1) The main product of knowledge work is not physical …
As to this whole item, be careful. “Knowledge work” is an intentional paradox: wherever a task has not been automated with a code-running robot — i.e. wherever there is work — there is knowledge work. It’s a PARTNERSHIP — between Big Perspective and Small perspective (e.g. the inventor of the assembly line, and its workers.) Every time you see Stupid Work that you don’t think is knowledge work, ask yourself, “1. If these ignoramuses wanted to really fuck up and cause destruction, how could they do it? 2. What system (of surveillance, communication, etc.) holds them accountable so that I don’t see that happening more often? (the point is, knowledge work is a partnership between KNOWLEDGE-work and knowledge-WORK.)

Ch. 26 of “the essential drucker” has the subheadings I’m looking for (Form and Function/From analysis to perception).

* The right answer will depend on the exact purpose for which the knowledge will need to be used. Any and all of the above are correct under different circumstances

Haha — well, you’re right, by definition; that should have been the clue that you hadn’t yet defined knowledge work usefully. It’s fine to define a word by context, but only if you take as given, a set of ways in which the word can and cannot be used. …

2) No task is ever complete and the answer will continually evolve …

I absolutely agree with you, but I prefer different analogies. Look how I defined knowledge work as a partnership between [nearsighted + work] & [farsight + work.] When you’re a member of a team (a network, a corporation, whatever), you have to make certain assumptions about what your teammates know, see, value, etc.: I call it “the fundamental problem,” “the mind reading problem.” This is where the liberal education comes in. Here’s an excerpt from something I wrote today, that you’ll enjoy:
* Freud: Socrates, why are you attracted to young boys?
* Socrates: The truth is just the opposite, Sigmund — I’m satisfying my desires with precisely those people that I find UNATTRACTIVE!
* Freud: Why?
* Socrates: … so that I can be SURE that what attracted me to them, was not their great beauty, but their great sense of justice. I’m controlling a variable.
* Freud: Which variable?
* Socrates: Here’s the truth: I’m controlling beauty, in order to figure out justice.
I like that last sentence: “Here’s the truth: I’m controlling beauty, in order to figure out justice.”

On second revision: maybe that’s what you mean by “three dimensions” below: a human’s actions can be understood only along one axis; he can only understand his actions by understanding them along one axis; when either viewer or actor engages in this process, he must tell himself (falsely, of course) that he’s holding the other variables constant.

3) The “rightness” of answer is highly subjective and dependent on each individual
… The more “knowledge-based” a problem is, the more ambiguous the answer will be

Look, my concern is that when we’re talking about “knowledge work,” we’re talking about teamwork, and so the words “question, answer, knowledge, subjective, ambiguous,” etc. are ALL in need of definition; so I don’t think we’re ready yet to use one to define the other.

I don’t know whether knowledge-basedness is a function of a problem. I need to interject here and say that I see two kinds of problems: “How do I create value for the customer?” and “how do I create value for the shareholder?” It’s just another way of rephrasing the problem that faced Rousseau, How do we organize selfish men to do good for others? I think, at BCG, you’re in a unique position, in that your customer’s customer is not your customer, so it’s easy, as you’ve said, for you to blend the two … I feel that Drucker would look at a lot of the “problems” you solve, like “who are the top 10 competitors in the industry,” and say, “who benefits from a right answer?” If the answer to that is, “An institutional investor” or “some shareholders,” then I hesitate to say that the original question was a “problem.”

4) The fact that a knowledge-based “work product” has a subjective value implies the importance of the relationship between participants of this transaction

Uh oh, “subjective value.” Just like “ambiguous” from the paragraph before. Look, I thought all value was subjective — I thought revealed preference was like the smoke rising from a distant fire —

“The exchange becomes highly trust-based”? Okay, so you’re saying that it’s much harder to verify the quality of a product that consisted of an genius’s research, words, analysis, etc than one that is made of steel and plastic, like a car, or even a shirt or an apple? Starting from the premise that you’re wrong and working from there, I’ll say that verifiability can’t be different in the two cases: the details that went into the car get verified when the car explodes and you die; those in the top-ten-research happen when your client’s company crashes and dies. The long and short is, how do you distinguish trust from any other force legitimizing the Division of Labor? Even if some things are more verify-ABLE by the college-educated customer than others, the whole POINT of the division of labor is that the same guy who constructed it a product is its least-cost verifier, even if for certain products the cost imbalance varies; but his verifying the quality gets absorbed into the sale price; on a higher level verification, verification that he’s DOING that takes the form of reputation in a free market (e.g. explosions of cars, evidence of … gee, how is BCG heled accountable anyhow?? is this a joke?)

What’s this “three dimensions” concept in 6)? Is it like the thing I said above about truth-justice-beauty, i.e. you have to pretend to hold variables constant?

Here’s how I think Drucker’s mind worked: He makes a recognizable dichotomy, and then empirically lays it upon the world in a way that accomplishes his goals. The most prominent example, to me, is “Value only exists outside a corporation. Inside a corporation are only costs.” This isn’t TRUE. It’s a choice he made. Peter Drucker was fundamentally an ethical man, and that’s how he was able to convince other people to accept his categorizations of functional parts — he spoke the language of greed to people who only spoke that language of greed, but he spoke ethical words to people who wanted to think that what they were doing was good.