resolution

rosybrown

Although this is not completely accurate, let us, for the sake of argument, represent our entire experience of computing as one of shuffling stacks and stacks of pages, which we could otherwise call pieces of paper, in a representational sense, and, on those pages, clouds of images. This, of course, refers to that part of the computing experience which consists of doing things on a screen, which is not the whole of computing - in a technical sense, it is only the edge of computing - but which is, of course, nonetheless the most prominent part of the typical user's experience of computing, or, at any rate, one of the very prominent and primary parts of that.

I feel a need to also state, though I have been warned this is only my obsession and I ought to leave it alone, that I am trying to correct, in that environment of things on the screen, a muddled and difficult situation. And some would argue that we simply need or ought to accept the difficulties of that environment, in exchange for the benefits it offers, benefits which are self evidently substantial and even miraculous, that, what can we expect, because we are asking for so much. This argument seems to be that it is the amount we are asking for that makes a muddled situation inevitable, and proceeds from there to vaguely suggest the solution to the problems of confusion and limitation we encounter in that environment is to somehow ask for less, that is, to limit what we ask for in one or another way.

This idea, that the solution to problems we encounter in visual computing on screens is to in some way limit what we ask for, is, in a way, accurate and, in fact, fundamental, but, as it is being applied, I do not think it is correct. I would ask what it is we are being asked to limit, in terms of our expectations, and I would say we are being asked to limit the wrong things. We are being asked, in a wide variety of ways, to limit what we are asking for, generally, and, to put it in simple and abstract terms, I think there is in fact no real reason, on a technical level or a conceptual one, for us to do so. What we need to limit is what we ask for at one time, or, to put it another way, on one screen. This may seem like arbitrary reasoning, but I think it defines something very important.

It is my contrary assertion that the whole question of what computing is for is being answered, throughout the industry, in a way that is in one sense fundamentally incorrect. The entire computing product has been constructed around this fundamentally incorrect assessment, is what I say, which, of course, means I am rowing against the tide in about as complete a sense as you could ask for. I have attempted to argue my point in the industry, and been summarily dismissed, but, I believe I am correct, and I remain determined to continue.

A philosophical point which I find interesting is the idea that paradox is inherent in the nature of reality. Perhaps this helps to explain my position. What I am proposing is very much computing, even computing as we know it. At least in a sense it incorporates possibly every aspect of computing as we know it, and even does not add anything which is not a component of computing as we know it. But, no, in fact, one component of computing as we know it is removed, and that is the muddle and limitation which is, in a classic example of irony, inherent to the contemporary experience of this medium which is all about orderliness and limitless capability. The result of the change I propose, then, will be precisely the same phenomenon, computing, but also dramatically different. And where I say that one - and only one - aspect of the contemporary phenomenon is being removed, I can reiterate that nothing is being added. Of course, a change is being made, and the form it takes is the reversal of the relationship of two components.

How can I describe what these two components are? I will say that they are two ways of describing something, and that that something they describe is, let me put it this way, what computing is about. Of course the question follows, then, what is computing about? And I just said there are two ways of describing that, but what they describe is one thing. Another way to describe the change I am proposing is to say that I am proposing an alternative to the way the question "what is computing about" is being answered. As I described it, the whole product, computing, and the industry which brings it to us, is currently formulated around one way of describing that, and now I would like to describe it in a different way, and by that means dramatically altar the whole phenomenon and the whole industry. Again, though, I am not removing or even altering any of the components of what computing is, and therefore I am not removing or altering any of the components that defined the industry which brings us computing. All of that remains just as it is, except for that removing of the one component of what computing is, the muddled and limited quality of the product, which, by their words and actions, everyone seems to suggest are inherent to what computing is, excepting myself, who now stand here arguing that that is an erroneous notion, and then this reversal of two components, which are not even physical components, but are, rather, conceptual ones.

So I have stated that these two components are definitions of what computing is about, or I called them descriptions of that. And, strangely, as I ponder this, it seems I could assert that it is not possible to say that computing is about some one thing. The conclusion I arrive at comes to resemble mystical notions. It seems I must say that what computing is about cannot be stated, but, in fact, that is not quite correct. Rather it seems that what computing is about cannot be defined in only one way. It seems to require two definitions. I am not entirely sure where this line of reasoning leads, but let us see.

In order to define what computing is about we indeed can begin with a single statement: it is about information. It is again interesting, though, to discover that, having made that statement, the term information seems to require definition, but an accident of precedence has occurred, here, because it might in fact be necessary to first define the term "about". If computing is "about" information, or, to put it another way, if it is about something, what does that mean, that it is about that something? In fact, the meaning of the word information begins to emerge when we investigate this. Here I can introduce another term, manipulation. What computing does is, it manipulates something that we call, or can call, information.

