Limits of Category Theory and Semiotics
They are wonderful tools to explain much of our world, but lack ‘The Right Stuff’ to handle the metaphysical underpinnings of anything near a Philosophy of Mind, Philosophy of Language , or a Philosophy of Learning.
This is, because Category Theory specialises on roughly half of the Noosphere. It does a wonderful job on exteriority, but cannot sufficiently describe nor comprehensively access interiority.
Therefore, as is the case with Semiotics, has limited metaphysical value with respect to philosophy in general.
For example: philosophies of mind, language, or learning are not possible using only category theoretical tools and/or semiotics.
Here is an example of one attempt which fails in this regard: http://nickrossiter.org.uk/proce…
(and here: VisualizationFoundationsIEEE)
Here are two problems (of many) in the paper:
4.4.2 Knowledge is the Terminal Object of Visualisation states:
“The ultimate purpose of the visualisation process is to gain Knowledge of the original System. When this succeeds (when the diagram commutes) then the result is a ‘truth’ relationship between the Knowledge and the System. When this process breaks down and we fail to deduce correct conclusions then the diagram does not commute.”
I want to also comment on Figure 3 (which also exposes missing or false premises in the paper), but I will wait until I have discussed the assertions in the quote above which the authors of this paper reference, accept, and wish to justify/confirm.
1) The purpose of a representation is NOT to gain knowledge; rather, to express knowledge. Also, truth has nothing to do with knowledge except when that value is imposed upon it for some purpose. Truth value is a value that knowledge may or not ‘attend’ (participate in).
1a) The ‘truth value’ of the System (‘system’ is a false paradigm [later, perhaps] and a term that I also vehemently disagree with) does not always enter into the ‘dialogue’ between any knowledge that is represented and the observer interpreting that knowledge.
2) The interpretation of a representation is not to “deduce correct conclusions”; rather, to understand the meaning (semantics and epistemology) of what is represented. ‘Correct’ understanding is not exclusive to understanding nor is it necessary or sufficient for understanding a representation, because that understanding finds expression in the observer.
2a) ‘Correct’, as used in this paragraph, is coming from the outside (via the choice of which data [see Fig. 3] is represented to the observer) and may have no correspondence (hence may never ever commute) whatever to what that term means for the observer.
The authors are only talking about ontologies. That is a contrived and provincial look at the subject they are supposing to examine.
There may (and usually are) artefacts inherent in any collection and collation of data. The observer is forced to make ‘right’ (‘correct’) conclusions from that data which those who collected it have ‘seeded’ (tainted) with their own volition.
‘System’ (systematising) anything is Reductionism. This disqualifies the procedure at its outset.
They are proving essentially that manipulation leads to a ‘correct’ (their chosen version) representation of a ‘truth’ value.
I could tie my shoelaces into some kind of knot and think it were a ‘correct’ way to do so if the arrows indicate this. This is why paying too much attention to a navigation system can have one finding themselves at the bottom of a river!
The paper contains assumptions that are overlooked and terms that are never adequately defined! How can you name variables without defining their meaning? They then serve no purpose and must be removed from domain of discourse.
Categorical structures are highly portable, but they can describe/express only part of what is there. There are structure, dynamics, and resonance that ontology and functionalism completely turns a blind eye to.
The qualities of Truth, Goodness, Beauty, Clarity,… (even Falsehood, Badness, Ugliness, Obscurity,…) can be defined and identified within a knowledge representation if the representation is not restricted to ontology alone.
In order to express these qualities in semiotics and category theory, they must first be ontologised funtionally (reduced). Trying to grasp them with tools restricted to semiotics and category theory is like grasping into thin air.
That is actually the point I’m trying to make. Category Theory, and even Semiotics, each have their utility, but they are no match for the challenge of a complete representation of knowledge.
Universal Constants, Variations, and Identities #19 (Inverse Awareness)
Universal Constants, Variations, and Identities
#19 The Inverse Awareness Relation
The Inverse Awareness Relation establishes a fundamental relationship in our universe:
Is Real World Knowledge More Valuable Than Fictional Knowledge?
No.
Here an excerpt from a short summary of a paper I am writing that provides some context to answer this question:
What Knowledge is not:
Knowledge is not very well understood so I’ll briefly point out some of the reasons why we’ve been unable to precisely define what knowledge is thus far. Humanity has made numerous attempts at defining knowledge. Plato taught that justified truth and belief are required for something to be considered knowledge.
