At the beginning of the sixteenth century, the printing press changed the world. With the Internet, we are on the verge of another such change as we head into the twenty-first.

Five hundred years ago, the automated reproduction of text improved the speed and accuracy with which religious thought, political and social ideas, literature, and news propagated through the civilized world. Not only did the printing press represent greater efficiencies, but it made entirely new forms of literature and thought possible—the novel, for example, was all but impossible before this advance. The spread of Protestantism required the printing press; how else could every household own a copy of the Bible in its own language?

Today we brace for a similar revolution, this time facilitated not by platens, paper, and type but by a network of wires that stretches around the world, through which computers relay information to one another. The Internet represents a quantum leap in speed and efficiency over its older sibling, the printing press, and we must suppose it will change not only the way media are distributed but the shape of human thought itself.

I believe the strongest clues to the way this will actually take place lie in the foundation of the Internet: its protocol. Like civil laws that guide us in our social and political life, and like moral principles that inform our daily decisions, the information-sharing protocols of the Internet will gird our civilization with order.

MEANING AND ORDER IN INFORMATION

Critics love to use the phrase, the test of time. If a work of literature, for instance, withstands this test, it must have some merit. What is the basis for this merit? There are really two intrinsic bases: the truth of that which is said, and the skill with which it is said. As Quintillian said, rhetoric requires "a good man speaking well."

These bases are inside the writing itself, but there are also factors outside the text: the binding, the cover, the numbered pages, the index, and the dry shelf on which the book is kept. These factors have to work in favor of the book if its information is to stick around.

How does this relate to the Internet? The hallmark of durable information is not simply good content but also good form. Durability is as much a question of structure as of content. Good structure, both internal and external to the information, creates and preserves the meaning of its content. And the Internet is nothing if not the soul of structure, communicating as it does through strict, simple protocols (http, smtp, ftp) and through predictable browsing formats.

Physical libraries have traditionally represented the strongest bastion against the loss of information. The Library of Congress catalogs and preserves everything from ancient philosophy to American folk music. The New York Public Library houses thousands of original modern manuscripts. Interest-organizations preserve texts which are related to their interests; my Reformed Protestant alma mater keeps several early editions of John Bunyan's Pilgrim's Progress and a first edition of his Seasonable Counsel in the "Bunyan Room" above the library.

When the Library at Alexandria was trashed by a mob of barbarians in a.d. 400, the greatest storehouse of ancient information moved from reality into myth. All but seven of Sophocles' 123 plays were burned in the stoves of Egyptian bathhouses. The civilized world regrets losing those writings, not for sentimental reasons, but because they would have advanced our understanding of how humans communicate with one another, as well as increased our ability to communicate with one another. Instead, the Western world was plunged into the Dark Ages for a thousand years.

Why did that library in particular mean so much to civilization? Because of the high quality of the communication it preserved—and so we are back to the intrinsic bases for durability of information. The destruction of the Library of Alexandria is perhaps the most compelling proof of the inability of information to stand on its internal merits alone. Texts need to be articulate, organized, indexed on the inside, and bound, shelved, and protected from the elements on the outside.

These principles of information structure are ever-important in the in formation age, as the volume of information globally grows exponentially. Matthew Butterick, president of AtomicVision, a Web site development company in California, made the following point in a speech to the Netscape Developers' Conference in 1996:

Suppose the phone company decided to publish everyone's names and numbers in random order. Clearly, all the data you need would be available to you, but just as clearly, it would be completely useless because it's not information. The real service that the phone company provides with the phone book is taking the data and giving it logical structure (alphabetical order), visual structure (a multi-column format) and physical structure (a bound volume). The value is not in the data per se, it's in adding structure and thus converting data into information. (http://www.atomicvision.com/manifest/design/fuse.html)

The phone book succeeds because it adds structure to chaos. For that matter, the phone number system succeeds because of its standards: in our country always a three-digit area code, always preceded by a "1" or "0", followed by a seven-digit number. In human communication, structure and predictability are everything.

