Definitions of systems LO28127

From: Daan Joubert (daanj@kingsley.co.za)
Date: 03/31/02


Replying to LO28095 and others in the thread

   Hallo Bob

   In response to your request for a definition of 'system':

   Having eagerly embraced the "systems concept" back in the early
   1970's, the definition - description perhaps - that seemed most useful
   to me is:

   "A system is a collection of interrelated entities that:
   * Belong together according to a simple rule and
   * Have a definable boundary that separates it from its environment"

   Like a few of the other definitions, this one implies and outside
   observer - as if the concept of 'system' is not an inherent or
   intrinsic quality, but only becomes meaningful when recognised as
   such. (And I am not properly equipped to take this line of thought any
   further!)

   The two qualifications are near synonyms, but the second is essential
   to ensure that the existence of 'the system' in - and therefore
   relationship with - the environment of the system is acknowledged.

   As a point of interest, I cannot recollect any of the definitions so
   far mentioning that a system exists within an environment - with which
   as a general rule it will also be interacting - although some implied
   this by demanding a 'system function' in terms of the processing of
   inputs and production of outputs. (A few make allowance for infinite
   systems, that would exclude the presence of an environment. I think.)

   Now when working in a near non-philosophical environment - as a
   systems programmer - I may well have considered the question "Do you
   mean to say the universe is not a system, since it presumably does not
   meet the second of these qualifications?" completely irrelevant for
   all practical purposes (on a par with the matter of how many angels
   could dance on the head - what about the tip?? - of a pin).

   While the above definition could be faulted because it fails to deal
   with the matter of the universe, I can imagine the extended debate
   that may well erupt here if exactly the same question is also asked of
   definitions that require a system to have purpose or intent in order
   to comply. Or of those that presuppose inputs, processing and outputs.

   Note that in the above definition the 'nature' of the entities is not
   fixed - they can be discrete "units" or (sub-) systems in their own
   right.

   There is a temptation to add the phrase "and interdependent" to the
   first part of the description, but that might be overkill and just
   trigger the splitting of hairs. The same applies if one were to
   qualify the "interrelatedness" by saying there is also interaction
   between any pair of subordinate entities, directly or indirectly so.
   Rather leave the matter of what the nature of the relationships are to
   the defining rule that brings the 'system' into the realm of being.

   Should one be talking only about organisational systems - people who
   get organised in some manner - that can be assumed to have some form
   of intelligence or at least self-awareness, as some of the supplied
   definitions appear to do, then it would be very difficult to exclude
   or disregard the concept of a 'purpose' , or intent. Speculating about
   what this might be could give rise to a whole new discussion.

   But that, of course, would be a completely different thread! ;-))

   With best wishes and kind regards

   daan

   Daan Joubert
   Extending his environment
   Roodepoort
   South Africa
   daanj@kingsley.co.za

-- 

Daan Joubert <daanj@kingsley.co.za>

Learning-org -- Hosted by Rick Karash <Richard@Karash.com> Public Dialog on Learning Organizations -- <http://www.learning-org.com>


"Learning-org" and the format of our message identifiers (LO1234, etc.) are trademarks of Richard Karash.