RealSoftwareDevelopment - a new methodology?
Submitted by Antony Marcano on Wed, 04/04/2007 - 01:43.
development methodology
At XTC this evening I was on my soap box about how I am fed up with the strict adherence to generic methodology and the abuse of the term "Agile".
There must be some way of reflecting reality... So, I wrote in my notebook - RealSoftwareDevelopement - a methodology.
(Interestingly, I notice the words 'real' and 'reality' mentioned by Jonathan Kohl from 7000 miles away at around the same time as me standing on my soap-box - we must be in ESP communication with each other).
As I took the journey home, I sketched out the idea...
Real Software Development is context sensitive, must adapt to growth in our understanding of what we are trying to deliver and evolves as we discover what works and what doesn't work. We draw on practices across numerous methodologies, like dipping into a toolbox, and settle on the combination of practices that work and are practical at a given moment in time.
Then, my thoughts were - but how do you know where to start? We need a starting point. Are there a set of core practices that we should start with on a new project then adapt? Perhaps...
If you are dealing with a legacy project, how do you know what to do no next? Symptoms! Are there symptoms for which there are common practices that often help address those symptoms?
Is it too complex to proceduralise or 'methodologise'? Probably, and thus skill comes into play...
Tonight, I am convinced that there is something we have lost... The idea of Agile Processes... a process that is adaptive and not bound to the flow of a given 'branded' methodology... Not labelled, not measured by a checklist of methods. A process that changes based on what is 'effective' in a given context.
I call this methodology "Real Software Development".... Isn't that what "Agile" was meant to be about?
There must be some way of reflecting reality... So, I wrote in my notebook - RealSoftwareDevelopement - a methodology.
(Interestingly, I notice the words 'real' and 'reality' mentioned by Jonathan Kohl from 7000 miles away at around the same time as me standing on my soap-box - we must be in ESP communication with each other).
As I took the journey home, I sketched out the idea...
Real Software Development is context sensitive, must adapt to growth in our understanding of what we are trying to deliver and evolves as we discover what works and what doesn't work. We draw on practices across numerous methodologies, like dipping into a toolbox, and settle on the combination of practices that work and are practical at a given moment in time.
Then, my thoughts were - but how do you know where to start? We need a starting point. Are there a set of core practices that we should start with on a new project then adapt? Perhaps...
If you are dealing with a legacy project, how do you know what to do no next? Symptoms! Are there symptoms for which there are common practices that often help address those symptoms?
Is it too complex to proceduralise or 'methodologise'? Probably, and thus skill comes into play...
Tonight, I am convinced that there is something we have lost... The idea of Agile Processes... a process that is adaptive and not bound to the flow of a given 'branded' methodology... Not labelled, not measured by a checklist of methods. A process that changes based on what is 'effective' in a given context.
I call this methodology "Real Software Development".... Isn't that what "Agile" was meant to be about?