To date, talent management's school of thought appears to premise systematic approach to explain how archetypes like traits (characteristics, personality, and strengths) allow each individual to capture, interpret, and utilise knowledge differently. It is the academics in consulting firms rather than academic institutions who have offered some practical tools. You may read the following articles.

Article Talent management in Thai cement companies: A study of strat...

Dear Veerendra,

I think it is possible to design these archetypes as frameworks containing critical knowledge for a certain domain. In my view, such a semantic construction should be characterized not by the quantity of knowledge stored like objects, but by the quality of filtering and integrating the essential knowledge for understanding problems from that domain. Knowledge is a nonlinear entity and it would be useless to amass a large quantity of knowledge without a good semantic structure and critical knowledge. The initial efforts to build artificial intelligence programs failed because people made no difference between linear and nonlinear processes.

Also, in each domain there is a certain knowledge dynamics and a certain life cycle for knowledge people should account for. The archetypes should reflect this dynamics. More important, there are three fundamental fields of knowledge which interact continuously: rational field of knowledge, emotional field of knowledge, and spiritual field of knowledge. These archetypes you are talking about should not contain only rational knowledge. Adaptation to contexts, as you mention should take into consideration cultural values and social differences coming from emotional and spiritual knowledge. Finally, it is important to switch from the well-known "stocks and flows" and "iceberg" metaphors to "energy" metaphor. Attached is a book I published recently in which I present this new perspective on knowledge fields and their interaction.

Dear Prof. Constantin Bratianu, I agree with you on quality of filtering knowledge for understanding problems in the domain and creating good semantic structures. I also intuitively agree on nonlinear nature of domain knowledge and futility of accumulating knowledge without these considerations. Can you please elaborate on non-linearity of domain knowledge? I need to appreciate it fully. Thanks for your answer.

In marketing communications and social media, it is important to segment our customers (and other groups with which we wish to communicate) into "personas."  These personas describe a typical member of the particular market segment, including demographic and psychographic information (interests, lifestyle, connection to our organization, etc.).

These descriptions can be very valuable in crafting our interactive communication with the group/segment.

Knowledge is the combination of symbolic (processing of symbolic representations), and non symbolic (processing of neuronal connections) capabilities. The former uses symbolic categories to build explicit knowledge, the latter uses weighted connections to compile implicit knowledge. From a computational point of view archetypes can be applied to both aspects.

https://caminao.wordpress.com/what-is-to-be-represented/knowledge-architecture/

Dear Veerendra,

In simple terms, linear processes are characterized by the fact that outcomes are proportional with inputs. For instance, a salary based on the the following formula  Quantum ($/hour) x Number (working hours) is linear. A salary based on negotiations in well-defined contexts is nonlinear. The metaphor for nonlinear processes is the butterfly effect: a butterfly produced some perturbations of the air in New York and they will generate a tsunami in Tokyo. For a more elaborate explanation please read the attached paper I published some years ago. If you have some other questions I shall be happy to answer to them.

Dear Prof. Constantin Bratianu,

I understand non-linearity in general being a systems engineering person. I requested to know non-linearity of knowledge. Please elaborate. thanks

Dear Veerendra,

For knowledge, we should start from the metaphor used for defining knowledge. According to Andriessen & Boom: "Knowledge is not a concept that has a clearly delineated structure. Whatever structure it has it gets through metaphor. Different people from different cultures use different metaphors to conceptualize knowledge. They may be using the same word; however, this word can refer to totally different understandings of the concept of knowledge" (Andriessen & Boom, 2007, Knowledge as love. How metaphors direct our efforts to manage knowledge in organizations, Key note lecture at 8th European Conference on Knowledge Management, Barcelona, Spain). 

When knowledge is based on the metaphor "knowledge as objects" then the linear rules that apply to physical objects are transferred to the knowledge and knowledge can be accumulated and stored like objects. When knowledge is considered as a field, coming from the metaphor "knowledge as energy" ( a metaphor proposed by Andriessen and myself) knowledge cannot be amassed like physical objects but it should be integrated in the semantic field that already exist. If necessary, the semantic field can be re-structured such that the new configuration to reflect much better the reality we describe. In this case, knowledge is treated as a nonlinear entity. 

Dear Veerendra,

While I agree with what Constantin (overview of his thesis) has suggested, I don't think that non-linear definition is where knowledge fits. What you're seeking to do is the condensation of information into bodies of knowledge that, at a lesser amount of time, yield some form of decision-affecting value. You can have rows and rows of data, when a query is invoked, the data turns into information (various levels of filtering). This is a common idea in computing, as you probably know.

