An information theory (part 2)

 

This page is published under the terms of the licence summarized in the footnote.

All free-to-read materials on the Avancier web site are paid for out of income from Avancier’s training courses and methods licences.

If you find the web site helpful, please spread the word and link to avancier.website in whichever social media you use.

 

This paper is a supplement to part 1 – a collection of related and additional observations.

Contents

How do messages/signals differ from meanings/information?. 1

Signal quality and interpretation issues. 3

The importance of information for life. 4

Continuous and discrete signals. 6

Abstraction levels. 7

Conclusions and remarks. 8

Footnote 1: Other views of “information”. 8

Footnote 2: on the confusion between data and information. 9

 

How do messages/signals differ from meanings/information?

Here, information is meaning created or found by an actor in some matter or energy that acts as a signal.

There is information potential in any variety detectable in a structure of matter or flow of energy.

There is actual information when actors use some information potential to recognise or describe something, or direct some action.

 

There is information potential in the variable

There is actual information when

angle of the sun’s rays.

a human reads the time from the shadow on a sundial.

a sunflower perceives the position of the sun and turns to face it

nerve impulses (electrical charges).

an actor responds by removing its hand from a hot plate

bending of a bi-metal strip.

a thermostat responds by switching a heater on or off.

movements of a honey bee.

honey bees dance to communicate a location of pollen.

open or closed state of an office door.

actors share a vocabulary in which an open door means “you have permission to enter”.

lengths of dots and dashes (in sound, light, braille…)

actors use Morse code to communicate.

quantity in a number.

an actor says 20 in reply to a request for a fact (say, the speed of a bicycle in miles per hour).

 

The last four examples are social systems in which senders and receivers benefit from sharing the meanings in signals.

Signal senders encode logical meanings in physical forms; signal recipients decode logical meanings from physical forms.

Information is created in the act of encoding meanings in signals, and obtained by decoding meanings from signals.

 

A social system depends on its members sharing information.

System actors must communicate by sending and receiving messages or signals.

All the actors in this paper are message senders or receivers (in evolutionary terms, computer actors are proxy human actors).

 

The system can only thrive if actors find the same meanings or information encoded in the signals they send and receive.

Actors create and find logical information in a physical signal by mapping its elements to a general pattern or language.

 

With no actors to look at it, a message (be it print on paper or magnetic patterns on a disc) it is just an arrangement of matter and energy.

The meaning of the message lies in the intention of the sender and the interpretation(s) made by receiver(s).

 

Different actors may find different meanings in one message.

When and where do those two meanings exist? Not simply in the material form of the message.

No actor can understand the material form of a message without a language, or be entirely sure how it might be interpreted by another.

Meanings arise at moments when actors create and interpret a message.

They do this by mapping elements of the message to a language known to the actor.

 

These three messages may help to further illustrate the difference between signals and meanings.

·         “Read the symbols in this signal.” Since the signal quality is OK, and we share the same language, you’ll read the meaning I intended.

·         “Read thu simbuls in this signal.” The signal is different (the quality is imperfect), yet the meaning you read is surely the same.

·         “A rose is a rose is a rose.” The signal is intended to be meaningless, yet you may nevertheless find some meaning of your own in it.

In other examples, ambiguities arise from the use of ambiguous words, or different dialects of a language.

 

In short, information is meanings created and found in messages or signals, by actors using a domain-specific language.

Social systems depend on messages – usually - conveying the same meanings to message creators and users.

 

What do meanings mean? Often, they are directions or descriptions.

Directions may be continuous or divided into discrete instructions.

Descriptions are usually divided into facts (tasty, tall, scary) about things (say, food, friends and enemies) that actors perceive as discretely identifiable.

Signal quality and interpretation issues

The physical content of a signal can be written and read with different meanings – accidentally or deliberately.

To repeat the example above:

A general tries to say “Send reinforcements; we’re going to advance” using the physical medium of sound waves.

A corporal hears the same physical signal as “Send three and four pence; we’re going to a dance.”

Thus, the signal receiver obtains information, but not the information the signal sender intended to convey.

 

Humans and computer actos can use Morse code to code/encode meanings in a physical signal.

Suppose senders and receivers use different versions of Morse code, then they will write/read different character strings.

At the first level of decoding, the signal receiver can read the message as a character string – with no apparent difficulty.

But at the next higher level of reading, the receiver will find it the characters do not form into words.

 

A poet may say their poem means (signifies) whatever meaning each reader finds in the poem.

Some systems thinkers take that idea to an extreme.

