Ok, well I've been studying and doing this methodology thing for 9.5 years now, and just named some limits for it. Not that you didn't already know that, it's just hard to know how to say it sometimes.

BertrandMeyers talked at TOOLS USA '98 about CodeIsm, ManageIsm, MethodIsm (to which I add ToolsIsm and PeopleIsm).

Anyway, I've been writing about the effect of people in projects now for a bit, and only just detected that methodologies describe ''roles'' on projects...  but ''roles'' don't show up on project teams, people do. And the people may not have the necessary characteristics for their roles, or they might just not like each other. I only detected this in 1999 when studying some really good methodology writing. The only thing that made those writings not practical in like was that they didn't mention the effect of individual people. So I'm starting to fix this in my own work, and putting it as its own factor in the CrystalMethodologies.

I find it refreshing to be able to state the limits (as well as the benefits) of Methodology. -- AlistairCockburn