And it follows from this, if you treat it as a formula of logic, that information is something. This again may sound completely meaningless, but I think we will find that it is in fact, first of all, an assertion which needs to be made, that it must be determined whether that is or is not so, and then that it is very significant. But what does it mean, that information is something? I propose that what it means is that information is things, and this is why we can say that we are manipulating it. To manipulate information is to manipulate, in a general sense, a phenomenon, and then, in a specific sense, particles of some sort that comprise that phenomenon.

What we are attempting to define, then, is this thing, a particle of information. Here is where we encounter those two definitions which are both both correct and necessary, but whose role in our practice of, information manipulation, and of thinking about that, ought, to my mind, to be reversed, in order to liberate ourselves from certain limitations which are, as I see it, holding back the progress of our industry, and of the experience of computing which we, by our industry, provide.

The one definition, then, around which I believe our current efforts so preponderantly revolve, states that information is ideas, or, to put it in a more abstract and more coherent way, it is meanings. Indeed, information is meanings, it is meaning, and its value lies in our exploration of the particles of meaning which comprise it.

We move those particles about, we stretch and squeeze them, we juxtapose and combine them in various ways, and we divide them into smaller and smaller parts, then recombine those parts, into larger parts - new things - and recombine and rejuxtapose those, and we observe the results that follow from  all these manipulation, and, ultimately, build elements of our existence ourselves by these means. That is what the information component of our existence is about, and that is what computing is about.

However, if we define information only as meaning, a level of abstraction remains which is, in the end, highly problematic. There is this other way of describing what information is, and what a particle of information is, that simply is essential to mastering the phenomenon. And here I must assert again that I am talking about, for my purposes of explication, an aspect of computing, visual computing on screens. It is the meaning of that phenomenon which I am primarily discussing. It is true, however, that what I am proposing has, I would hazard, profound implications as regards the "back end", that is data.

This last idea can, I think, shed light on the confusion which I say has arisen around the question of what computing is, and around the proper relationship between components of the phenomenon, which I say we must establish, in our thought, to resolve the muddle I say we are in.

It is, upon reflection, natural, if in a silly way, to think that the more fundamental nature of meaning is to be found in data. First we are thinking that meaning must be extracted from something, in other words that meaning is an extract, and this leads us to think that meaning is a form of abstraction. We are abstracting something into some kind of medium, say, words, and we think that process is meaning, or is meaningfulness. Then, because we are not completely satisfied with the result, we are looking for something even more fundamental, and this seems to offer itself in the form of an even greater abstraction of information, data. There we find something which is the essence of abstraction. The parts are defined with completely strict exactness and ultimate simplicity. In there must lie the actual meaning of meaning, its fundamental meaning.

And, as usual, in some sense this is true, and data is the essence of meaning, but, again as usual, it is not the sense we think it is. The sense in which it is true is this: data describes things. And the fact that data is completely abstract, and strictly defined, makes it possible to manipulate data, and that makes it possible for us to manipulate the descriptions of things contained in data, which in turn makes it possible for us to manipulate, in a sense, the things thus described. And it is this ability to manipulate things that ultimately defines meaning.

Now, I have been referencing a kind of manifestation of things, images, on a screen. I am, I think, building a case that that, and nothing else, is the primary essence of what computing is, at least with reference to itself, which sounds like yet a third nonsense assertion, and which, once again, is the opposite of nonsense. It may be possible to argue that there are assorted ways in which we experience data, the contents of data, and images on screens is a prominent one of the, but the actual question is: is experiencing the contents of data the primary essence of what information is, and the standard answer would be, I think, that it is not, and that it rather is a secondary essence of what information is, and that what information essentially is is the information itself, as stored in data.

Viewed this way, the conventional way, images on screens are not the primary meaning of information or of visual computing on screens, but rater are a mere tool we use to access that primary meaning. And this is, I am asserting, incorrect. It is the reverse of the truth. Images on screens are the primary meaning of visual computing on screens, and data, and the descriptions of things it contains, are a tool that is used to access those images.

The functional meaning of information as experienced through visual computing on screens, then, includes three components, images on screens, which could be called a concrete phenomenon, and the data used to generate those image, which is also a concrete phenomenon, and then the descriptions of things that data contains, which is an abstract phenomenon. And it is the incorrect tendency to think that abstraction is synonymous with meaning, and that that abstracted layer is thus the real meaning of computing, which confuses the definition of what computing is, and, by doing that, limits our access to what it actually is. In fact, the essence of what computing is is not one or another of these things, it is the combination of the three of them. But we are also encountering, now, another question, which is that of the meaning of meaning. We will find some kind of answer in something of a misnomer of my own, which I produced here. In fact, I have already defined it as an experience of the ability to manipulate things. However, I was referring to the ability to manipulate images on screens, and that must now be investigated further.

First, to avoid confusion, I would note that many of the images we see on screens are letters, and what they comprise, texts. These images are not, except in an esoteric sense, meaningful themselves - they do not, in themselves, represent anything - rather, they are themselves a form of data, and that data is used to describe things, and thus to give us access to the meaning of those things, in the way I've described that is done, that is, by allowing us in some sense to manipulate the things described.