They say that meaning is only found after a signal event – in the use(s) made of the signal.

 

This way of thinking about communication may lead you to Luhmann's "autopoietic social system."

There are reasons to doubt the value of Luhmann's theory.

First, it radically changes the meanings of “autopoiesis”, “society” and “system” understood by others.

More importantly, society depends on listener/readers (usually) successfully construing what speaker/writers mean.

This means signal senders and receivers give the same meaning to the contents of signals.

This implies the actors must share a vocabulary and grammatical rules – be they innate or learnt.

The importance of information for life

The physical forms in which signals can be stored and conveyed, using matter and energy, is not the focus here.

A dash of physics or chemistry might help you understand how information can be stored or conveyed.

But information is better understood in terms of its role in biological evolution and social systems.

 

It has been said that “all life is information processing”; certainly, information is needed for life.

Even primitive actors need information about the state of the world they live in.

They need to monitor and direct the state and behaviour of themselves and their environment.

An actor’s very survival depends on it perceiving and modelling those realities well enough.

It needs sensors to detect the state of itself and its environment.

Those states must be represented in energy flows sent to body parts or organs.

 

The importance of information for social and business systems

The ability of actors to describe the world evolved alongside biological evolution.

Evolution led to actors with brains that can remember past facts about the world (e.g. where a food item was buried).

And then to social systems in which animate actors communicate.

“Social” implies actors in the society cooperate, which in turn implies they communicate, share information using a common language.

“System” implies actors in the system follow given rules when responding to information received.

 

Social systems, information and communication existed before human actors,

However, the range of communicable information vastly increased with the ability to describe things in a verbal language.

Humans use a wide variety of signals to share information that describes things, which they need to monitor or direct.

They create information by organising matter. E.g. by writing a letter, or painting a picture, or placing a chess piece.

The retrieve information from organised matter via energy flows. E.g. the light reflected from a painting or chess board.

 

Social systems range from the informal to the formal.

In evolutionary terms, business systems evolved from social systems.

Humans formalised social communication to standardise the transactions of government and commerce.

In short, information existed eons before human and computer activity systems.

How do social system actors share information?

Social system members have three mechanisms for sharing the language, rules and state of a social system.

·         Breeding - by inheritance from parent to child (languages and rules only,  not state information)

·         Communication - by messages sent from senders to receivers

·         Recording - by storing in a shared memory that all actors can access.

 

(By the way, actors and societies may evolve as a result of copying errors in breeding, communication and recording.)

How do social system actors store information?

In evolutionary terms, all information storage must have started off as fuzzy and imprecise.

Biological memory mechanisms are fragile, prone to error, decay and loss.

Humans evolved the ability to communicate logical information using physical sound waves.

Then to store information (including the state, languages and rules of a social system) in persistent storage materials.

 

In 2,500 BC Sumerians stored their literature, laws (behavioural rules) and business transactions (information) in libraries of clay tablets.

(Sumerians had shifted from picture-writing into cuneiform, which means "wedge writing" in Latin.

Cuneiform was written with a wedge-shaped stylus, similar to ones used on today's hand-held computers.

Writing was inscribed into damp clay tablets, which were then baked until hard.)

 

A social system’s state also includes the current state of inter-actor communications.

A request-reply communication requires that signal creators and users share a language.

But moreover, the requester must remember the question to which an answer is sought, and associate any reply with that question.

So, social communication requires that a requester’s brain remembers the state of an inter-actor request-reply communication.

And in this way, the current state of inter-actor communications is distributed between the minds of social system members.

Does communication imply intention and mindfulness?

Information can be extracted from a signal, even though the sender had no intent or mind to send a message.

When you read the time from a sundial’s shadow, you don't attribute any intent or mindfulness to the sun.

(Though the creator of the sundial was mindfully intent on building it to block the sun’s rays.)

 

However, our focus is on social/business systems in which messages have both senders and receivers

Some receivers may not successfully obtain the meaning intended by the sender.

But a social system's very survival depends on enough receivers sharing senders’ intended meanings.

 

Words like "intent" and "mind" are questionable in a general information theory.

You might say a human, elephant or dog "is minded" or "intends" to send a message, within or across species.

You might feel uncomfortable about saying the same of a bee or an ant.

You probably see honey bees as automata without minds or intentions to do things.

 

It seems our concept of mindfulness is tied up with the self-awareness we see in at least some animal species.

And it seems reasonable to speak of "intent" and "mind" where social system members are self-aware.

 

By the way, some regard as jellyfish as social system all to itself.

