The Librarian’s Dilemma
Every document is different. Every document writer has his or her own viewpoint and purpose. How do you match documents up with readers or students?
Almost all library systems are based on a combination of document type and subject matter.
Maps have to be stored differently than magazines which have to be stored differently than large reference books that don’t circulate which are handled differently than audio recordings, etc. etc. But on the computer everything is just a digital file. So the choice seems obvious: Organize by subject.
Well, that might work for non-fiction, but what about novels, poetry, artwork? These are usually organized by author/artist and genre. OK for most people. Unless you’d like to read a novel about time travel set in Boston in the late 1800’s. Unless you happen to know that Edward Bellamy wrote such a book, you’d need a good cross-reference system to find that one.
And there are similar problems with non-fiction. You will find books that cover the theory of a technology along with instructions on how to use it and sample projects. Other books may cover only theory while others may have only sample projects. Do you put them all together under the same subject?
The above term is meaningless to most people, but I love its mystic ambiguity. Graph theory is a mathematics that seeks to describe the connections between the various elements of a system. Most subject systems and file systems, more or less by necessity, use a tree structure to relate their elements. In a tree structure, there is only one unique path to each element from the root. This works for trees, but in a real tree the end node of every single path is a leaf. In a system of data, you don’t have that kind of uniformity between nodes.
I know of no basic computational system that has overcome this. The closest I know about are computer networks. And the logic involved to make these things work is quite intricate. Yet it is intuitively obvious that in the real world there are an endless number of possible paths between point A and point B. The choice of which path to take is a very human activity. It involves intuition, guessing, luck, hard choices, possibly even physical violence.
A system of data may not be that dramatic, but the point is that without human choice, it’s just a big pile of undifferentiated data.
LRH says in Logic 10 “The value of a datum is established by the amount of alignment (relationship) it imparts to other data.”
So I looked at various LRH basics to see which ones might be useful in helping me align the kinds of data I was working with.
I knew I wanted to use the Dynamics. These can be used to assign scope to a document. They also provide an outline for a list of subjects. I would try to use both characteristics of this scale in my scheme.
The cycle of action in its many forms also seemed like a useful concept. This gives a general context to a document. Is it about beingness, or doingness, or havingness? But I also had to factor in the concept of abstractness.
Using the analogy of a map, it is difficult for us to locate something in more than 2 dimensions at a time. And while there are ways to communicate a third dimension, they don’t work will if the upper layers mask the lower ones. Therefore, it is relatively easy to show mountains on maps, but relatively difficult to show caverns.
On top of this, the human-computer interface remains two dimensional. And while you can simulate the exploration of a 3d space on a computer, the software is complex and it is difficult to indicate how exactly to get to the desired destination.
Thus, 2 dimensions, with only an indication of a third, was as complex as my basic navigation system could be.
If you use certain kinds of software very much, you will run into the concept of mapping. In abstract terms, this involves taking one structure, such as a table, and creating a correspondence between every point (or node) in it to a point on a different structure, say a tree. The term, I suppose, comes from map making. The problem there is usually to represent a curved surface (the earth) on a flat sheet of paper. In this case, the mapping can be accomplished through a rather straightforward deformation function. This is topology. I haven’t studied it.
But what I needed to do was map a table, or grid, to folders in a file structure. If your grid is 5 by 9 (as mine currently is) this means you have 45 nodes in a lattice configuration to 45 nodes in a tree configuration. These are the 45 top-level nodes in my data structure, so I wanted them all to be top-level branches in my tree. Since I needed both structures to be human-friendly as much as possible, I wanted folder names that would preserve the basic order of the system when they were arranged alphabetically.
LRH has a saying that the physical universe always works 180 degrees different from your own universe. In your own universe, if you want something you get it and if you create something it’s yours. In the physical universe if you want something you can’t have it and if you create something you have to sell it in order to survive.
And so it is with human subjects. Theoretical physics would seem to be a specialized subject buried deep below many layers of “easier” subjects, yet it is the subject that comes closest to describing the most basic characteristics of all reality. In general, subjects are studied backwards, starting with the effects that are most easily noticeable and working back towards suspected causes. This is 180 degrees from how things are created, and I wanted to follow a creation-based paradigm for all my data. That’s what I am used to in engineering, and I didn’t want to change the paradigm in other fields just because they are not yet as easy to apply or understand as engineering is.
My first shot at a system
The dynamics start with 1, self, the effect, and go up to 8, god, the cause. I wanted to reflect this in my system, but also retain a distinction between the “old” gods and the “new” ones. So I modified the dynamics into a list of effects: universes, stars, planets, mockups or societies, machines, life forms, humans. Then added to these freedom and a new civilization.
Then I came up with five aspects for these effects (I could have had more): beingness, theory, mathematics, parts and systems.
I based this on what I experience as the sequence of making something. One starts by deciding to be the maker, one then gets an idea of what one wishes to make, predicts its operation (by mathematical simulation), fabricates its parts and then assembles them into a working system.
Mapping these to a set of alphabetical words expressing similar concepts I came up with:
be, idea, math, part, system for my set of first key words, and
anew, freedom, human, lifeform, machine, mockup/society, planet, star, universe for my set of second key words.
Deciding what subjects or items should fit into each of these 45 top-level combinations has been a real challenge. And it is actually a kind of philosophical exercise. Was psychology used before humans ever existed? I say “yes.” So that puts it under ideas-social, not ideas-human. Interestingly, art (taking my cue from The Factors) begins as a subject up at the universe creator level. That doesn’t mean that I will put all examples of human art in that category, though! It is assumed that concepts that show up “high” in the data structure will apply to all levels beneath them. Thus there is art in stars, planets, societies, machines, life forms, and even humans. I consider that the fact that this is recognized is more important than the confusion caused by having to decide where to put a particular file that has to do with art.
After working with this system for a few days, and making a printed table to help me keep track while I move my files from a more traditional structure to this new one, I am getting more comfortable with it. And initial experience indicates it will result in a more balanced distribution of files among the categories with better integration of subject matter. These are important considerations to making the whole system workable.