Agile Index?

A couple of weeks ago Ron, Glen, and I had a back-channel conversation that I'd like to introduce parts of here for discussion.  We were talking about how better to discuss the adoption of XP / Agile in various domains / contexts.

Bill Walton wrote:

> I've been thinking about coming at the context /
> domain question from a different angle.  What 
> I'm thinking is a CMM-like ranking where, on the 
> low end we've got a centralized C&C structure, 
> on the high end we've got decentralized, self-
> organizing, XP-like principles, values, and 
> practices throughout the business, and in the 
> middle we've got various mixes that suit a 
> business at a particular point in time depending 
> on a bunch of issues we'd try to identify.  
> XP is XP.  We don't talk about modifiying XP.  
> We talk about modifying / ranking business by 
> Agility level.

Ron Jeffries wrote:

> It's an interesting idea. I don't necessarily believe 
> that "XP is XP" but whatever it is it's not a thing 
> that one modifies, I suspect. Agility level is 
> interesting as a "ranking" idea. (Probably don't 
> want to call it ranking, now that I think of it, since 
> that suggests that more agile is better. While I 
> suspect that more agile IS better, I question 
> whether it's good marketing in the circles you 
> are trying to influence to say so.)

Bill Walton wrote:

> I know what you mean with the "ranking" 
> implication.  I've wondered about that for a while 
> with CMM, in fact.  I remember reading studies 
> that said that going from CMM L1 to L3 produces 
> productivity improvement, but continuing forward 
> from L3 to L5 takes that downhill.  Always 
> wondered about the marketing aspect of that.
> Also have had the feeling (never seen it explicitly 
> stated) that what L1->L5 is really about is insurance 
> against a specific kind of risk.  Never seen it named, 
> though.  Anyway, I thought it might be interesting to 
> examine agilie practices in that same light, perhaps 
> making the risk curve explicit.

Are any of you aware of any work that's been done along these lines, i.e., characterization of either development methodologies or business cultures / structures / etc. in terms of agility / risk "curves"?

Would a discussion of that topic be of interest to any of you here?

Best regards,
Bill