My brother, The Libre Hacker, is now running his own Gemini server. For those unfamiliar with Gemini, it is a protocol used for non-web browsing. Or, from the Wikipedia article:
The Gemini protocol is an application layer protocol for distributed hypertext information systems that provides access to simple, primarily textual documents in Gemini space.
https://en.wikipedia.org/wiki/Gemini_(protocol)
Essentially, it is like Gopher but with TLS. It takes a very minimal approach to web browsing and presents pages much as you would have seen them in the 1990’s. It’s pleasant in that it doesn’t bog down with CSS or javascript, but it does lack the finesse we are used to in the modern web experience.
Anyways, the interesting part is finding a way to browse it. You can’t use a regular web browser to browse the non-web. This turned out to be quickly resolved as there are several Gemini browsers available for Android, so I installed Ariane, which worked perfectly on the first try. However, I couldn’t find any Ubuntu Gemini browsers in the repository, and had to build one from source code. I tried building several, and if you are using Ubuntu 20.04, by far the easiest to build and to use was Castor. The read me came with Ubuntu build instructions, and they worked to a ‘t’. It went like this:
$ sudo apt install build-essential rustc cargo libgtk-3-dev libgdk-pixbuf2.0-dev libssl-dev -y
$ cd ..
$ mkdir castor
< Downloaded Castor 0.8.16 from their website and extracted into castor folder >
$ cd castor
$ cd castor-0.8.16/
$ ls
$ make
$ sudo make install
$ castorAnd that was it. In less than 5 minutes I was browsing the non-web! I spent about 15 to 20 minutes monkeying with Lagrange and another 10 minutes with Kristall. I’m sure they are good browsers that work great, but Castor was by far the easiest for me to compile and set up.
In case you give it a try, you can go to the Gemini discovery page to find search tools, index’s, and more.
gemini://discovery.geminiprotocol.com/Linux – keep it simple.
To clarify, the text/gemini document (our version of HTML) is designed purposely to present only the raw text content and organizational markup, so as to not let anything get in the way of presentation of the content and structure of the document. Other media can be linked to a text/gemini document (for example, an image) but media cannot be embedded, properly speaking, and the document cannot control how the content is present. This prevents tricks, e.g., like blocking access to information with a wall of JavaScript, or crowding out a piece of site content with a dozen ads.
The Gemini protocol (our version of HTTPS) is designed specifically not to use (and to make it difficult to add in) those sorts of headers that are used to violate your identity and privacy (for example, there is no user-agent field).
Good point, thanks for clarifying!