I would, personally, advise you to stay away from semantics (termed information sciences) and specific ontology definitions that have their own thesauri of some kind, unless they can be standardized (or you just need to do so). If you still go with that approach, you will probably be better suited in the adoption of an existing body of knowledge because whatever you do from start will probably result in ambiguities due to non-standardized definitions.

So, for a knowledge base hypothesis, where you begin is where you form that common ground.

Based on my studies, I don't think knowledge is linear (this is where I differ from Constantin) - in computing, we work with data (abstract data that does not need to make any sense), we convert data to information with structures for context while this information is converted to knowledge by certain filtering. As such, if you argue that archtypes are key to understanding this body of knowledge, it still needs some form of context since to truly be able to rely on some form of blue ocean strategy on knowledge structures is not, in my opinion, possible. Particularly cross industrial application-wise, it seems undoable of not very very difficult.

So, I would argue against the premise that knowledge is non-linear - rather, knowledge is linear because data is an abstraction, information is non-linear and knowledge, by the time it undergoes the filtering process, becomes linear for effective decision making.

I would, again, advise that you stay away from semantics unless they are certain sources that you think are key to your formulation of whatever theory you intend to undertake. 

IIn my experience, I noticed that sharing knowledge into the group requires more than intellectual linkage between members: it requires a special kind of common identity with the object to be exploited. Simply put, the members must share a deep understanding about the object, its importance and potentiality. Understanding the ideology and symbolism that underlies the object, in several domains, will make it possible to create new solutions, products and services.

Dear Veerendra,

I agree with arguments presented by Santosh, but only within the paradigm of INFORMATION conceived by Claude E. Shannon: "The fundamental problem of communication is that of reproducing at one point either exactly or approximately a message selected at another point. Frequently, the messages have meaning; that is they refer to or are correlated according to some system with certain physical or conceptual entities. These semantic aspects of communication are irrelevant to the engineering problem" (C.E.Shannon, A mathematical theory of communication, Bell System Technical Journal, 27, 379-423 & 623-656, 1948). Thus, Shannon used the concept of INFORMATION as a pure mathematical entity, being devoid of any meaning, and having with the energy signals used for message transmission only a statistical correspondence. He states that very clearly: "If the number of messages in the set is finite then this number or any monotonic function of this number can be regarded as a measure of the INFORMATION produced when one message is chosen from the set, all choices being equally likely" (C.E.Shannon, idem).

I used a different paradigm, that of the Knowledge Management. In this paradigm, the central concept is KNOWLEDGE, which means a result of processing INFORMATION and DATA, with respect to a field of meanings or a certain semantic. In this paradigm, the knowledge interpretation comes from the metaphors we use. When we use the concept of knowledge field, then we should consider nonlinearity as an important attribute of knowledge.

Thus, it is necessary to define first the paradigm you are going to use and then to start building up archetypes. These should be consistent with the main hypotheses used in defining the paradigm. However, we should make the difference between the Information Systems Management and Knowledge Management. My previous posts are consistent with the paradigm of Knowledge Management.

Dear Veerendra,

I agree with arguments presented by Santosh, but only within the paradigm of INFORMATION conceived by Claude E. Shannon: "The fundamental problem of communication is that of reproducing at one point either exactly or approximately a message selected at another point. Frequently, the messages have meaning; that is they refer to or are correlated according to some system with certain physical or conceptual entities. These semantic aspects of communication are irrelevant to the engineering problem" (C.E.Shannon, A mathematical theory of communication, Bell System Technical Journal, 27, 379-423 & 623-656, 1948). Thus, Shannon used the concept of INFORMATION as a pure mathematical entity, being devoid of any meaning, and having with the energy signals used for message transmission only a statistical correspondence. He states that very clearly: "If the number of messages in the set is finite then this number or any monotonic function of this number can be regarded as a measure of the INFORMATION produced when one message is chosen from the set, all choices being equally likely" (C.E.Shannon, idem).

I used a different paradigm, that of the Knowledge Management. In this paradigm, the central concept is KNOWLEDGE, which means a result of processing INFORMATION and DATA, with respect to a field of meanings or a certain semantic. In this paradigm, the knowledge interpretation comes from the metaphors we use. When we use the concept of knowledge field, then we should consider nonlinearity as an important attribute of knowledge.

