Along the "Information Superhighway" one may now find service stations: sites whose intent is to make money. But most of these are merely front-ends to an already-established order entry system. To tap the vast potential of this new market, companies must understand its environment and surmount several challenges. I see at least three issues that need to be addressed: locatability, applicability, and security.
Locatability has to do with how potential customers find a site. While URLs are now included in many TV and print ads, I don't expect most web-surfers to type in those URLs (unless they are very easy to remember). They will want to consult a list instead. Unfortunately there is not yet an established "Yellow Pages" for the Web. Various search engines try to locate keywords within page content, but the results are often unsatisfactory. (A search for "Javascript example" might turn up a page whose content is "I hate HTML gimmicks -- Javascript, for example.") The most well-known Web organizer, Yahoo, is on the right track in two ways: categorizing entries, and creating regional sublists. The next step may be to include capsule descriptions (for a fee) and a standard set of keywords identifying specific goods and services.
That brings us to applicability: how well a product fits into the Internet conventions. Yes, you can order a pizza from your Web browser; but except as an advertising gimmick, there's probably not much profit for suppliers. Phone ordering works too well in this case, so few pizza stores can afford to tie up a phone line connecting to the Net (nor can its employees spend the time to read E-mail in a timely fashion). The best opportunity of this kind would seem to be for mail-order companies. Here response time is not an issue, and the use of hyperlinks, animation, and form fill-in could make an electronic catalog much more interesting and convenient than the printed version. Another opportunity which is Internet-specific is document retrieval; in this context, a "document" can be a book, photograph, technical report, or anything else that can be sent electronically. However, some of these are more practical than others. If you are looking for a paperback novel, the content may be too large to store on a local disk, and there's no convenient way to take it with you (unless you have a laptop, I suppose). Technical reports and reviews are more suitable, as they are smaller and usually read all at once (to be either discarded or archived afterwards). But there is a problem for suppliers: unauthorized distribution. Even with the availability of copying machines (from Xerox or others), it takes time and effort, as well as supplies, to make hardcopies for other people. But electronic distribution can be accomplished as fast as you can say "E-mail".
And so we come to security, which includes a number of separate issues. Two are worth noting: authorization and encryption. Authorization means ensuring that only a paying customer can make use of a product or service (and also that only the supplier can obtain payment, and only for the agreed amount). Encryption refers to a process for hiding information from unauthorized users while it is in public territory (e.g. being sent across phone lines). Many organizations are working on these issues, and some solutions are already in place, but again there is no generally-accepted mechanism on the Internet yet. In the case of authorization it may not be possible to adopt a net-wide solution, since the huge number of users precludes any centralized database. Companies will probably adopt modified versions of their current methods, such as the use of pre-established accounts or going through a trusted intermediary (credit card company or bank).
In summary, there is still much work to be done before the Internet becomes a viable marketplace. But even so, one thing is certain: given the vast potential for profit in this arena, someone will find ways to exploit it.