Recently, Ton Zijlstra wrote about the concept of “federated bookshelves”, in which he referenced a post by Tom Critchlow on a similar concept called “Library JSON”. I have played with this idea based on Ton’s proof of concept. I also read through the postings flowing from Tom Crichlow’s post in 2020, and the ones from Ton Zijlstra’s post this year, and thought it would be helpful to provide a chronology of the development of this idea. To this end, I have created a Github repo with a chronology and links to tools and other distributed book info concepts that I have come across in reading on this topic. Pull requests and other comments are welcome! I am planning to spend more time on this in the next few weeks, so hopefully more prototypes and ideas to come….
Software Development
There are 96 posts filed in Software Development (this is page 5 of 10).
Federated Bookshelves Update #2
After reviewing a Selfoss issue on Github which described the behavior I saw in my first post on this topic:
This depends on the mime-detection on your server. You need to map .opml to the right mime-type
application/xmlin your webserver.
and this MDN page on updating mime types:
- If you’re using the Apache web server, check the Media Types and Character Encodings section of Apache Configuration: .htaccess for examples of different document types and their corresponding MIME types.
I decided to look at my own web hosting to see if I could find and modify the .htaccess file. For Bluehost, there is an .htaccess file within the public_html folder. I thought I would start with this one and see how things went.
Looking at this example configuration file for media types from HTML5 Boilerplate project, I saw this:
AddType application/xml xml
I then added this line to my .htaccess file:
AddType application/xml opml
After uploading that addition, my OPML file renders the same way that Ton’s file does – success!
Federated Bookshelves Update #1
Ton gave me a suggestion to refer to the XSL file on my local domain. I did this, but still only saw the raw OPML appear. After some thinking, I thought about MIME types, and wondered if this was set up on my web hosting to serve OPML as XML. I uploaded a copy of the file with a .xml extension, and everything worked! I have found some resources for updating MIME types, will post later after some experimenting.
Getting started with Federated Bookshelves
Some time ago, I read a post from Ton Zijlstra on the concept of federated bookshelves. His original post was riffing on an idea by Tom Critchlow called Library-JSON. Recently, Ton revisited this topic, and in a short period of time created a working prototype OPML file and a XSL stylesheet to view the OPML file.
I finally had some time this weekend to play around with this. Here is my work so far:
- My first booklist in OPML format
- The transformed file
I had some problems using Ton’s XSL file. On my server, when I entered the URL of the OPML file, all that was displayed was the OPML file. However, I found a tool to apply the XSL transformation, and the resulting file was a valid HTML page (the “transformed file” above). I will continue to investigate this and also reach out to Ton to see if I did something wrong.
I like this – it drives home the need for planning to ensure that projects can be completed successfully.
Free book on Node.js programming – over 400 pages!
Enjoyed this reference to “Why you need a programming checklist” from Andrew Shell…
Some things just keep on movin’….
Link to David Brin article