While directly linked images provide at least a minimal acknowledgment of the linked website, they also place a burden on the that site's bandwidth. Some people complain that this is a form of stealing in its own right. On the contrary, this clearly shows that the original website is distributing copies of its own image document (primary mode). Traffic to and from a website does, indeed, increase each time a document is requested and transmitted. This is an inevitable consequence of any communication between computers. In theory, however, the communication is of benefit to all parties--the party that wants to view the image, the one that wants to display it, and the one that posts it on the Web for display.
        These transmission costs are like the service charge for a cashier's check. Some banks ask their customers to pay a "service fee" when they issue them a cashier's check, usually for a large amount of money. Yet when the customer chooses a cash withdrawal rather than pay, these same banks inevitably waive the fee, because this "service" is in their interest as well as the customer's. Similarly, although websites provide a service to whoever views their material, this service must also have been in the interest of the webmaster and any contributors in order for them to have provided unlimited and free access to it in the first place. Thus the burden of increased computer transmissions is offset or balanced by the additional promotion those images receive.
        The structure of the World Wide Web is based upon links. That is the mechanism by which people navigate from a search engine to the webpages they want to view. It is the links on search engines and other html documents that advertise and promote webpages and thereby make their presence felt. A document free floating in cyberspace, without any links to it, would be invisible, virtually non-existent. Thus links are necessary. In addition, their potential for expanding traffic to any specific document is very powerful. Each link funnels a percentage of a webpage's traffic to the next webpage downstream. For direct links the percentage is 100%. For click links it is the percentage of viewers that choose to click on a particular link. Click links can send a stream of computer traffic through a series of webpages converging with other similar streams to eventually form a raging river of traffic to some well-connected image document. It is this geometric progression of links radiating upstream from a document that gives the Web its power, empowers popular webpages and images, and can sometimes overpower them, even though the burden for each single request and transmission is not great and not affected either by the route it takes or by the nature of the final connection--typed, click link, or a direct link.
        While the owners of documents are necessarily saddled with the burden of primary mode transmission, any distribution in a secondary mode will help them avoid those transmission costs. In other words, there is a trade off between control of access and transmission costs. Owners of popular image and html documents may be willing, even eager, to sacrifice control over their distribution in order to minimize their transmission costs. For this very reason whole websites are sometimes copied and loaded on other servers and the result called a mirror website. This can be accomplished, on a smaller scale, by loading the copy of a single image document onto a new website. The copy can then be embedded in one or more webpages using direct links to the new website, while connecting the embedded images to the original website with click links, and thus providing acknowledgment. The percentage of traffic flowing through the final connection will be reduced from 100% to the percentage of people that choose to click on the reverse link.
        Loading someone else's documents on your own website, of course, deprives them of control over their distribution and thus requires their permission. While this should reduce the ongoing operational burden of computer transmission, it imposes a communicative burden in two directions--one upon the person requesting such permission and another upon the person receiving, considering, and replying to those requests.

        We have established that the World Wide Web is a global copy machine for digital information and discussed some of the possible ways these copies might be shared. Realizing that securing permission from image owners is desirable even when it may not be mandatory, we now turn our attention to how that might best be accomplished--communication. It sounds so easy. Image owners could simply post the terms of use on their website along with an e-mail address for image link permission requests. Whoever makes direct links to images at another site could alert the original website to the existence of the connecting webpage, so that they could inspect it, evaluate the manner of linkage, and respond. In theory, this communication process should be accomplished within a few days, long before the rest of the cyberworld has time to find and make their own connections to the new webpage. Unfortunately the reality of the Information Super Highway is much more complicated.
        The Information Super Highway, after all, not only gives us instant access to information, it also bombards us with it, spam and viruses included. The floodgates are open; the in-box fills up; and we have to deal with it. Opportunity may be boundless, but the competition is fierce. Since we can do more, we must do more. Our lives become frenzied trying to keep up. Websites are continually under construction. Links go dead for want of maintenance. It is not surprising, therefore, that many websites lack any contact information or that some posted e-mail addresses are out of date. Sometimes this failure to provide an e-mail contact is a purposeful strategy to protect busy webmasters from what could be a deluge of e-mail, legitimate or otherwise. If they do not want to discourage all contact, a webmaster may provide a postal address or a special form submission webpage. This filters contact with the cyberworld down to those who are willing to make the extra effort.
        Other factors may also discourage people who would like to make contact. On websites with numerous webpages, contact information may not be on each and every page. Although, in such cases, links are often provided to a page with the information, more and more powerful computers manage to squeeze more and more information onto individual webpages. This information overload can make it difficult to find the appropriate link and/or to find the right contact information even when viewing the correct webpage. This is especially true if the person is not a native speaker of the language used on the image owner's site.
        The search for contact information is one obstacle to communication, but it can usually be overcome. It is mostly a matter of having enough time and making the effort. The next obstacle is more difficult to surmount. After link permission has been requested, the most common response in my experience is silence--no response. We can only speculate on the possible causes: (a) lack of concern about linkage, (b) lack of time to evaluate requests and respond, (c) language barriers. Most people who have responded to my link requests have responded positively, granting permission. Aware that the structure of the World Wide Web in based on links, many people probably do not perceive a need to grant permission for them. Alternatively they may just be too busy to give serious attention to the request. Any language barrier will compound the problems of understanding the request, evaluating the webpage that wants to link, and formulating a response. I used to send English requests to Japanese websites whose images my Japanese students had chosen to illustrate their research papers until an irate Japanese company complained to my university about my sending requests in English. It took me a couple months to work up a bilingual version of the request. I have posted it on the Web (www.aichi-gakuin.ac.jp/ ~jeffreyb/ permission.html) and now refer all Japanese webmasters to that webpage if they prefer Japanese.
        The point to note here is that communication in the real world, even with the Internet, represents a real burden to both parties, the website requesting permission and the website that has to evaluate the request and respond. Some webmasters try to minimize their burden by posting pseudo-legal notices and/or giving formulaic responses to requests. Though this may be a viable tactic to try to preserve whatever rights exist for materials knowingly placed in a copy machine universally accessible to the cyberpublic, the actual legal implications of these notices are extremely difficult to assess. As a global network the World Wide Web is potentially subject to a multitude of uncoordinated and possibly conflicting national laws. How have those laws have been interpreted, reconciled, and applied by national or international courts is unclear. While I would be happy to hear from anyone who is privy to such information, that line of inquiry will have to remain beyond the scope of this paper.