The largest developed specific system theory seems the "system theory" by Niklas Luhmann which comes as a general theory but is in fact a "social system theory". I think this creates all types of jargon problems. For example he defines what a "system" is, but this conflicts with the view of an engineer or an software developer or a community founder.
Why should this be important to us? Luhmann built a theory of social systems -- and online communities are - at least in our view - social systems. But our online communities today would not qualify in his eyes, because his "system" term includes independence from individual elements and the ability to reproduce its elements. The MeatBall wiki for example might currently die, when the server is turned off, or the founder looses interest. It's not clear how a wiki will continue to exist after 10 or 20 years. How can it continue? Will people care for?
Luhmanns theory seems unsatisfying for community developers, because it starts at a very high level of "system" and "stability". An engineer or software developer will usually insist to be free to call anything a system that consists of reasonably related elements (or objects). A founder would like to talk and think about his site and growing community as a developing system. Stable or not, growth or not, reproducing or not, only the future can tell.
Another thing that looks weird, is how a system is defined. I suppose the typical engineer would describe a system by listing elements and relations. Not so Luhmann, because social systems are not that way describable. He looks to make a decision for an element whether it is in the system or in the environment. This decision is identical to a difference that you can perceive. This difference creates a boundary. Therefore (Luhmann) no system without constituting boundary. No system without an environment.
This "no system without environment" feels very uncomfortable for an engineer who is typically abstracting from the environment and looking at his systems in complete isolation.
This "no system without boundary" also it feels uncomfortable for a community founder, because he likes to think of his system as an open system, where anyone may walk in, start writing and become part of the community. But this seems a myth, wishful thinking. People are welcome, but only if they respect the existing traditions, if they share enough of the values, if they play to the rules. Otherwise conflicts are inevitable that lead to inclusion or expulsion relative to that boundary. Maybe Luhmann is not that wrong at all. Maybe online communities would work better by accepting boundaries as necessary and by clearly defining them.
ChristopherAlexander looks at systems in a quite different way. The notion of a system (as a subsystem) and its environment hardly exists. Neither does the idea of elements as simple parts or objects within the system.
Instead there is the idea of wholeness that contains all centers and their almost countless relationsships. Centers can't be identified with objects: the whole is a center, objects are centers, boundaries are centers, patterns of centers are new centers, even the void - the absence of centers - is a center of its own.
The Luhman idea of stability and reproduction is also very foreign to the world of Alexander. Systems are observed as living and developing systems. Their unfolding happens in small steps, each serving a specific purpose, increasing the value of the system, preserving the current structure as far as possible.
Patterns are observed as history-proven elements of life. The PatternLanguage is the result of this empiricial work. Structure-preserving transformations emerge as the key processes to create living systems.
Typically there is no technical system theory, maybe because such an abstraction is of philosophical nature. Systems are everywhere in the natural sciences and in technology and they are typically dealt with using causal and mathematical models. Fuzzy information or stochastic processes are integrated numerically or statistically.
A technical system theory merges well with object-oriented thinking. Systems consist of elements (objects) that are specified by properties (classes, attributes) and relationsships (interactions, methods).
Thinking is analytical in nature: one expects to understand a system by understanding its parts and their interactions. The search for knowledge tends to go into the details. Its easier to publish hard uninteresting details than soft interesting theories (TheAcademy). Lots of knowledge about how synapses work, little knowledge about how our thinking brains work.
Would be nice to have one.
Social system theory put to work:
If MeatballWiki died today, we could resurrect it from the Google cache and have it running elsewhere without too much effort. The legal right to resurrect it is hazy, though, unless we could argue we are doing so as "MeatBall" the organization and not as individuals.