For an upcoming project at TOPP, we're talking about setting up an optional scope contract [PDF] -- where we specify the time, cost and quality, but leave the actual scope of work open.  This approach has many advantages, which I'll just quote from Beck & Cleal's document:

• Customers can change their minds

• Suppliers aren’t encouraged to sacrifice quality as soon as something goes wrong

• Customers’ and suppliers’ interests are contractually aligned

• The knowledge that both parties gain during the project can influence the finished product.

In my experience so far, it has been much easier to set up agreements like this in the private + nonprofit sectors than in the public sector.  Typically, public sector contracts must begin with detailed requirements (beginning with an RFP then a final scope of work), to ensure that the requesting agency doesn't get screwed over.  The problem with this approach, of course, is that you don't always know what you need at the beginning of a project, or to rephrase, that's when you know exactly the least about what you'll be making. So my question for you, internet, is have you had experience making optional scope contracts work in the public sector? // thanks Nate for turning me on to this idea at last year's Nonprofit DevSummit