PROTOCOL AS THE FOUNDATION OF ORDER

In all good literary criticism, the critic pays attention to the form's inherent qualities; you don't judge a novel as though it were a poem, or a pop song as though it were a opera libretto. In seminary we called this a hermeneutical evaluation. So, in order to begin to form a basis for criticism of the Internet, we need to mark out its essential qualities. These qualities begin in its physical form (its very wires and protocols) and extend into its applications (i.e., existing Web sites, LANs, and other systems).

On the Web, the first four letters of every URL (Universal Resource Locator) are always the same: "HTTP." HTTP stands for "Hypertext Transfer Protocol." http is the basic set of regulations that govern the movement of browser-oriented data over the Internet; it is an open standard. It's free, so any browser or operating system can be written to comply with it.

Elsewhere on the Internet, although you probably don't ever see them, the letters "SMTP" are included in the transfer of every email message you send and receive. They stand for "Simple Mail Transfer Protocol." They tell your computer "Here comes Simple Mail. It's about to transfer. Prepare yourself!"

These rules for transfer remind me of one the innovations of last century. I found this on the Web:

American inventor, pioneer, mechanical engineer, and manufacturer Eli Whitney is best remembered as the inventor of the cotton gin. He also affected the industrial development of the United States when, in manufacturing muskets for the government, he translated the concept of interchangeable parts into a manufacturing system, giving birth to the American mass-production concept. (http://invent.nforce.com/book/book-text/108.html)

Another nineteenth-century figure who made a name for himself by following Whitney's model of open standards was gun designer and manufacturer Samuel Colt, the inventor of the revolver. Colt's pistols and rifles made mass-production of bullets and gun parts possible, as well as the revolving bullet chamber inside the weapon itself. Five bullets of exactly the same dimension could be fired from the same barrel. Bullets could be calibrated to fit different standard barrel sizes. This replaced the awkward, time-consuming, and imprecise process of muzzle-loading. Later, Colt invented the first machine gun, which was put to devastating use in World War I. There was a saying in the old West: "God made men; Samuel Colt made them equal."

The open standards of the Internet render every developer and software company equal. Everyone has access to the same basic set of resources, just as all painters have access to the same palette. It's the same way with the standards of good literature and effective communication. Every writer or speaker knows that, in order not to drown in the swim of history, he or she must observe standards which are reapplied and reinvented for each cultural era. There is a protocol in Chaucer to be followed. This is what accounts for timelessness in literature. (Many postmodern critics would disagree with this logic. Tough.)

The wonder of the Internet is entirely made possible by a rigid basic structure. Here, as in other forms of communication, predictability is everything. One of the most-often accessed sites on the Web is Yahoo (http://www.yahoo.com), which is nothing more than a text directory. Many other prominent sites—Hotwired (http://www.hotwired.com), Slate (http://www.slate.com), Salon (http://www.salon.com), ESPN (http://espn.go.com), the Microsoft Network (http://www.msn.com)—are similarly built around textual, linguistic communication. Their purpose is to convey information, sometimes editorial and essayistic, as in the case of Slate, and sometimes functional, as in the case of Expedia, the online flight reservation system. Sometimes they convey images, database information, stats, reporting, and opinion, as in the case of ESPN. On these sites, graphics are usually representational, not realistic. Audio clips are a sidebar item. Scan-friendly layouts, intuitive and consistent navigational systems, and bullet-pointed indexes are key to their success. All visual effects drive content.

Unlike broadcast media, the Internet began as an interoffice communication tool for the Pentagon, and then for military intelligence on a larger scale, and then migrated into university use. It was essentially a nationwide LAN (more accurately a WAN, or Wide Area Network). The Web began with text-only browsers whose revolutionary feature was Hypertext—Lynx, for instance—until graphical browsers became more and more popular. E-mail has almost always been text-based, except in the silly things Netscape is trying to do with html-formatted messages (which is essentially like sending someone a Web page in stead of an e-mail). But there is always a little craziness at the inception of something new.

Visual and logical order is more important than ever with the advent of the information age. To preserve and access information, we have to impose order. We don't want to make the same mistake the Alexandrian mob made.

