The World Wide Web is a global copy machine attached to a digital library. This is no accident. It was purposely designed that way to allow and encourage copying and sharing of digital resources. Instead of buttons this copy machine utilizes URL addresses and the Information Super Highway to distribute digital copies around the world. People at computers ''push the copy buttons'' whenever they type URL addresses

into the address box of their web browsers.
        Webmasters post original source documents, including html documents and image documents, on special computers known as servers. These source documents are the blueprints for webpages. The servers that store these original source documents can transmit copies to any computer connected to the Internet. The receiving computer is called a client. Browser software on the client computer reads an html document and any related (jpg and gif) image documents and combines the text and images into a single browser image called a webpage. All the text is contained in the html document and each separate image is contained in its own distinct source document. Suppose you view on your computer a webpage that has text and three images. Then you have downloaded copies of four source documents: one for the text and three separate source documents for the three images. It seems like you are viewing an image out somewhere in cyberspace. You may imagine that you are viewing an image from another computer, but in reality every webpage you see "on the Web" has been created on your computer from copied source documents.

        An html document contains not only the webpage text, but also the URL locations of all links and all images displayed on the webpage. Any computer that opens the html document will automatically request copies of the image documents at the specified URL locations. That is why connections that produce embedded images are sometimes referred to as direct links. The standard hyperlink, that seminal feature that makes the World Wide Web a web, initiates a request for copies of the linked document if and only if the user chooses to click on the link. Unlike these click links, direct links do not involve any specific action on the part of the viewer. This is the way all images are loaded onto a webpage even when they come from the same website as the html document. Each image document arrives after the html document, and the image and the html text are then combined on the client computer. Unlike printed media, all webpage images are copies that have been "pasted" into their webpages. Although images appear to be an integral part of the webpage, they are posted on the Web as separate and distinct entities with their own URL addresses.

        When you view a webpage, only a single URL address appears in your browser's address box--the address of the html document. This contributes to the mistaken feeling that the webpage is a single entity. Image documents, however, do not need to be combined with an html document to be seen. They can be viewed individually simply by typing their URL address into the browser address box. The image will then appear all by itself in the browser window in its original size.
        The fact that any image can be displayed on any webpage or linked to it creates a great deal of flexibility for the entire Internet community. The owner of an image--the person on whose website the document is posted--can insert that image into any number of his own webpages by writing that URL address into the corresponding html documents. The same holds true for other webpage authors. Anyone can write an html document which will download and display the image. This, as stated above, is exactly what the Internet was intentionally designed to do--share computer resources. Hyperlinks have made it simple and convenient to make these connections.
        They have also made it simple and easy to break connections. That is why there are so many dead links on the World Wide Web (Wales et al., 2005). A webmaster links to an external webpage or image without the need for any coordination or communication. If, however, as often happens the external webpage or image is moved, removed, or renamed that hyperlink fails to work. It is here that directly linked images have a great advantage over click links. When direct links go dead the image disappears from the webpage leaving behind an empty box. Because it is visible, anyone who views the webpage knows immediately what has happened, and corrective action can be taken in a timely manner.
        These are the essential characteristics of the World Wide Web: it is both a decentralized digital library of html and image documents and a global copy machine. Documents can be linked and images inserted with great ease. Ideally the process is a symbiotic one in which document owners and webmasters interact in a cooperative, productive manner to build a continually expanding and improved Web of resources available throughout the world. The world is not an ideal one, however, and issues come up. I would now like to explore four of them: control, credit, transmission costs, and communication.

        The Internet in general and the World Wide Web in particular are maintained and used by such a great diversity of people that it may be "the most ... extensive medium of personal exchange" in history (Wales et al., 2005). There is an overtly commercial sector, which seeks a direct profit for providing access to its documents. The economic law of supply and demand dictates that such access be restricted. The same is true of websites that handle information of a private nature or where users may be engaged in fierce competition with each other, such as college and financial aid applications. This paper will not concern itself with such websites, except to note that they employ a number of successful strategies to limit access and exert control over their materials. Such websites typically use a system of usernames and passwords. Other strategies include the blocking of direct links from external html documents and disabling the print function on the client browser.
        Although webmasters and the managers of their servers can limit access to the documents on their website, most choose not to do so. They want as many people as possible around the world to view the text and images they provide. That is precisely why they posted them on the World Wide Web. Some web authors go so far as to include counters in their html documents in order to measure the traffic and then boast of how many hits their webpage gets, how popular it is. The problem for those who wish to retain control of their materials is that to the extent that you provide access, you give up control. There is a trade-off between the two. Artists, including photographers and writers, are particularly sensitive and vulnerable. They understandably want to retain artistic control of their creations.
        Comparison with a typical printed product, such as a book, will illustrate what a great difference the Internet makes. Books are printed in limited numbers. People pay for them, read them, and sometimes pass them on to friends. Some books are put in libraries, where many people can borrow a single book. Most libraries have copy machines, and it's not unusual for people to copy sections of some books. Copying an entire book, however, is time consuming and costly as well as obnoxious. It is almost always easier and less expensive to buy the book.
        The non-commercial webpages under discussion in this article are posted on a global copy machine for any and all to access free of charge other than fees paid to an Internet service provider. Anyone who has viewed a webpage has requested and received a temporary copy of its html document and of each image it displays. Making more permanent copies of the images or text is often as easy as click-and-drag or copy-and-paste. There is virtually no economic incentive or barriers involved since the text and images are freely available on the Internet. Any computer anywhere in the world that is connected to the Internet can copy such source documents. Unlike a library of printed material, you do not actually borrow or directly view a webpage. Instead, you get a copy of its blueprint--the source documents; the webpage is constructed on your computer; and you never return the copies. It is as if a library printed up copies of its own books and gave them away for free. Copies of documents on Internet servers are transmitted directly from the websites of owners to client computers around the world. Let us call this the primary mode of distribution.
        The primary mode occurs as a chain of events that ends with the original website--the document owner's website--transmitting a copy of the original source document to a client. We will call him the primary client, because he receives the document from the original website. This chain of events, furthermore, always begins with the primary client. There are three possible routes the chain may take according to whether the request for the document is initiated with (a) a typed URL, (b) a click link, or (c) a direct link. When a primary client types out the URL address, he probably has a good idea of what document or kind of document he is requesting. The search for webpages, however, often begins by going to a search engine, like Yahoo or Google, with a URL address that is easy to remember.

After that it usually continues with a series of click links each of which leads to (ie. downloads) a single html document. The client makes a conscious choice to click words, phrases, or images contained in an html document. The words, phrases, or images will often give them a hint as to the document or kind of document that they are requesting. A click link to an html document may also download multiple directly linked image documents. This is an automatic consequence of viewing that html document. Where a click or direct link leads, that is which documents it tries to download onto the client computer, of course, is decided not by the primary client or by the original document owner, but by a third party--the author of the html document. Yet document owners are still in full control of this mode of distribution. The website can distinguish between a typed request and a link request. Websites can and sometime do refuse external link requests for documents. The owner can impose limits on this access or even remove some documents and thereby end access to them completely. What owners cannot control is the secondary mode of distribution.
        The secondary mode occurs as a chain of events that does not include the original website. A primary client can transmit document copies to other client computers via e-mail attachments. These secondary clients, or any primary client, can also upload these documents onto websites other than the original, where they will be available to the entire cyberworld. As long as access through the primary mode is unlimited, access through a secondary mode merely serves to augment the distribution of those documents to people who already have access. Rather than unauthorized access the main issue is acknowledgment.