(At a brief glance, "The Social Metabolism: A Socio-ecological Theory of Historical Change" by Manuel González de Molina Navarro, Víctor M. Toledo looks relevant here.)

Continuous and discrete signals

The term analogue is used to describe signals or data that vary continuously, as the hands on a clock face do.

The term digital used to describe signals that are chunked into discrete units, as you read from a clock that displays discrete numbers. 

It is now common practice to convert continuous analogue signals into discrete digital signals, and vice versa.

The discreteness of facts in social communication

The universe may be continuous, but animals do decode apparently continuous signals into discrete facts.

Honey bees watch a dance and decode it as showing the distance and direction of a pollen source.

You see a rainbow as distinct bands of colour; you hear a continuous stream of sound as composed of discrete words.

You perceive a stream of light as revealing your lunch to be collection of discrete food items sitting on a discrete plate.

Discriminating between these things is useful; it helps you to spear discrete food items and eat them.

 

“At a conscious level, humans tend to interpret the world as discrete things.

We pattern match, then label (not necessarily verbal), then move our limited attention on, which is part of a survival strategy, to categorise things as safe or dangerous.

'Chunking' may better describe the operation (https://en.wikipedia.org/wiki/Chunking_%28psychology%29 ).

Most descriptions of chunking don't do justice to the profound way in which it avoids information overload.” Ron Segal

 

How our brains and computers work at the neuron or electronic level is not important here.

At the level of consciousness and social communication we chunk our perceptions of the universe into discrete entities

Perceiving and describing discrete entities (you, me, your lunch, or a bicycle ride) is how we make sense of the world.

And we describe those entities in terms of discrete facts (you are tall, I am old, lunch is tasty, and this bicycle is speeding at 20 mph).

These descriptions are models or coded representations of the states of things we perceive as discrete entities.

 

If a signal can be perceived as composed of elements, then each element may be used to convey unitary fact (e.g. surname).

You can create or find composite facts in a structure of elements by matching it to the rules of a grammatical structure.

E.g. full name = forename + list of middle names + surname.

The digitisation of facts

The entry to the information age was associated with the digital revolution that happened as business communications were computerised. 

Most communications are now sent and received by computers in a binary digital form.

This means that information is encoded in signals/data flows using the binary number system as series of discrete digits 0 and 1.

The binary digits may be represented in physical signals as the presence or absence of physical things.

Or as the contrasting values of a physical quantity such as voltage or magnetic polarization.

 

How to encode human-level verbally-expressed information in binary digits, and decode information from binary digits?

This is usually done step-by-step, in a series of transformations, down and up a communication stack.

Abstraction levels

The table below shows three levels of abstraction away from the behaviour of entities in the material world.

At each level, actors create and use data that conveys information from or to actors at the level below.

 

At this level of abstraction

these actors

work to

and so convey information about 

3 meta system

methodologists

define notations for modelling data and process structures

The language systems designers use at the level below

2 system description

system designers

define data types and process flows that using the notations above

The language systems actors use at the level below

1 operational (run-time) system

system actors

read and write messages containing data items conformant to data types above

The attributes and behaviour of real entities at the level below

0 material (run-time) world

external entities

do stuff that is describable in messages above

 

Conclusions and remarks

This paper says

·         Information is a meaning created or found by an actor in some matter or energy that acts as a signal.

·         Data is a signal or signal element that can be created or perceived by an actor, and mapped to a general language or ontology.

 

There is more at http://avancier.website on:

·         the evolution of description and communication, read A description theory

·         the formalisation of types used to describe things, read A type theory

·         knowledge, information and data as a hierarchical structure, read Data, Information and Knowledge.

·         other information theories.

Footnote 1: Other views of “information”

This work may be seen as philosophy based on a scientific view of the world; especially on biological evolution and the science of digital information systems.

For us, information appears in the act of informing – in the communication of knowledge from senders and to receivers.

The elements of information are descriptive or directive qualities (including quantities).

 

Scientists have developed a broader view of information.

Information Philosophy (I-Phi) <http://www.informationphilosopher.com> is a philosophical method grounded in science.

Its definition of information is any quantity that can be understood mathematically and physically.

That includes information in the structure of any physical object, like a snow crystal, a star, a biological entity, genetic code, cell structure etc.

Any physical structure is thus an "information structure," from which information can be abstracted as meaningful knowledge.

 

However, a message that is not actually meaningful to any sending or receiving actor is not a useful concept in this work.

 

Social, business and software systems all depend on communication.

