Next: Staffing Up: Planning aspects Previous: Planning aspects

Organizing information on the Web

The raison d'être of the Web server is the information provided on it, and data organization is probably the single most important design issue in setting up a Web site. This issue needs to be thoroughly discussed with all parties concerned as the implications are very far-reaching.

The most important point to grasp here is that once a hierarchy of information is in place, altering its structure will prove very difficult. In large part this results from the fact that any Web user, anywhere in the world, may have created their own links to the documents within that structure and the Web site has no way of knowing if this is the case, and if so how extensive the links are. If documents are deleted, moved or renamed the integrity of the Web is compromised. The only current way round this problem is to leave a place-holder document as a substitute, containing either a link to the new location or name, or a note to the effect that the document has been deleted. This is simply not practicable on a large scale. Of course, if a document is deleted and someone subsequently creates a new document with the same URL, people following old links may get very confused indeed!

Server machines should be assigned logical names or aliases that reflect their rôles, for example www.site.domain, gopher.site.domain, or ftp.site.domain. Thus if these services are all being run on a single machine, the machine might have several different logical names.

Creating a directory structure can be full of pitfalls for the unwary! The following example may help to illustrate the case in point. If the R&D department within the organization XYZ puts up a Web server independently, it may just choose the host-name www.xyz.org and create a document containing the departmental phone list at the top level of the hierarchy, which would be accessed by the URL: http://www.xyz.org/phonelist.html. If other departments within the organization subsequently want get in on the Web action, how do they differentiate their documents from those of R&D? A logical component of the hierarchy has been omitted. The design of information hierarchies can be quite subtle and requires careful deliberation.

An organizational mechanism will be needed for editorial staff to update information content for Web documents that are under the overall control of system administrators. Some organizations use version control software for keeping track of changes to documents, which can be used to regenerate old versions of documents if necessary.

If you have a body of information that you are publishing on the Web, you may want to consider running other Internet information servers, such as Gopher or FTP, in parallel, in order to reach a wider audience. If your organization does not yet receive USENET news you may wish to consider this. There are a number of news groups that cover the Web and other information systems.

Spinning the Web by Andrew Ford
© 1995 International Thomson Publishing
© 2002 Andrew Ford and Ford & Mason Ltd
Note: this HTML document was generated in December 1994 directly from the LaTeX source files using LaTeX2HTML. It was formatted into our standard page layout using the Template Toolkit. The document is mainly of historical interest as obviously many of the sites mentioned have long since disappeared.