Running a Gopher Server in 2018

Original published by Paul Scott at

If you read the History of Gopher then you might want to make your own server. The process isn’t complicated, thanks to an application called Pygopherd. You will need a Linux machine or virtual machine (this article uses Ubuntu) with Python installed (default on most modern distros). There are just a few steps to follow.

Step 1: Make sure that pygopherd is in your repositories.

$ apt-cache search pygopherd

Step 2: Install pygopherd

$ sudo apt-get install pygopherd

Step 3: Configure pygopherd

Open /etc/pygopherd/pygopherd.conf in your favorite editor. The config file is well-commented and mostly self-explanatory. The most critical change is the name of your server, which allows a gopher client to find your server. In your text editor find “servername” and configure:

servername = <IP or Domain>

Your IP address is available in /etc/network/interfaces.

Step 4: Set up the server

Stop, start, and restart your gopher server

Your gopher server is up by default. Start/stop/restart the service using:

$ sudo /etc/init.d/pygopherd start
$ sudo /etc/init.d/pygopherd stop
$ sudo /etc/init.d/pygopherd restart

Change your default gopher directory

By default, the directory where your gopher-related files are stored is /var/gopher. To change this directory open up pygopherd.config and edit the setting root=/var/gopher (located around line 120 in the config file) to your desired directory. Don’t forget to add a gophermap to the new directory. (For more about gophermaps, see next.)

Step 5: Configure the gophermap and Add Content

Your gophermap is an index for each folder in your gopherhole.

Technically, a folder doesn’t need a gophermap. However the gophermap file can be quite useful. It gives you more control over the structure of your gopherhole, lets you add links and text to your pages.

With a gophermap your server will appear as a list of files and links, with comments and other information that you might choose to add. Without the gophermap your server will appear as a raw list of files and folders, navigable in a gopher client but with less information for your users.

If you choose NOT to use a gophermap then you must delete the default gophermap file. A gopher client will look for the gophermap by default and if it’s there then it will override the other content in your folder. If you do use a gophermap then you will have to update it every time you add or delete content from the folder.

Gophermap file syntax

In a Gopher menu’s source code, a one-character code indicates what kind of content the client should expect. This code may either be a digit or a letter of the alphabet; letters are case-sensitive.

RFC 1436, the Gopher technical spec, defines 14 item types. Type 3 is an error code for exception handling. Gopher client authors improvised item types h (HTML), i (informational message), and s (sound file) after the publication of RFC 1436.

Canonical Types
0   Text File
1   Gopher submenu or link to another gopher server
2   CCSO Nameserver
3   Error code returned by a Gopher server to indicate failure
4   BinHex-encoded file (primarily for Macintosh computers)
5   DOS file
6   uuencoded file
7   Gopher full-text search
8   Telnet
9   Binary file
+   Mirror or alternate server (for load balancing or in case of primary server downtime)
g   GIF file
I   Image file
T   Telnet 3270
Non-Canonical Types
h   HTML file
i   Informational message
s   Sound file

To set up a very basic gopherhole, you need only three item types: i (informational message), 1 (link or subdirectory), and 0 (text file). Edit the gophermap file at


A gophermap is formatted as a set of lines, one for each item, using tab-delimited fields with the following syntax:


T      The item type code (e.g. 1, 0, i)
label  A human-readable label for this item (e.g. Books, Movies)
path   The relative path to the target file
host   The hostname of the target gopher server
port   The port that the gopher server is listening on (usually 70)
       A literal tab character (ASCII character 9)

Note that there is no space or tab between the item type and the label.

You can omit the host and port fields if your items are on the same server as the gophermap. You can omit everything but the item type and label for informational items.

So suppose you use the default setup of /var/gopher as your home folder for content and your main gophermap. In that folder you have the gophermap; an “AboutMe” file and three subfolders for a blog, movie reviews and vacation photos.

The main gophermap might look like this, with informational items to explain each folder.

iWelcome to my Gopherhole!
0A little bit about me.<tab>about.txt
iHere's my Blog, "The Technical Technical Writer".
iReviews of some of my favorite movies.
iPics from my vacation in San Diego
1Vacation Photos<tab>vacation

Each of the subfolders (blog, movies, and vacation) will have its own gophermap. The blog and movie reviews will be text files (item type 0), while the vacation photos will be image files (item type I).

Add content

This is the simplest part of setting up your gopherhole. To add content just add files and folders to your gopher home folder and/or subfolders. Update your gophermaps (if you have them) and you’re off!