Our focus is on information created and found in messages sent or stored with an intended purpose.

Business systems integration (a focus of enterprise architects) depends on information flowing in messages and stored in memories.

In today’s “information age”, businesses look to capitalise on information they glean from messages that have been stored in huge quantities.

 

Our information theory takes a "survival of the fittest" view of how information processing evolved in social systems.

It views business systems as formalised social systems, in which actors communicate to cooperate in activities.

 

Claude Shannon developed “information theory” about the limits on signal processing operations such as compression, storage and communication.

Shannon wrote "The fundamental problem of communication is that of reproducing at one point either exactly or approximately a message at another point."

His concern was maintaining signal quality in communication; his theory is about the physical content of a message.

Shannon wrote: "Frequently the messages have meaning"

But as long ago as 1956, Boulding observed that Shannon did not address the meaning of communication in social or business systems.

 

For us, the fundamental problem of communication is that of knowing what a message means, at a logical level.

A system can only thrive if actors who receive messages/signals find the same meanings/information that senders encoded in those messages/signals.

And this can only happen if senders and receivers both map elements in messages/signals to elements in a shared language.

To create and find logical information in messages and memories, actors must map discrete elements to elements types in a shared language.

Footnote 2: on the confusion between data and information

This morning (28/10/2015), I received the email below from a recruitment consultant.

Enterprise Information Architect: London - £90,000 per annum plus benefits.
My client, a large and iconic London brand is now looking for an innovative and experienced Enterprise Information Architect.

He/she will focus on developing and leading the exploitation of data and information architecture in conjunction with business area leads, business partners and project delivery teams.

 

How does the recruiter expect us to differentiate information from data?

Consider a particular telephone number, or customer address. Is that a data item or an information item?

Consider a message containing customer name, address and telephone number. Is that a data message or an information message?

Does it make any difference who creates the item or message - a human or a computer?

 

Consider the generalised structures for items and messages – simple data types and complex data types or data structures.

Consider the structure for a telephone number, or an invoice. Is that a data type or an information type?

Does it make any difference whether the type is defined by an international standards body or by a programmer?

 

Consider the higher level meanings of messages that are created in message by senders and obtained from by receivers.

Does it make any difference who creates the item or message - a human or a computer?

Resolving the confusion between data and information

Business system design usually addresses the plural (sets of things) rather than the singular (individual thing).

Business processes follow the logical paths defined in general process types.

Processes monitor and direct business entities and events that instantiate defined types.

Messages convey meaningful information about business entities and events by giving values to their typical qualities.

 

A business needs senders to encode meanings in messages, and receivers decode the same meanings from those messages.

To this end, the mapping of data types to generalised meanings must be agreed and defined before messages are sent.

Thus, business systems formalise what is fuzzy and ambiguous in purely social communication.

Social system members may share their own language for describing the entities and events of interest in their domain.
The domain-specific language used in a digitised business system is commonly defined in what is known as a ontology

A data model is an ontology composed of generalised data types (e.g. Customer, City) instantiated as data items (e.g. Jack Jones, London) in particular messages and records.

This language stands apart from, is independent of, particular actors, particular signals, and particular information.

 

The proposal here is: Data is a signal or signal element that can be created or perceived by an actor, and mapped to a general language or ontology.

Honey bees use the language of dance to signal the locations of pollen sources; situation-specific data is found in particular dance movements.

The bees’ language is generalised; it is independent of particular bees, particular dances and particular pollen sources.

Accepting the confusion between data and information

So, why are the terms “data” (signal) and “information” (meaning) so confused in enterprise architecture?

Because systems are built on the assumption that physical data carries the same logical information to all senders and receivers.

It is assumed that the meaning actor A gives to data on posting it will be read with the same meaning by actor B, in a different time and place.

And that data stores record the current state of business entities (and/or the history of business events) in a shared memory all understand the same way.

 

In practice, the term data and information are used interchangeably and used so loosely.

You generally can’t assume either term has any one specific meaning.

Some equate data with information; some equate data with signals.

(A sister paper called Data, Information and Knowledge addresses the wide variety of ways people distinguish the terms.)

 

 

 

Footnote: Creative Commons Attribution-No Derivative Works Licence 2.0      3/24/2017

Attribution: You may copy, distribute and display this copyrighted work only if you clearly credit “Avancier Limited: http://avancier.website” before the start and include this footnote at the end.

No Derivative Works: You may copy, distribute, display only complete and verbatim copies of this page, not derivative works based upon it.

For more information about the licence, see http://creativecommons.org