{"id":10824,"date":"2025-06-06T03:00:00","date_gmt":"2025-06-06T07:00:00","guid":{"rendered":"https:\/\/www.both.org\/?p=10824"},"modified":"2025-05-31T16:31:21","modified_gmt":"2025-05-31T20:31:21","slug":"viewing-pdf-documents-in-console-mode","status":"publish","type":"post","link":"https:\/\/www.both.org\/?p=10824","title":{"rendered":"Viewing PDF documents in console mode"},"content":{"rendered":"
\r\n
\r\n \r\n <\/i>\r\n <\/a>\r\n 1 <\/span>\r\n<\/div><\/div>\n

I have a Raspberry Pi 3 that\u2019s sat unused for the last few years. I used to run the Raspberry Pi as a personal \u201chome server,\u201d such as a printer server to share my old \u201coffline\u201d laser printer on my home WiFi. But ever since I bought a new printer a few years ago, I haven\u2019t had a good use for the Raspberry Pi. Until now.<\/p>\n\n\n\n

I started a retrocomputing experiment<\/a> to see if I can still be effective as a technical writer, but doing it without a graphical user interface. I\u2019m doing everything at the command line. To do that, I installed Fedora Minimal on the Raspberri Pi\u2014this gives me only the packages that are necessary to boot the Raspberry Pi. Fedora Minimal doesn\u2019t provide a full desktop environment or even many of the command line tools that I\u2019m used to. If you want to use a command that\u2019s not installed with Fedora Minimal, you have to install the package.<\/p>\n\n\n\n

While this might sound restrictive to some folks, I like it because it lets me explore the minimum<\/em> set of tools that I need to be effective as a writer, without loading a graphical user interface.<\/p>\n\n\n\n

Most things are easy<\/h2>\n\n\n\n

As someone who grew up with computers in the late 1970s and early 1980s, and used DOS from the 1980s into the 1990s, and run Linux since 1993 before graphical desktops were common, I am already comfortable using the command line. I\u2019ll gladly use ls<\/strong> and cd<\/strong> to navigate my filesystem, and less<\/strong> and vi<\/strong> to view and edit content.<\/p>\n\n\n\n

I learned how to write text documents with nroff<\/strong> when I was an undergraduate student using our campus computer lab. At the same time, I quickly learned how to use groff<\/strong> and LaTeX on my Linux desktop at home to provide more interesting formatting for other documents, such as equations and charts in my lab reports for physics classes.<\/p>\n\n\n\n

These days, I write most of my first drafts in Markdown, then convert to something else to make final edits or to transmit the file to someone else. Writing in Markdown with a plain text editor like vim<\/strong> makes it easy for me to focus on what I\u2019m writing and not how it will appear on the page.<\/p>\n\n\n\n

So far, I didn\u2019t need to change how I used Linux at the command line to do my work. I could still use vim<\/strong> to write in Markdown, and pandoc<\/strong> let me convert from Markdown to something else. The only change was that I could only work on one thing at a time\u2014there\u2019s no switching between \u201cwindows\u201d when you\u2019re working entirely at the command line.<\/p>\n\n\n\n

I also used the experiment as an excuse to format documents<\/a> using groff<\/strong> with the ms<\/code> macros. But that\u2019s where I ran into trouble. While groff<\/strong> is a superset of the old nroff<\/strong> and troff<\/strong> formatters, and I could run nroff<\/strong> to format documents as text to view on my terminal, viewing a PDF generated from groff<\/strong> seemed like a major obstacle for doing my work at the console.<\/p>\n\n\n\n

Using the framebuffer<\/h2>\n\n\n\n

Fortunately, that wasn\u2019t the end of my experiment. When booting Linux on real hardware connected to a display, you\u2019re actually at the console<\/em>, not a terminal<\/em>. The difference is that a terminal (or terminal emulator, if you\u2019re doing it from a window on your graphical desktop) only provides basic access to the display. You\u2019re going through an emulated<\/em> layer.<\/p>\n\n\n\n

However, the console is different. With the console, you can directly leverage the display hardware on the system. This is usually done through the frame buffer<\/em>, which is the same way that your Linux desktop puts the system into graphics mode. Except when you\u2019re running entirely in console mode, you can use other tools that take advantage of the frame buffer.<\/p>\n\n\n\n

One program is fbi<\/strong>, which is a f<\/strong>rame b<\/strong>uffer i<\/strong>mage viewer, available at kraxel\u2019s GitHub<\/a>. This is actually a collection of programs that can use the frame buffer; displaying images is just one feature.<\/p>\n\n\n\n