Apr 09 2009
April 27th, central London: What is FitNesse and should I use it?
I’ll be doing an introductory talk on FitNesse on 27th of April in central London.
I’ll introduce FitNesse, explain its basic features and present when you should use this tool, when you shouldn’t and how to use it efficiently.
The event is free but up-front registration is required. for more information and to register, see:
http://skillsmatter.com/event/open-source-dot-net/what-is-fitnesse-and-should-i-use-it
![]() |
![]() |



Hi,
A mutual person we both know(but who which I prefer to be very disassociated from) setup fitnesse where I work. The attempt was to try and run our entire integration testing primer on the said platform using Cruise Control. It was ropey at best. Do you think this is was right approach? It has left a foul smell in my mind when regarding fitnesse i.t.o. of capability and maturity but hindsite tells me that it is not the product, it is the person who sold it the wrong way that did the damage.
~f
i think that every team and project is unique so i wouldn’t be able to comment on that without actually seeing what the project is about. fitnesse is very good for business logic tests, and if the complexity of the project and integration tests are in the service layer, that makes sense. if the complexity is in the ui, then it probably doesn’t make a lot of sense. again, hard to say without actually knowing anything about the project.
The project uses an MVC(JQuery+ASP.NET), with IoC(Castle Windsor) and DDD+TDD. It was a royalty posting system which used excel and email through web interface with a separate windows service for background service processing. Simply put it has the ability to generate statement templates, mail them, then receive filled out financials and then do file revisioning using Check In/Out through the interface after which the data is posted up in a P+L. It felt like fitnesse was going to be used as this big stick to assert the correctness of all functions from the business end, but the way it was going we eventually found ourselves writing non-production code features to inteface with the fit tests just to get them to work. It was also very data-centric. This was time consuming and the tests were extremely fragile through cruise control. Something did not feel right and it was taking too much time, so I chucked out Fitnesse + Cruise, installed Team City, rewrote the integration tests using NUnit and run them nightly on a separate build configuration. After all that I still feel like my bias toward fitnesse is as a result of the wrong approach not the product.
then come to the event and see where fitnesse really works
I will be the guy standing in the back with a bazooka …