Re: Identifying the Problems Addressed by the User Stories

On Fri, Jul 10, 2009 at 2:57 AM, Oshani Seneviratne<> wrote:
> Hi all,
> I think it would be beneficial for the final deliverables of the TFs
> if we identify the underlying *higher level* problem(s) that each of
> the stories convey and need to be addressed. So, for example, based on
> the stories related to privacy on the wiki [1], I believe the
> following would be some suitable "privacy problem categories" :
> A. Access Control:
> Stories related to improper access, users not thinking much about the
> repercussions of not putting proper access control to their online
> social data, etc.
> B. Data Usage:
> Stories related to personally identifiable information leaking out
> from the user's social network. Once a user publishes some data about
> herself, or her family or friends, can somebody else use it? under
> what conditions?
> C. Trust:
> Can we really trust people to be who they say they are on social
> networks? (very much related to the identity-based user stories in the
> wiki). Things like age verification which needs to rely on a third
> party should also be considered.

I like these categorisations.  However I think Trust should read
"Trust and Reputation".  Trust, as I understand it, is the probability
[0,1] that a given action in the future will be performed as
described.  Reputation is a summation of pervious data points to give
a score that can be used to to calculate trust.  Reputation is easier
to capture using machine readable data.  The algorithm to calculate
trust can be quite complex and will often use subjective factors,
based on users and groups.

> If it makes sense to categorize user stories like this, I propose that
> we include a new section titled "Problem" or "Problem Category" (or
> any other suitable name) in the user story template at [2]. The
> alternative is to restructure the wiki based on these categories. But
> that could be problematic if a particular user story has more than one
> category.
> I wonder whether a similar approach would work for the use cases in
> Portability and Context TFs?
> [1]
> [2]
> --
> Oshani

Received on Friday, 10 July 2009 10:24:20 UTC