Whether you're a current or potential Literary-Machine user, be sure to check
out these screen shots and writings . . .
from Gunnar's desk . . . Discreteness — Symbols
What Words Do
The Concept Level
The Symbol Factory
The 2nd Abstracting Step
The Dropping Mechanism
History versus Relations
Variability & Continuity
How to Use it
Not Grammar and Operators
Finding the Next Thought
Do we really Communicate?
Cognition, Causation, Time
Everything in the Literary Machine is about words. These words comprise your Dictionary, which is a file. The
words you add to it are numbers in this file. So, you can change a word's spelling or ending without affecting
its links to other words or its behavior in the system.
The Literary Machine's words, are discrete phenomena. That is, either they exist in the current database, or they do not; either they have a particular connection, or they do not. This discreteness is characteristic of thinking. True, our experience of the world is varied and rich in nuances and degrees, but this experience is the effect of combinations of discrete phenomena, many of which occur together by chance. In the next section, I show how to use combinations of the word-symbols in your Dictionary to generate various shades of nuance and degree.
In LM words vary as they do in language. For example, dictionaries is variation of the word dictionary.
To represent such variation, a word appears in any number of forms called inflections.
The Inflection Object keeps track of a list of inflections for every Dictionary word. Unlike linguistic inflections, however, LM inflections can be similes or synonyms — any word that you consider interchangeable with that Dictionary entry. (The synonym set of that entry.)
The Literary Machine is a symbol factory. In it you can combine Dictionary-word entries into database objects
called concepts. Thus you build new symbols (and cognition values) hybridized from existing ones.
You may think up your own ways to use this middle layer of definition. The symbol factory is as flexible as your imagination: it does not determine what you make out of it. (Read more)
I often use LM Concepts to define combined or extended ideas, as in the concept of "love" + "relation". I could also use concepts to harbor near-synonyms in a group, thus making the expression more precise.
The LM concept is a step in a symbol reference chain that accommodates our preference for slack in meaning and economy in symbolization.
Dropping a Concept on the Desktop to Produce the Related Note-Card Items
Selecting Representative Words in Items to File Them Under
The Literary Machine handles not only its own symbol factory, but also information in files outside the database.
In fact, the seamless integration of internal and external data is one of the most powerful features of LM. It
is extremely flexible in allowing you to create and alter interrelationships between the two.
Via the standard Windows clipboard, you can import and process any text from outside sources. The Machine can accumulate and rearrange text items into longer concatenated texts as projects, and it can send these projects to the clipboard so you can apply them to other documents or text boxes.
You can also arrange projects into a traditional outline tree. This feature allows you to, say, form a book with chapters and form a hierarchical outline of each chapter's text.
Ideally, to meld with the system of human perception, The Literary Machine should treat many media of expression, including sounds and moving pictures. An ambitious ideal! As a modest beginning, it already lets you store sounds or pictures with your text on the cards.
The LM system stores textual information as items on Item cards. These note cards appear in Item windows on
the LM desktop. Item texts are not automatically connected to objects in the Literary Machine database system;
instead, they are indirectly classified by free association with keywords.
For all practical purposes, keywords are concepts, which are built by Dictionary words — singly or in combination. Since a concept may thus consist of several Dictionary words (without a designated header word), LM marks-up an item with an arbitrary word (a keyword) from the concept. At a first, this strategy might appear senseless, but your work in the program gradually shows you how useful it is.
Project Window's Tree Pane
At the heart of Machine operations is the "drop" mechanism.
For example, drop a Dictionary word or a project on the desktop to make all its related items appear. Or, drop
a flyword on the Dictionary to search for, and scroll to, that word.
The drop mechanism is a mouse-driven technique that uses the computer screen area as efficiently as possible. It allows you to use your spatial or graphical cognition for grouping and connecting things. Nevertheless, it bears no resemblance to the conventional mind-chart paradigm. In fact, LM's departure from traditional mind-mapping and concept-mapping methods is by design. While graphics are instantly recognizable as icons and do aid memorization, I am skeptical of them as a stimulus to creativity and as an environment for human thought-pattern programming.
What we've explored so far benefits the analytical side of human mind, the part that asks, What if . . .
? or Could not this or that be changed?
Yet we also have other needs. We need to describe developments in their temporal perspective, and when we do, we must avoid too much analytical "what-if-ing."
The same Dictionary word can live in both environments. Take the word politics. We may explore politics as the common understanding that underlies the arena of possible social change. Or we may explore politics as a factor in human life that competes with other interests. Either way, the premise could be "anything goes."
The dialectical opposite is exploring politics as an easily recognizable entity having an organic development in time.
In the dialectical setup of The Literary Machine, your text narrative is the equivalent of the second perspective. Your Project window might contain headers of these continuing texts, also called "streams".
