I've been thinking about this problem recently -- and it's a hard one -- because we've got a corporate client with a "decision maker" who's got no technical background, but keeps making technical decisions anyway. He does this by listening to advice from me and other technical consultants, and then ignoring that and doing whatever his old buddies at the DoD suggest.
The problem is further compounded by the fact that you can gather together four different technical generalists, ask them a specific question about your project, and get four different -- but equally correct -- answers.
For example, a good one might be: "Should I use Ruby on Rails, Python, WSGI, Java, or PHP for my project?" There's not really much in the way of technical reason to use any of those methods over the others; it comes down to a matter of personal preference on the part of the technologist.
Trying to resolve the matter through discussion is probably out, since the discussion is likely to devolve into arguments over minutiae between the different approaches.
So, I can't think of any way for a non-technologist to ask for technical advice from anyone, and get the answers they really want. It's too much a matter of dabbling in a field where you don't even know enough to know what you don't know.
You might try this instead: write up an RFP that doesn't spell out the technologies that should be used. Instead, spell out your requirements, and be specific in terms that make sense to you:
* The features listed must be N% complete in X days;
* Total development time and cost should be Y and Z;
* The project will be maintained at J rate over K months, meeting L, M, and N requirements...
...Then shop that around. If a technical person thinks they can meet those requirements using their favorite technology, then let them have at. What does it really matter to you if they use PHP or Java, so long as it works?
Set up a contract that gives both of you escape clauses early on in the development project, with minimal losses. e.g., "The developer must complete W feature by Q date, to project manager's satisfaction, or the project manager may choose to abort the project and pay the developer $ amount for their time so far."