Fortunately, the very basis of computing is order, so that if barbarism were to become the norm on the Internet, at least it would not be barbarians that held court. It would more likely be scientists: those in charge of standards, those who write the protocol. Digital media is composed entirely of digits—zeros and ones. Faster and more powerful computers require greater efficiency. Faster networks require more efficient wires, and more sophisticated data compression techniques. In a sense, high-tech can never be lowbrow, at least at its structural level.

And though the wires of the Internet may be used to convey video entertainment, and audio events as well, and even giant jpegs of naked people in compromising positions, the essence of the Internet is not non-interactive media such as these. It is dynamic information. And the production of dynamic information is best done with an understanding of the deep order of durable literature.

THE RELATIONAL DATABASE

I work for a company that helps other companies manage and use their information more efficiently. Our primary strategy in this effort is to create databases and integrate them with Web sites. Most of our clients have databases or some form of data management already in place, and so we do our best to work with those.

For instance, one of our clients builds air compressors. They use distributors to sell these compressors all over the world. They manage a large group of sales representatives to sell the compressors to the distributors. To keep track of these two groups, and to en courage fairness, they have divided the world into territories, to which both sales reps and distributors are assigned. These three information groups enjoy an ever-evolving relationship. For example, when a contract is negotiated with a new distributor, whose offices are located in a given territory, a sales rep must be assigned to that distributor—either a rep already working that territory, or a new one who will be assigned to that territory. Sometimes our client has to create new territories by subdividing existing ones.

In any case, whenever a new contract is drawn up, the existing base of our client's information has to be modified. Also, news of the modification has to be shared with relevant parties. In the past, this information has been maintained by two secretaries working in separate departments at our client's HQ, one using a Microsoft Excel spreadsheet and one a Filemaker Pro database. When ever a change took place, a fax would issue from one department to the other, the change would be entered manually, and the HQ periodically would mail out a revision of the "Distributor List," a printed document which indexed all three groups. The printed document contained mistakes, as all printed data-documents do, irrevocable until the next printing of the document. This is the old way of doing things, in their "legacy application," to put it in software developers' jargon.

To smooth over these mistakes and reduce much of our client's clerical overhead, my company created a relational database which unifies the whole system, storing information associated with each of the distributors, each sales rep, and each territory. It is composed of a table of information called "Distributors," a table called "Sales Reps," and a table called "Territories." These three tables are dynamically related, so that when a modification takes place in one that is pertinent to some information in another, that pertinence is preserved. The management of the database takes place via password-protected "administrator" Web sites which can be accessed anywhere on the Web.

How boring is that, you ask? Well, it sounds dull at the outset, but I can as sure you that the staff who handled the changes prior to the conversion find their new system quite exciting. Information maintenance is not an entirely human job.

The tools we use for this kind of development are, as I said, database (which for us takes the form of Microsoft SQL) and the World Wide Web. We build databases and put them on the servers that host the Web sites, and we write scripts that allow the Web sites to ask questions of the databases. This makes the Web sites useful, not just pretty, which is one theme we emphasize in our own promotional material.

This work we are doing is based on two great technological advances. One is the Web, which is really not a source of its own information, but is rather a way of sharing and viewing information that already exists (and some of which is developed specifically for the Web). The other one is the relational database, which is a relatively new phenomenon. The relational database can support Web content.

A traditional database is essentially a reference chart. The multiplication table we memorized in grade school is a database. You follow the row you want ("6," for example) to the point at which it intersects with the column you want ("9," for example) in order to find the product ("54").

The relational database model was proposed in 1970 by Dr. E. F. Codd at the IBM research labs in San Jose, California. He began to write something he called Structured Query Language (SQL), which would form the basis of the model. Ten years of laboratory research led to IBM's 1981 release of SQL/DS. Big companies such as Oracle, Relational Technology, and Microsoft followed with their own implementations of the language, and by 1988, SQL relational databases were being developed by over 200,000 computer users.