Throughout the history of the theory of knowledge (epistemology), others have done their best to add to Plato’s work or create new or more comprehensive definitions in their attempts to ‘contain’ the meaning of meaning (knowledge). All of these efforts have failed for one reason or another.
Using truth value and ‘justification’ as a basis for knowledge or introducing broader definitions or finer classifications can only fail.
I will now provide a small set of examples of why this is so.
Truth value is only a value that knowledge may attend.
Knowledge can be true or false, justified or unjustified, because
knowledge is the meaning of meaning
What about false or fictitious knowledge? [Here’s the reason why I say no.]
Their perfectly valid structure and dynamics are ignored by classifying them as something else than what they are. Differences in culture or language even make no difference, because the objects being referred to have meaning that transcends language barriers.
Another problem is that knowledge is often thought to be primarily semantics or even ontology based. Both of these cannot be true for many reasons. In the first case (semantics):
There already exists knowledge structure and dynamics for objects we cannot or will not yet know.
The same is true for objects to which meaning has not yet been assigned, such as ideas, connections and perspectives that we’re not yet aware of or have forgotten. Their meaning is never clear until we’ve become aware of or remember them.
In the second case (ontology): collations that are fed ontological framing are necessarily bound to memory, initial conditions of some kind and/or association in terms of space, time, order, context, relation,… We build whole catalogues, dictionaries and theories about them: Triads, diads, quints, ontology charts, neural networks, semiotics and even the current research in linguistics are examples.
Even if an ontology or set of them attempts to represent intrinsic meaning, it can only do so in a descriptive ‘extrinsic’ way. An ontology, no matter how sophisticated, is incapable of generating the purpose of even its own inception, not to mention the purpose of the objects to which it corresponds.
The knowledge is not coming from the data itself, it is always coming from the observer of the data, even if that observer is an algorithm.
Therefore ontology-based semantic analysis can only produce the artefacts of knowledge, such as search results, association to other objects, ‘knowledge graphs’ like Cayley,…
Real knowledge precedes, transcends and includes our conceptions, cognitive processes, perception, communication, reasoning and is more than simply related to our capacity of acknowledgement.
In fact knowledge cannot even be completely systematised; it can only be interacted with using ever increasing precision.
[For those interested, my summary is found at: A Precise Definition of Knowledge – Knowledge Representation as a Means to Define the Meaning of Meaning Precisely: http://bit.ly/2pA8Y8Y
Does Knowledge Become More Accurate Over Time?
Change lies deeper in the knowledge substrate than time.
Knowledge is not necessarily coupled with time, but it can be influenced by it. It can be influenced by change of any kind: not only time.
Knowledge may exist in a moment and vanish. The incipient perspective(s) it contains may change. Or the perspective(s) that it comprises may resist change.
Also, knowledge changes with reality and vice versa.
Time requires events to influence this relationship between knowledge and reality.
Knowledge cannot be relied upon to be a more accurate expression of reality, whether time is involved or not, because the relationship between knowledge and reality is not necessarily dependent upon time, nor is there necessarily a coupling of the relationship between knowledge and reality. The relationships of ‘more’ and ‘accurate’ are also not necessarily coupled with time.
Example: Eratosthenes calculated the circumference of the Earth long before Copernicus published. The ‘common knowledge’ of the time (Copernicus knew about Eratosthenes, but the culture did not) was that the Earth was flat.
HUD Fly-by Test
Don’t take this as an actual knowledge representation; rather, simply a simulation of one. I’m working out the colour, transparent/translucent, camera movements, and other technical issues.
In any case you may find it interesting.
The real representations are coming soon.
Information Visualization Is Not Knowledge Representation
(Lynda.com – Overview of Data Visualization)
Information Visualization Is Not Knowledge Representation
This great video from Lynda.com shows how the processing language/interpreter is great for modeling information.
With such a multitude of interesting ways to model data, we find it hard to resist the temptation to call this knowledge, but it’s not!
All of the wonderful representations here still require us to interpret their meaning!
What if there were a way to present knowledge in which our own understanding is not required to interpret them? What if our understanding of what we have presented to us becomes part of the presentation itself, and in fact, influences what we take from that representation?
We obviously need knowledge representation that can provide their meaning on their own for only they can provide a true understanding of their inherent structure and dynamics.
You see real understanding is the personalization of knowledge into your own mind. If your mind cannot dialog with that knowledge, it’s not really yours and if your mind does all the work, it’s only information.
Heurist.me (Carey G. Butler) 