Now, it is possible to depict things directly using images on the screen, but it must be kept in mind that the things thus depicted are not, in fact, concrete realities. Let us investigate that further.

First, it is possible to directly manipulate certain things that we could say have their own independent existence. And it is possible to directly perceive those kinds of things, and it is also possible to represent those kinds of things as images on screens. [However, in the later case, those images, which exist for our purpose of manipulating them, are no longer things independently.]

For example, we can generate images of something which has its own independent existence, let us say, a city or town, and then we can view those images, and we would say we are viewing the city. But if we wanted to directly interact with the city, we would need to go there. The imagery, in this example, allows us to manipulate the city by carrying it through space to our location. Of course we are not carrying the city itself, we are carrying images of it.

The other thing is, we view, on the screen, images of things that do not even have an independent form outside of the data. And it follows from both these cases that what we are manipulating, via images on the screen, is something that in a sense does not exist. In a very real sense, that is the whole point.

However, while this thing we are manipulating, this universe of descriptions of things, may not in reality exist, it is a thing, which is to say, it has a form, and here we are approaching, in fact, the concrete purpose of this essay, which is to describe a product, or the use to which a product would be put, which is to allow us to interact with a form, with some thing, that does not exist. And the problem we encounter in contemporary computing is this: we don't know what form that is.

It is true that we know what the forms of many things in this thing, the contents of data, are, but somehow what the whole form of it is is unknown to us. This is in fact how the misnaming of meaning causes problems. If we really think about what form is being given to the whole body of the contents of data, it is the form of data: memory banks, arrays, words. And I think we are discovering, or we would, if we payed attention to our results, that this is an extremely inefficient form to assign to a whole body or world of information. The structure of data is efficient for creating individual things that do not exist, but something else, in the way of a form, is needed to assemble those things into an experience. Identifying that other form is also what this essay is all about.

This form, of course, is completely before us. Its unseemly obviousness seems to be one of the reasons it has been relegated to a supporting role, instead of being treated, as it ought, as it must be, as one of the two fundamental structures for information, the other one being data. I am speaking of space, the structural geometry of the world.

We could say that, to find an efficient structure for representing worlds of information we must look both in the medium in which information is created, data, and outside of it, and outside of it is the world, what we may call ordinary reality, and the organizing structure of that is space.

In space, of course, there are many things, and we can begin to see its efficiency as an organizing structure by, say, going down the street, which is lined with houses, and eventually coming to another street, and going down that, and turning this way a number of times until we arrive at a place where there is a collection of shops, and finding the one among those that offers the particular kind of item we are seeking, and going up and down its aisles, looking at items on shelves, until we find that item. We then move to a location that serves a special function, and move certain objects about a little, and this moving about of certain items records our permission to now remove the item we located to a new location of our choosing. And then we navigate streets again, perhaps to several other locations, and finally returning to our home, and all of this we do with a complete sense of fluidity and an essentially complete sense of orientation that contrasts with the world of information in computing as we know it.

Even the existing implementations of space in computing somehow seem very clumsy and tentative by comparison. Why is that? I would argue it is not, as most would say, because of some inherent difficulty presented by the problem of applying the structure of space to the computing environment, that it is not because of that at all. I believe it is because space is being treated as a foreign structure in a world, the world of computing and computer development, that sees data and data structures as what is native to it, or, really, what is proper to it. The effect is twofold: on the one hand the systems that are needed to implement the structure of space in data are treated as foreign objects. In order to insulate the pristine environment of pure data from them, they are rapidly encased, when they intrude, in insulating cysts of several kinds. These capsules are supposed to help. For instance, they are designed to make the offending data construct look not like a data construct but like something that it is not, a "tool" of some sort, such as an enclosing box, or a button. being typical components of these kinds of things. And then, at the same time, space is only partly implemented. The assumption is it cannot be fully implemented, and the following assumption is it need not be fully implemented, that a partial implementation is "plenty". This returns us to the idea that we are being overly demanding, and cannot and ought not to, "for our own good", expect an unlimited experience.

And, thirdly, the question will be asked - it is usually not so much asked as answered - what are we supposed to do with "space", in computing. Of course, for certain special applications, it is useful or important. For kids we have games, for real grownups, design tools. If ordinary people really demand it, we'll give them a smidgen for design purposes, too, though they are probably not up to using it.

In fact, let us address that last idea. In order to introduce things into space constructed in data, we must perform certain data procedures. These are considered to be very inaccessible to ordinary people, thus the elaborate procedure of constructing those tools that are supposed to help us. The truth is, these procedures are simple and direct, while the tools, which are very difficult to construct, end up being very clumsy and limited. I say - and I will be energetically challenged to prove it, this is true - that if they were given to us very directly, we would have no trouble at all building spaces in data.

But I still need to answer the question of why we would want to do it, and why it is such an urgent necessity. And now perhaps I ought to describe an experience, of this, as I imagine it.

Chapter 2