The difference in the relational model is that it allows data to be dynamically related to other data. For instance, if you are an Excel spreadsheet user, you know that you can enter a formula in one cell that sits in the background of the spreadsheet and performs a function on other data. You can have a "grand total" cell that dynamically adds up the cells above it, so that when you change any number in the list, the total changes itself and remains accurate. This is a simple example of a dynamic database. Whereas, if you changed any number in the grade-school multiplication table, it would render some of the other information incorrect.

DYNAMIC INFORMATION

I have said that the orderliness of literature—its structure—has everything to do with its durability in the library. I would like to apply that premise to the new kind of information that is conveyed over the Internet, which is known as "dynamic information."

What's new about dynamic information is that it doesn't exist anywhere. Whereas in a physical book a reader encounters static information and draws his own dynamic conclusions, the data base-supported Web site automates some of that process. Dynamically generated content comes into existence in the same way a calculator's answer comes into existence: on the fly. It's something like one of those old "Choose Your Own Adventure" books Bantam publishes, except with infinite possible conclusions.

Consider the following situation:

I've had my eye on a car for sale: a white 1986 Chevy Suburban Scottsdale, parked in front of a liquor store on Hampton. It has 109k, is touched slightly with rust, and boasts absolutely no creature comforts. It appeals to my Germanic ethnicity. Recently I took it for a test drive with its owner, Tom, sitting at the other end of the bench front seat. "I bought it from an older man who passed away. He did to it all the things that old men do; you know, change the oil every 3,000 miles, replace the belts and hoses periodically." I believed him, because the thing drove and handled very solidly. "Like a rock," one might say. Unfortunately I don't have the $5995 Tom's asking.

Also I have no need for a truck. I'm not a carpenter or a painter; I'm an ex-English-student who spends almost the whole day on the phone and "surfing the Web." I want a truck because I'm a wuss who wants to feel like a man by driving a genuine dad-car. Not an SUV, mind you; no, a half-ton 2wd Chevy, complete with vinyl flooring. "You can just spray it down," Tom said, "if you track mud in."

But because I'm an info geek I also have an advantage: I know exactly what kind of truck I'm looking at within an hour of seeing its for-sale sign. My power tool in this situation is Microsoft's Carpoint (http://carpoint.msn.com), the one-stop buyer's guide to new and used cars.

Its best feature is an interactive Kelley Blue Book (http://carpoint.msn.com/kbb). "What Is Your Car Worth?" the section begins. It then provides a field for you to enter your zip code and a pull-down menu to enter the year of your car. When you've selected a year, another menu appears with a master list of makes; then a menu with models, then trims. Once all four menus have been selected, a flashing "Find Out" button becomes active on the right side of the screen.

Clicking that takes you to a more specific screen, where two drop-down menus need to be selected, labeled "engine" and "transmission." Since these menus are dynamically generated from the information you entered on the previous screen, they are customized to your car's make and model.

Once you fill out the final field, specifying mileage, and press "enter" on your keyboard, the Carpoint database begins miraculously calculating your car's value. The first number it brings up is trade-in value. As you fill out the rest of the fields on the page ("Drive," "Option Package," "Audio," etc.) the number at the top changes slightly. If you click on "What is my car worth if I sell it myself?" at the bottom of the page, you will be taken to a similar screen entitled "Kelley's Suggested Retail Value." My target Suburban listed at $3925 trade-in, $7,225 retail; now I know Tom's price is fair.

From either of these screens you can click on "View Classifieds Listings" in the general navigation on the left. This takes you to a screen where you enter your zip and the kind of car you're looking for, and it takes you to a list of vehicles for sale in your area. By setting sliding-arrow options at the bottom of the screen, you can control the year, price, and mileage of the listings in view.

Carpoint does exactly what a data base-driven site ought to do: it keeps you from physically having to scan through a number of charts and perform small calculations ("add x dollars for this many miles in excess of average; refer average mileage table, page so-and-so"). These are tasks an Internet server can perform easily and quickly for thousands of end-users simultaneously.

