Consumer Rankings

The ranking software I presented this month is only in its earliest stages and will undoubtedly have improved by the time this issue of Linux Journal reaches your hands.

Along with improvements in the user interface and the possible addition of a hierarchical category system, there are several related applications I would like to add. One would return the highest-ranked items within a category, allowing a user to find the best choice without having to wade through dozens or hundreds of reviews. Hard-core users of this system might want to read only those reviews which have appeared since the last time they visited the site. Some additional personalization, including the use of cookies to remember the user's name and e-mail address, would reduce the amount of typing someone would have to do. An editing facility for system administrators will undoubtedly prove useful as the system grows. Finally, it is always nice to provide a search function, just in case an Italian restaurant is accidentally classified as a café.

The programs we examine this month demonstrate that it is not particularly difficult to produce a simple user ranking system. This is especially true if we take advantage of the power a relational database affords us in storing and retrieving data. Best of all, subscribers to my e-mail list can now spend their time trading information, leaving the recommendations to a set of CGI programs.

Resources