More Laws of Software Knowledge Process
More laws and observations

These are also included in my book and are extensions of the three laws.  I have rewritten them for the more general case of knowledge as opposed to knowledge intended to be inserted into the software medium:

• The Corollary to The First Law: You cannot have a process for something you have never done and do not know how to do.
  This is pretty evident  [1].  That said, you can have a meta-process at some higher level of abstraction.  One can always have a process for any activity as long as it is at some abstracted level of context.  Such meta-processes are of little practical use in actually directing the target activity.

• The Reflexive Creation of Results and Processes:
  1. The only way that effective results can be achieved is through the application of effective processes.
  1. The only way that effective processes can be created is through the achievement of effective results.
  These statements were originally couched in terms of building/creating (software) systems.  I've rewritten them using the more general term "results" (whatever they might be).  Basically, these statements are saying that the only way to devise an "effective" process to do something is to actually do something.  Equally, the only way to produce a result effectively, is to apply a proven process, one that has already been shown to produce the required result [2].  These statements are both mutually and internally self-referential. 

• The Lemma of Eternal Lateness: The only processes we can use now were created in doing things in the past.
  Again this was cast in terms of software development projects.  The thing that defines (or should define) a new software project is that certain elements of the project are, well, new.  If the goal of a project is to do something which has already been done, the developers should simply use whatever was already dome, right?  So this lemma only applies to new things.  If what we are doing is the same, or close, to what we have done already, we should just use that process, or perhaps reuse the results of that process.  If we are doing something really new, something we haven't done before, any process we can use really applies to something else.  It might work; but it likely won't.  

• The Dual Hypotheses of Knowledge Discovery:
  Hypothesis One: We can only "discover" knowledge in an environment that contains that knowledge
  Hypothesis Two: The only way to assert the validity of any knowledge is to compare it to another source of knowledge.
  We have already addressed this.  We can only learn from a book that contains the knowledge we need to learn [3].

We see, over and over, that whenever the word "knowledge" appears, a recursive loop is right behind it.  Such is the nature of the beast.

FOOTNOTES 

[1]   There is the "accidental competence" process, of course.  Simply working at something you have never done will (might) ultimately produce something that is  "correct".  In its extreme this is the monkeys-at-a-typewriter-producing-the-works-of-Shakespeare approach.  However, if we are tackling something that we have never done, but which is somewhat close to things we have done, we can often effectively transfer from that experience to the one we are currently wrestling  with.  The closer the new thing is to the old thing, the more effective we will likely be in this transfer.  But the converse is also true and if something is entirely new, we cannot have a prescriptive process for it.
[2]  The key word is "effective" which is doing a lot of heavy lifting here.  Effectiveness is a qualification of both/either the process and the result and is an echo of the earlier assessment of knowledge.   If we've never done this before, how would we determine if it is effective?

[3]  This references  knowledge "discovery".  But it is reasonable to asert that, in the larger sense, it is actually knowledge creation.