Paper documents are heavy and they take up a lot of space. When we have converted paper documents to electronic media, they weigh nothing and take up very little space. One of the most significant themes of the so-called 'Electronic Revolution' of the past fifteen years has been the push to covert these paper documents to electronic storage.
Most of our modem paper is very acidic. This paper deteriorates rapidly. Almost all our newspapers, magazines and books have been printed on paper which is turning brittle, yellowing, and crumbling to dust with every passing day. Within fifty years, over half the material printed during the Twentieth Century will be reduced to motes of dust.
The documents left from earlier centuries are more stable, and may survive many centuries to come, but with the industrialization of the l9th Century, and the ever-increasing demands for more reading material, paper became a mass-produced product, created for the passing needs of a day.
The need to convert these cheaply produced documents was one of the more compelling drives behind the earlier waves of this 'Electronic Revolution'. It is still a critical point today, but the 'Revolution' has been stymied by the magnitude of the challenge. We came up against several serious bottlenecks.
The mass of data to be converted is one very serious hurdle. The historic documents of this century would, if put together into a single archives take up the space of several of our western states - Colorado, Arizona and Utah combined. But of course, centralizing these documents together in such a way is an impossibility, so no centralized or comprehensive scanning effort is possible.
Such programs to scan documents have arisen from individual archives, libraries, corporations and universities, each converting its own documents for its own purpose. This effort has been underway since the early 1980's. More recent developments such as the Internet are making it possible to synthesize these specialized electronic archives into larger and more generalized data systems.
In this way we have overcome one of the hurdles before us, the fact that printed information is dispersed throughout the world, with no single guiding principle of organization. By approaching this mass of data through many small, piecemeal efforts, we can coordinate these individual databases through shared media, and create a structure of data which is finally much larger than any single group could have built through a centralized method, while at the same time avoiding all the costly organizational and management issues attending more ambitious efforts.
But we are left today with another, perhaps even greater hurdle to converting our historical records to electronic memory. In order to scan these documents into an electronic format, we must first remove each document from the file or the box which houses it. We must then scan the document, each one at a time, then return the document to its file or drawer, ensuring that we have not misfiled it. If the document is stapled, taped or glued to another, we must separate the sheets, scan each document separately, then file all documents into an envelope, or re-staple them, so the integrity of the original is not lost. If the document is bound in a binder, or a book, we must remove it from its binding, scan, then rebind it.
These steps are extremely labor-intensive and time-consuming. it is as if the entire 'Revolution' has got caught in the slow lane behind a heavy, slow-moving and overloaded convoy. We could go faster if we didn't have to spend thousands of hours removing documents from files and putting them back into the same files.
A single records box storing 1 cubic foot of documents can hold as many as 1,000 photographs or 1,500 individual sheets of paper. Granting only ten minutes total scanning time for each document (which, with the transportation, temporary housing, and maintenance required by each document, is a generally conservative estimate) we can spend up to 15,000 minutes, or roughly 250 hours scanning one cubic foot of paper. Given that the federal government of the United States alone generates at least this much paper with every passing moment, it is clear that our efforts to scan paper documents to electronic media are falling forever behind.
The time required to scan documents represents the second bottleneck we've faced in our efforts to convert the data of history. There's so much of data, it takes years of painstaking labor to scan even the smallest portion, and the effort to do this is very expensive.
In order to overcome this second hurdle, we must look outside the fields of document scanning, searching for techniques that can produce mass results. This can be done by mechanizing the detail work, moving it from the techniques of individual handcrafting (which is basically what most scanning is today) to an assembly-line approach. Given the infinite idiosyncracies of paper documents, is such a possibility reasonable?
Medical imaging provides us an answer. Medical scanning devices have been brought to a pitch of discrimination that makes it possible for us to scan human bodies down to microscopic levels, without having to first take the human out of the body. If we can do that, then we can certainly develop techniques that Vill permit us to scan file drawers without removing the documents from the files. This we will accomplish through a marriage of document scanning and medical scanning technologies.
Instead of spending thousands of hours removing documents from and returning them to files, we will someday have developed the means to scan documents in situ. When this happens, the ongoing process of the quantification of historical data will go through another period of radical acceleration. Where it now takes years of effort to scan the contents of an archives into electronic media, it will take only weeks or days. This development may also represent a coming wave in the further course of our present 'Electronic Revolution', setting off another Gold Rush of activity by making access to historical data aga@in much easier and much less expensive to work with.
How long? How long before we have scanners that Will scan whole file cabinets instead of single documents? No time soon. Unless work is going on at this moment to develop this technology, then we must expect at least ten years of research and development and probably as many as twenty, before the idea finally becomes a workable prototype. This basic research is by no means cheap, but there is a clear, though presently unrecognized need for it.
Developing a such a mechanical means to scan large numbers of documents in an inexpensive way will give the company that develops it a key position in a large field, one that is yet largely unexploited, and one that otherwise deals with this challenge through the basic (and often prohibitively expensive) technique of applying more hands to the task.