10.1 Why index?

It's no good having useful information on your site if your users can't find it. It's hard to predict how they'll look for it, but if you have a full-text index of your pages, you won't have to guess how the users will look for information, and they'll have the best chance to get hold of what they need.

Unless you go to some trouble to prevent it (as described in Chapter 7), your site or at least those pages that are linked to from elsewhere is searchable by external search engines. But they won't do this job very effectively.

There are two problems with using external search engines for this function. One is that they can't find files that they don't know are there if items are present but not linked from pages that the search engine has seen, they can't be seen or indexed. The other problem is that it's more user friendly to offer a search function on your home page than to tell your users to go off to Google for it. (If you decide to use an external search appliance, you can offer a form on your home page that has as its action a script that runs on the appliance in order to get this capacity without running it on VMS.)

If your site is very small, you can, of course, implement a search function that searches through your Web tree for either specific strings (even using plain old DCL SEARCH) or regular expressions...