I guess I'm one of the younger members of this community, but don't let that fool you!
I have applied more of a vanilla implementation of FDD last year in my final year project at uni (16 developers), and recently just some aspects of it to my last J2EE project, which i worked on by myself reporting to two managers.
I have many views and still many questions on agile development. It's funny to me how so many IT professionals are looking "the next big thing" to fly straight into their laps, specifically I'm talking about those that view agile as being Fred Brooks'(sp) silver bullet for development.
I think to expect a vanilla implementation of FDD or any of the other agile practices to give a particular organisation an advantage is naive. Mind you, like waterfall, you should try to practise the vanilla versions at least once, but I feel the key lies to being able to assimilate the practices that the organisation found value adding from each vanilla project post mortem and combine them, whilst noting what was poor. This resonates with me as I don't think any organisation can benefit by trying to "follow a standard". In the end, it's what the organisations practises that will deliver results to clients, hopefully under budget and on time. It's at this junction where you can exploit FDD's main power as a development framework; its flexibility. Unlike XP, which is fairly rigid in its 4 core values and 12 practises, FDD will scale, from a 1-2 person project to a >=100 person project (How? I will write a story on that soon!).
I haven't read all the backlogs but I'm sure it would have been mentioned. FDD can be agile; it can be whatever you want it to be. All it does is say ?Here is a roadmap for how to get your project completed on time". There are many roads that lead to on time, and which road you take will be influenced by quality of people and their understanding of technology. It's up to each organisation to take that map and find which road will give it most success for future projects.
If you're looking to FDD to be you're Holy Grail for development methodologies, then you best be looking for werewolves too. You might find one, odds you won't. FDD can give you a great head start in the crusade. It's up to each organisation to tailor it to achieve their goals.
Also, if anyone reads this and holds any sway of power in a company based in Melbourne (no I'm not looking at Jeff De Luca at all!!); please give me a job!!!!.. if anyone read the Australian IT (20/05/03) on outsourcing and the cost on Australian graduates or on the lack of jobs for Australian graduates in general, you would understand the plight of the Australian IT grad. I think like most graduates if any company is willing to invest in constant personal development of employees they will be rewarded with low turnover and a high skill set (and if you think technology or process mean more than people in the golden triangle you have really been missing out on core SE theory!). Mind you, you need quality to begin with, and hell I'm not going to brag but I have quality written all over me (right?!?!.. right?!?! ha-ha)!! I'm also wondering the long term benefits of outsourcing, especially to India, but I?ll get to that another day.
Currently, I am involved in creating an open source FDD project administration tool, as well as writing up a nice report on document driven development with FDD based on an industry Uni project last year. I?m also interested in methods of extending FDD for UI (other than Gavin Baker?s, and David Anderson?s proposals), and UI patterns for java swing (Hierarchical MVC etc).