Thus, it is necessary to define first the paradigm you are going to use and then to start building up archetypes. These should be consistent with the main hypotheses used in defining the paradigm. However, we should make the difference between the Information Systems Management and Knowledge Management. My previous posts are consistent with the paradigm of Knowledge Management.

Dear Veerendra,

I agree with arguments presented by Santosh, but only within the paradigm of INFORMATION conceived by Claude E. Shannon: "The fundamental problem of communication is that of reproducing at one point either exactly or approximately a message selected at another point. Frequently, the messages have meaning; that is they refer to or are correlated according to some system with certain physical or conceptual entities. These semantic aspects of communication are irrelevant to the engineering problem" (C.E.Shannon, A mathematical theory of communication, Bell System Technical Journal, 27, 379-423 & 623-656, 1948). Thus, Shannon used the concept of INFORMATION as a pure mathematical entity, being devoid of any meaning, and having with the energy signals used for message transmission only a statistical correspondence. He states that very clearly: "If the number of messages in the set is finite then this number or any monotonic function of this number can be regarded as a measure of the INFORMATION produced when one message is chosen from the set, all choices being equally likely" (C.E.Shannon, idem).

I used a different paradigm, that of the Knowledge Management. In this paradigm, the central concept is KNOWLEDGE, which means a result of processing INFORMATION and DATA, with respect to a field of meanings or a certain semantic. In this paradigm, the knowledge interpretation comes from the metaphors we use. When we use the concept of knowledge field, then we should consider nonlinearity as an important attribute of knowledge.

Thus, it is necessary to define first the paradigm you are going to use and then to start building up archetypes. These should be consistent with the main hypotheses used in defining the paradigm. However, we should make the difference between the Information Systems Management and Knowledge Management. My previous posts are consistent with the paradigm of Knowledge Management.

If you approach this issue from a knowledge management perspective then you can begin to understand the difference between tacit and explicit knowledge-

http://www.knowledge-management-tools.net/different-types-of-knowledge.html

You should be able to capture the explicit knowledge pertaining to the set of domains that you are interested in if you are aware of the essential steps that need to be followed. However capturing all of the tacit knowledge is a lot more difficult and I would suggest this is the greater of all of the challenges that you face.

I agree with Deborah in making the difference between explicit and tacit knowledge. The tacit dimension has been introduced by Michael Polanyi: "I shall reconsider human knowledge by starting from the fact that we can know more than we can tell" (Michael Polanyi, The tacit dimension, Peter Smith, Gloucester, Mass., 1983).

Dr. Deborah / Dr. Constantin, Is tacit knowledge a damper on the idea of capturing domain knowledge in archetypes? 

Perhaps the explicit/tacit distinction corresponds to the symbolic/non symbolic distinction I mentioned earlier.

Dear Veerendra,

I don't that archetypes are related only with knowledge. Knowledge is more than that.

To understand a phenomena you must be opened to it and imerge on it. Doing it, it will be up to you to realize if the archetype is related or not with the knowledge. 

Afonso

Hi Veerendra

Yes managing tacit knowledge regardless of the domain is the greater challenge and or barrier for knowledge management initiatives.

Deb

Dear Veerendra,

I would like to introduce a rather different perspective from the interesting debate above. I come at the issue from the perspective of a practitioner. In the past I have wanted to talk about the activities of my practice domain, operational research (OR). OR is a consulting activity and I suggested three archetypes covered much of what we do: 'smart bits', 'helpful ways', and 'things that matter'. Smart bits alludes to the development of mathematical solutions to such problems as scheduling, stock control and logistics management. Helpful ways involve more general investigations into issues and problems to try to find ways of addressing them in some way. Things that matter are projects that try to help to contribute to the great social and environmental issues of the day. Each involves a knowledge of technology, of intervention process, and societal context but the emphasis in each archetype is different. Both knowledge and crucially knowhow is involved. The archetypes have no theoretical underpinning but just seem right to those involved - it therefore has proved useful in communication and shaping discussion. So in answer to your question, I would suggest you should consider (i) 'characterising' (or some such fuzzy concept) as well as 'capturing' (often a rather too demanding task), and (ii) adding knowhow to knowledge (practitioners, as I am sure you know, usually trade on knowhow and often rely on the client for knowledge).

Richard (ex consultant)

Hi, I've used ontology to achieve a kind of system that from documents in multiple types which are processed and matched with the ontology. Please see my contributions.

