Information Provider

There are many ways of making your new or existing data available on the "web" . The best method depends on what sort of data you have. (If you have any questions, mail the www team at www-bug@info.cern.ch.). See also: Web etiquette . How can I help ?

You have plain text files

If you have some plain text files then you can easily write, or generate using a script, a small hypertext file which points to them. To make them accessible you can use either anonymous FTP , or the HTTP daemon .

You have a NeXT

You can use our prototype hypertext editor to create a web of hypertext, linking it to existing files. This is not YET available for X11 workstations. This is a fast way of making online documentation, as well as performing the hyper-librarian job of making sure all your information can be found.

Using a shell script

An HTTP daemon is such a simple thing that a simple shell script will often suffice. This is great for bits of information available locally through other programs, which you would like to publish. More deatils on writing servers using shell scripts under unix , or in DCL under VMS .

You have many files

In this case, for speed of access, the HTTP daemon will probably be best. You can write a tree of hypertext in HTML linking the text files, or you can even generate the tree automatically from your directory tree. If you want to generate a full-text index, then you could use the public domain WAIS software - your data will then be accessible (as plain text, not hypertext) through the WAIS gateway .

You have an existing information base

If you have a maintained base of information, don't rush into changing the way you manage it. A "gateway" W3 server can run on top of your existing system, making the information in it available to the world. This is how it works:

Menus map onto sets of hypertext links
Different search options map onto different "index" document addresses (even if they use the same index underneath in your system).
Procedures used by those who contribute and manage information stay unaltered.

If your database is WAIS, VMS/HELP, XFIND, or Hyper-G, a gateway exists already. These gateway servers did not take long to write. You can pick up a skeleton server in C from our distribution . You can also write one from scratch, for example in perl or sh . An advantage of a gateway is that you can maintain your existing procedures for creating text and managing the database. For more information, see:

A case study of one system .

Making a W3 server for existing data.

Allowing multiple selections

_________________________________________________________________

Tim BL