Seilevel
Seilevel Home
Back to Blog Home - Requirements Defined

Monday, December 03, 2007

Using PDAs for Requirements Gathering

I heard a talk at RE07 called Exploring Scenario Forms and Ways of Use to Discover Requirements on Airports that Minimize Environmental Impact by Maiden, N., Ncube, C., Kamali, S., Seyff, N. and Grünbacher, P.


First I want to say that hands down, Neil Maiden gets my vote on best speaker I heard at RE07 – if you get a chanced to hear him, he’s very fun! Anyway, more to the point, at RE06, this same group of people presented a paper on using PDAs for requirements gathering. I posted about that here, including my noted skepticism about the usefulness of PDAs for requirements efforts.
Well, they reported back this year on a project they were doing that used the PDAs. First, I have to say the project was really interesting! They were working with an airport to try to minimize the impact on the environment by reducing the noise and gas emissions. I believe they studied everything from the airplanes on the ramp, to the luggage carousels, to the car traffic around the airport.


They used their PDA tool on this project (though the actual focus of the research was on some visualization techniques). As an example, they would go to an aircraft as it parked at the gate, to talk to the people in context. They recorded the conversations to the PDA and attached them to the appropriate pre-defined event in the PDA. He gave a great example of how handy it was. He was walking through the control tower and saw a small door and asked “What’s in there?” Turns out that inside that room was an actor no one had identified in the project yet. So he could just zip in, have a conversation he recorded on the PDA and zip back out.


In their research, they found that they gathered 8 times more requirements while doing the sessions in context than they did gathering them in workshops. And of course the implied link is the PDA tool lets you do it in context, though arguably there are other tools to facilitate this if not a PDA.


All of that said, I must admit I am less skeptical than I was a year ago that there are such scenarios that such a tool could be useful; though I still think most projects won’t need this. I’d be interested to hear if anyone had success using tablet PCs for this as well.

Requirements Defined Newsletter Bookmark and Share

3 Comments:

Anonymous Anonymous said...

Using PDA isn't important. Imagine that the technology changed, so what. The real find here is that their process worked better than workshops.

In workshops, you will give more notice to those with the louder voice. The people with the louder voice don't do the real work, so the requirements will be skewed.

The speaker didn't address what they will do with all this research. More than likely, it will be the executive sponsor who speaks with the loudest voice, and thus much of the reserach will be thrown away to ensure that the executive sponoser gets what they want, rather than what they need.

Workshops speak for the executive sponsor and then leave the executive sponsor to drive the requirements volatility process, which is an inherent process in requirements elicitiation. Nobody really builds software to do the real work. Somebody always resorts to the post-delivery, construction of desktop applications to ensure that the real needs are met to the degree possible. The desktop apps are off the books, so who cares.

Actually, the developers care, but only because of the pain involved with all the volatitily, not the gross cost of it post delivery.

12/03/2007 1:16 PM  
Blogger Unknown said...

So, all they did was use their PDAs as digital voice recorders and perhaps categorize them? I dont know that this really speaks to the utility of PDAs, as it does to doing audio recordings of users.

I have been doing the Six Sigma Voice of the Customers thing where we go interview customers on their turf and record the interview so that we can capture their "voices" in their own words. When the user declines to be recorded, I attempt to type up their responses verbatim.

It sounds very similar, though less formal. The basics are the same - go to the user on their turf, in their workplace, record what they are saying, and do good requirements work (e.g. open-ended questions, probing questions, etc...)

12/03/2007 1:27 PM  
Blogger Joy said...

I started a thread on the message board where we can discuss these comments:

http://www.seilevel.com/messageboard/showthread.php?p=3688&posted=1#post3688

12/03/2007 6:02 PM  

Post a Comment

<< Home