Yes, there is a need to capture knowledge of domains to create applications, products, services and platforms to serve customers and end users. This process for capturing the domain knowledge and reusing them into applications is simple but not easy. Lot of work and resources are required. 

I have some experience in capturing domain knowledge in the context of Product development. The potential is high and returns are significant, however most of the knowledge techniques (like KBE and KCR are still limited to family of parts. 

Take a look at one such application...

Conference Paper Knowledge-based Enterprising (KBE) Strategy for Lean Product...

In some sense, it sounds like you are talking about "General Systems Theory" - and the general idea from systems thinking that we might be able to identify ways of thinking/understanding that are common across all disciplines. 

Second, we've been using a gameified approach to surface, clarify, and co-create knowledge in the form of the ASK MATT game (award-winning paper attached). We find this works great for strategic planning because it gets all the ideas on the table. And, instead of searching for the best and ignoring the rest, it integrates all understandings to provide an immediately useful action map.

Veerendra, hello. I suggest you give it a revised dissertation Rene Bedard HEC Montreal, especially his thesis on Philosophical Rhombus provides a methodological framework that can be of great interest for research. Cheers

Archetypes are the products of systematic analyses of systemic behaviour. We tend to accept through "halo effect" that every systemic behaviour can be approached through systematic approaches and outcomes would accrue. At times the results through this approach may not yield the desired results. Thus archetypes need to be carefully designed for better life cycles and of course reuse. Unfortunately however, this is not easy. 

Why can't a software application, or a business process map, or Brian Prasad's methodology (see above) each be archetypes of domain knowledge? Each captures re-usable knowledge to some extent. 

Mark - that sounds reasonable. Perhaps the key is to evaluate how much re-usable knowledge each archetype captures?

But what determines 're-usability'? I mean, it all falls down to the process of discretion ultimately. If you need judgment to define parameters for knowledge application, you are probably going to use a computational environment to break it up and discover information in context. If you really need to drop down to that level, why not stick to human intuition that is faster (if application expert is experienced), cost effective and quick than formulation of such, so-called, archtypes that will just not be adaptable to cross-industrial scenario?

And even human input, do you really need a model for that? Imagine the documentation!

The entire trouble with such definitive models is that they need a methodology, which needs application expertise, which again relies on human discretion. With automation tools for computation that needs referencing, real time calculation is a complex process. Can a software even perform ever changing algorithms at real time for aids in decision making (knowledge archtypes)? And how reliable will it be... at all? I personally think that the premise is one of those things that look good on paper... in industrial application, it may prove to be a theoretical 'good to have'... at best. Not to mention, the transition of application know-how between experts is another problem. You need semantics, established vocabulary that every subject expert agrees with... major major engagement. 

Santos, I would suggest that there is no point trying to capture the tacit knowledge that is required from the expert human that interacts with a piece of software or methodology. Knowledge archetypes can perhaps only capture explicit sub-sets of knowledge.

The parameter variables in a software "archetype" would need to be set up to suite a company in a domain (by a human). The fixed data structure would hopefully be flexible enough to work across domains. 

ERP companies like SAP try and create software that works across all companies in a domain (but I suspect there is a lot of messy customisation judging by poor track record of projects). Perhaps my question should be: Is explicit software that successfully works with humans in a domain not a knowledge archetype? After all some human expert shared their knowledge with the programmer.

You make a good point about the need for speed (which would suggest that intuition would be the way to go). However, while intuition is useful for making quick decisions, it is not so useful for making successful one. Also, you make a point about the information being in a particular context. I've found is that the most useful knowledge is self-contextualized. For example, with Ohm's law, changes in volts, amps and ohms are understood in relation to each other. The application then becomes obvious - one uses the self-contextual knowledge only in those cases where the concepts relate to the situation (one having amps, volts, and ohms).

Hi Mark, you've probably hit the nail with that question. And even with industrial focus, the application of a computational aid is very limited in adaptation. The even the best software do not adapt to an industrial operational pattern, it tends to hard code best practices with industrial input and then provides another (separate) solution to process automation - for which change management becomes a critical factor to manage. Sure, not all changes are bad, but given the extent of changes, a software system, even if engineered with these knowledge archtypes, will end up needed adaptation, often at an Enterprise scale. So, in the end, the process of computational automation is for the sake of computational automation rather than benchmarking or access to formal bodies of knowledge (with these  reference models). Perhaps, this idea is suited to other industries, F & B? Or, Logistics (without automation or efficiency layers applied). 