(To underline the dualism , you could in fact name projects as Dictionary words. Drop a Dictionary word in the project box to create a project of the same name.)
In LM you use the item texts as fragments to form the project's contents. You can change the typographical setting to Book Mode for viewing them all, one-after-another, as though together they comprise a book.
Project in Book Mode
You can use the Literary Machine Symbol Factory to develop your own private system, using the words around you. The more you work with it, the more you increase efficiency and your refinement in communication.
Cognitive Science has no simple answers so far, but it is evident to everybody that the meaning of words depends both on context and on ourselves as individuals. You may either wonder about how we can differ (assuming that cognition runs in common programs), or the other way around. Building your own dictionary (lexicon) with special links is a good beginning in thinking about this.
I think our brain may have developed its high priority for abstraction for the sake of processing economy. It must deliver results on time. This beneficial effect in impreciseness, or slack, has guided me in making The Literary Machine. The feeling of explanation pending should mirror reality.
When you look at a rune stone, what do you see? You see a stone, a delimitation of reality. The stone is a solid block of context, really. You see one or two names, perhaps also a geographical name.
Nouns. The beginning and the essence of language as operational knowledge.
Sure, grammar and verbs and the rest are good, but not really essential for the Viking or the Literary Machine user. Nonetheless, once a basic node model consisting of such substantives ("names" and "predicates" in fact) has been laid, most mind-creativity software heavily elaborates on operators and attributes. See for instance the theory around topic maps!
This inclination responds to our noblest feelings for grammar, but it also invariably leads to rote reliance on a few trick formulas that expand the "box" but little and to a maelstrom of messy theory detail. It is unclear who should use it, when to use it, and what to use it for.
Researchers in the social sciences are usually unaware of the good old saying that a theory should be simple. That is, efficient.
Grammar and linguistic perfection are impressive. In creative-thinking tools, however, honing them is to aim wide of the mark. It is like honing our impressive visual perception abilities or our even more impressive ability to pedal a bike.
(2005: On second reading I must add that I consider Natural Language work in AI to be highly productive and relevant. LM can be seen as referencing certain
semantics and ontology issues.)
Thinking can be wandering freely, or it can be a journey to a destination. In other words, it can be a course of
exploration and discovery or of road-mapping. In destined thinking, our last thought and our next thought follow the program.
Free thinking, creative thinking, is all about finding the next thought (or impression). Here we are at the busiest of all philosophical crossroads!
LM's unique contribution to the thinking-tools industry is its enhancement of the free-thinking process. We are not bookworms; we are more like computers with a limited ability for parallel processing. A freethinking mind looking for the next interesting thought is not like a marshaling yard. Yet, the much-heralded mind-tree graphical approach boxes it into one. So mind-charting and outlining tools are of limited value in thinking exploratively . . .
. . . but are very useful in writing a piece like this.
Screen Layout Command: Tile Items
Most thinking-tool development rests on a more or less outspoken assumption that thinking and communication
are two aspects of the same thing. This mistake is catastrophic.
(Philosophers like Descartes, Kant and Moore have struggled with this issue for centuries, but they often relax the next second to tell us how everything works.)
Yes, we use language both for thinking to ourselves and for speaking to others, but that is no guarantee that we get the same message from our language that somebody else does. Innately aware of language's true nature, we formulate it differently for export than for private contemplation.
The Literary Machine concentrates on this private cognitive work and should support memory, concept-building and association abilities. Communication is a secondary task. So far Literary Machine users have shown no interest in using LM for communication. (As a matter of fact, I did have other expectations.)
Cognition, causation, time
Human cognition spans reality with many overlapping and partially contradictory ways of understanding. At best it is a patchwork of compromises with many white areas on the map.
As intelligent humans we look for causal laws. We sometimes find them and believe that they work, at least locally in their particular cognitive and scientific environment. Equally less than perfect, we often contend ourselves with rules and forecasts based on statistical evidence.
Always trapped in our limited cognitive abilities, we strive to understand. The Literary Machine works with your mind, as it is, not against it.
The dictionary contains the words (ideas) that you in fact work with. The program contains many associationist features, like delivering your last visited thought again based on previous association or its being the most recent one.
Time is the strong, moulding and communicative dimension of a cognition representation system. LM considers time (alias time-lines, developments) to be a special kind of understanding. Time gives definite outcomes. Ideas about possibilities and causal effects are somewhere else in our mental awareness.
Book writing is often timeline driven: "This is what happened to it." The Literary Machine creates a balance between development accounts and the open system-like view.
Among other things, The Literary Machine is a production side hypertext tool.
The Literary Machine project rests on some very basic principles:
Test The Literary Machine. You might get the idea for some other great experiment!
Screen Layout Command: Sort Items Diagonally
these other writings in English by the program author, Gunnnar Sommestad: