. When a toaster breaks down we just take it for
  repair saying "It doesn't work". This usage model is inappropriate
  as a Mukta Software user should be able to formulate complaints
  more precisely and perhaps even take care of simple fixes or
  improvements.
  The Indian user is used to doing this for his/her gadgets for
  a number of possible reasons.
  (a) The local repairman is a monopoly and is over-burdened or is too
      far away.
  (b) It is cheaper to repair things if one knows what is wrong.
  (c) Due to the self-sufficiency attitude inculcated in our freedom
      struggle.
--------------------------------------------------------------------------------
   We need to distinguish between two types of users when we try to
   train them as users of Mukta Software.
   A. The elite and small fraction of users (in India) who have already
      been introduced to computing via the home/appliance computing
      environment promoted by the proprietary model; the Windows(TM)
      operating environment is the dominant player here. While the
      number of such users is small it is significant as most of the
      people involved in decision making come from this background.
   B. The large fraction of users who have had (little or) no prior
      exposure to computing. This includes children and most Indians
      who live in villages.
   It is perhaps not surprising that (A) is the more difficult category
   to deal with as they have a well-formed idea of what they expect from
   computer systems and they would like to provide such systems for (B).
   At the same time as (A) are quite knowledgeable they can help
   significantly if they learn the bug-report aspect of Mukta Software
   usage. In addition, reviews of Mukta Software from members of this
   community will go a long way in influencing others.
   As far as (B) is concerned, the main issue is finding enough people
   to perform the training function; "training the trainers". Needless
   to say a majority of the trainers are in the (A) group. The actual
   training material (software and documentation) is readily available
   but needs to be translated and adapted to actual requirements.