User stories for Ruby build

User motivations:

A pulselab analysis will want to use Hunchworks because:

They have an incomplete understanding of a problem. They might have a hypothesis that explains the problem but they don't have enough evidence for it, and are unable to gather enough evidence themselves, either because they need extra effort to do this, or they don't have access to the evidence they need. They might have micro or macro data but not the other and need both; their understanding of the problem may be incomplete; they might not have direct access to data for neighbouring countries, or they be having problems converting data into compelling evidence.

Example hunch types are:

• I think something's happening, but I don't understand what it is.

Hunchlabs have an interesting cultural element - using a hunchlab changes the emphasis from analysts not admitting that they don't know, to openly sharing their uncertainties and collaborating on making best sense of them.

User stories

As an analyst, I want...

• I want to test a hypothesis that I have about something that I think may be happening in my area of interest, so I can alert policymakers earlier about upcoming crises. I look to see if there's a similar hypothesis in the system, and if there isn't, I create a page for the hypothesis, add a note about it and look for relevant indicators in the datasets available to me. I might also share this hypothesis with other analysts that might have an interest in it, or be able to help me gather evidence and prove or disprove it.

• I want to see if anyone else has a hypothesis that might interest me, so I can get early alert about a crisis in my area. I look at the hypotheses from users in my groups and users with similar interests to me. If I find an interesting hypothesis, I open it up to see what evidence and notes have been gathered so far. I might put a 'watch' on it, add my name to a group formed around it, or add my own notes and evidence to it.

As a lab manager, I want...

• I want to see all the hypotheses that are relevant to my lab.

As a network moderator, I want...

• I want to maintain a list of relevant and timely hypotheses.

• I want to notice that two or more hypotheses are about the same problem, and be able to mark that they're connected.