Other features of Carpoint include downloadable surround videos of a number of new cars, side-by-side comparison charts of any two new or used cars, a handy car-loan payment calculator, and a section in which you can research the specs of almost any car produced in the last 20 years (fuel economy, wheelbase length, safety features, etc.). For instance, I just found out that the 1998 Bentley Brooklands, assembled in Great Britain, had a starting price of $155,900 and has a turbocharged 6.75-liter v8 which develops 300 horses at 4000 rpm. There is no fuel-economy data for this car, but the side-by-side comparison shows me that a 1986 Chevy Suburban is cheaper and not as powerful. See how useful Carpoint is?

DEEP ORDER

I hope that physics is not the only field left in which deep structural order is the basis for advancement. I have most of my formal training and practice in poetry, and in a strange way, I feel as if I've temporarily left my own field, be cause it's gotten to be too much about you and me. The judgment of poetry in America today has become so tied up in academic politics and personal feelings that it has become weak. Books and authors have proliferated, but their work sounds more and more miscellaneous, more arbitrary.

In literature built to last, nothing sounds arbitrary. A good poem from the 1920s doesn't sound worn out:

This Is Just To Say

I have eaten
the plums
that were in
the icebox

and which
you were probably
saving
for breakfast

Forgive me
they were delicious
so sweet
and so cold.

—William Carlos Williams

Great literature never sounds old. Antique, maybe, but never like yesterday's fashion statement. Reading a modern translation of a very ancient text earlier this year, I was moved to tears at the relevance of it in my contemporary context. My wife and I had our first baby in March, 1998, and we've been worried about not having enough money to support our little family; and I work in downtown St. Louis, which seems ever more vacant and desolate; and I found this old poem—a poem that moves so intuitively as to be almost illogical—right on point:

Unless the Lord builds the house,
its builders labor in vain.
Unless the Lord watches over the city,
the watchmen stand guard in vain.
In vain you rise early and stay up late,
toiling for food to eat—
for He grants sleep to those he loves.

Sons are a heritage from the Lord,
children a reward from Him.
Like arrows in the hands of a warrior
are sons born in one's youth.
Blessed is the man whose quiver is full of them.
They will not be put to shame
when they contend with their enemies in the gate.

—Psalm 127

The power of good literature is unmistakable, and has much to do with its protocol: how it balances so steadily between familiarity and surprise, how by its ordered presentation of content it teaches us to remember what it is saying. In this way, words well formed entertain us, nourish us, and give hope to people who often feel hopeless. Think about the speeches of Martin Luther King, Jr. Who can help but remember?

As a Christian, I would argue that the creation of order from orderlessness is among the holiest and most primal of activities. The first two verses of the Bible say, "In the beginning God created the heavens and the earth. Now the earth was formless and empty, darkness was over the surface of the deep, and the spirit of God was hovering over the waters. And God said, 'Let there be light,' and there was light" (Gen. 1:1-2).

The imposition of order is not re served for God, as evidenced in Genesis 1:26-27, in which God says, "'Let us make man in our image, in our likeness, and let them rule over the fish of the sea and the birds of the air, over the livestock, over all the earth, and over all the creatures that move along the ground.' So God created man in his own image, in the image of God he created them; male and female he created them."

Even in this text, there is a literary device at work, intended to help you remember:

formless = empty

in our image = in our likeness

God created man in his own image = in the image of God he created them

Such patterning can be found through out Scripture: "From six calamities he will rescue you; in seven no harm will befall you" (Job 5:19); "Some trust in chariots and some in horses" (Psalm 20:7). And similar devices are found in other works of the oral tradition, such as the Scandinavian epic, The Kalevala, and all through the Iliad and the Odyssey.

Order on the Internet must resemble the kind of order that has given meaning and durability to other forms of information through the centuries. If we are to communicate well in this medium, we'll have to buck our cultural philosophy of practicing "random acts of kindness and senseless beauty"; we'll have to work in specific ways to achieve sensible beauty. Our Web sites, databases, search engines, e-mails, FTP sites, and software interfaces, should be purpose-driven, bearing the best kind of useful information, rather than an ocean of unformed data.

Aaron Belz is an interface designer and a poet. The description of Carpoint first appeared in slightly different form in The Riverfront Times.

Copyright © 2000 by the author or Christianity Today/Books & Culture Magazine. Click here for reprint information on Books & Culture.