A newsreader is a program that users invoke to view, store, and create news articles. Several newsreaders have been ported to Linux. We will describe the basic setup for the three most popular newsreaders: tin, trn, and nn.

One of the most effective newsreaders is:

$ find /var/spool/news -name '[0-9]*' -exec cat {} \; | more

This is the way Unix die-hards read their news.

Most newsreaders, however, are much more sophisticated. They usually offer a full-screen interface with separate levels for displaying all groups the user has subscribed to, an overview of all articles in each group, and individual articles. Many web browsers double as newsreaders, but if you want to use a standalone newsreader, this chapter explains how to configure two classic ones: trn and nn.

At the newsgroup level, most newsreaders display a list of articles, showing their subject lines and authors. In big groups, it is difficult for the user to keep track of articles relating to each other, although it is possible to identify responses to earlier articles.

A response usually repeats the original article’s subject, prepending it with Re:. Additionally, the References: header line should include the message ID of the article on which the response is directly following up. Sorting articles by these two criteria generates small clusters (in fact, trees) of articles, which are called threads. One of the tasks of writing a newsreader is devising an efficient scheme of threading, because the time required for this is proportional to the square of the number of articles.

We will not go into how the user interfaces are built here. All newsreaders currently available for Linux have a good help function; please refer to it for more details.

In the following sections, we will deal only with administrative tasks. Most of these relate to the creation of threads databases and accounting.