Hi Steven, noted on your findings. The same can be said of relativity. That explanation is a very good one actually. But then, we are probably looking at measurable attributions. I wonder if quantification of factors is a recommended approach, especially across such widely applied bases. I mean, what are the parameters? Surely, not all factors can be attributed to quantifiers? Or can they? Especially given intuition or cognitive structures - they are just too relative.

Dubery, friendly greeting. Very interesting your question, from my research experience, I believe that if you can capture the knowledge, not only from the archetypes, but also from social representations. We recommend reviewing the doctoral thesis "Facilitators of organizational knowledge creation",EAFTIT University, Colombia,  this paper develops a theory approach Sergio Moscovici, and although it does not resemble Jung, may be evident that if possible.

Santosh - yes - attributes/variables/concepts/things/processes/dimensions should be scalable. That is to say, we should be able to count them or understand them as existing on some scale of lesser to greater. It is a poor option to say that there is just one of something, or that it is an absolute (e.g. "life is everything" - perhaps not a great example... I'm finding it more difficult to think "atomistically").

There is another part of that however, It is useful to hold open the possibility that something may be an important part of the theory even though we are not (yet) able to measure it. So, in evaluating theory construction, I allow for a "black box" (of course, as above, it should be scalable). Naturally, having too many black boxes makes the theory meaningless, so I recommend that each black box have inputs and outputs (variables upstream and downstream of causality) that are measurable. That way, we can gain insight into what is "in" the black box.

Thanks,

Steve

Hi Juan, Noted with thanks.

Hi Steve, noted, but the quantification of factors poses the issue of individual impact, right? I mean if decision-making is to be quantified, then we have the issue of the impact each of these factors on the model's product. How do you demonstrate this idea, which makes sense in the qualitative area though.

I think the latter factor (what you call a black box) can be percentile based significance, probably a constant that can be defined, it is understandable. But then, that's my issue with archtypes, you know, the need for static variables... when the strength of each attribute can differ across subjects. Even if we do allow for dynamic attributes, we will still be left with ambiguities... which ultimately require some form of discretion. Defeats, the purpose, right? How do you get around that?

Santosh - Generally, the IPA method is used to evaluate the potential usefulness of a conceptual system (e,g, mental model, theory, etc.). The measure of Systemicity (Depth / interconnectedness) suggests the level to which the model will be useful for decision making. However, that only applies for the concepts that are explicitly within the model. Any concept/variable/factor that is not inside the model would - yes - by necessity be left to discretion.  

It would certainly be possible to build a model with more concepts (the Complexity or Breadth measurement of IPA). However, it would be difficult or impossible to create a model containing all concepts - although at a very high level of abstraction it might be possible (although the usefulness of such a highly abstract model might be questionable).

Then, there is the "emerging unknowns." That is to say (related to Gödel's proof) if one were to create a model of everything, that new model would increase the number of things in the universe - and would not be taken into account by the model. Also, that new model would allow more effective action, thus leading to more things in the universe - also not anticipated by the model.

Someday... we may develop a model which contains itself? Perhaps one where the variables are non-static?

Thanks,

Steve

Many will think that contrary to information, knowledge by nature cannot be reused ...

Remy - is that knowledge? How many times has it been reused?

So, let us posit that there is information (data) that is disconnected (as in a database or telephone book). And there is knowledge (that is interconnected as in a theory). Neither one is possible without the other - they are two sides of the same coin.

The third dimension is application, which (due to changes in time and space) is never the same thing. Although, it seems that the times and spaces are similar enough that we can bridge differences through the use of inference - or perhaps a theory combined with data. And, of course, we can make mistakes and learn from them which provides more information and more knowledge!

Thanks,

Steve

Steven,

"let us posit that there is information (data)"

Disagree. Data has to be processed into information which can then be put to use by agents acting in given contexts according to specific purpose.

Remy. A difference of verbiage. Some distinguish between data and information, others between data and knowledge.... Which is why I use multiple terms to clarify.

Another way to gain greater clarity in these kinds of conversations might be to use diagrams. For example: https://insightmaker.com/insight/35235/Arango-Thesis-of-Information-a-Workspace-for-Collaborative-Structuration

By placing each concept (variable) in a box, and identifying causal relationships between them, we can more easily communicate our understandings... and evaluate the model as a whole to create an improved understanding.

Thanks,

Steve