2018-12 Update
While I liked the idea of having a decentralized way of commenting, there are a few obvious issues with this system:
- Serving a third party script is bad, be it disqus, and the shiny new thing.
- Web Annotations clearly didn’t take off :P.
Thus, I chose not to continue the experiment when my blog moved to it’s new home.
While updating my blog I was looking for a way to make it somewhat interactive. Disqus is the standard option nowadays, but there are some concerns ….
Here’s a really nice post summarizing the issues with disqus. Here’s an exerpt:
- Load-time goes from 6 seconds to 2 seconds.
- There are 105 network requests vs. 16.
- There are a lot of non-relevant requests going through to networks that will be tracking your movements.
I encourage you to read it, it’s an eye opener. Also the suggested solution to use github comments on pull request for blog post comments is pure genius.
If you’re not an adtech paranoid, the load time alone should be enough to convince you.
Another point is the interactive aspect. Comment boxes are boring, most of the time people want to react/discuss a specific topic of an article. Having that discussion take place at the location where that topic is mentioned is way more efficient.
Introducing Web annotations
The W3C has come up with a standard for annotating web documents. The basic idea is to empower the user to discuss documents without actually modifying the document itself; you can comment a website even if the website does not have a commenting system. For those interested, you can read the draft here.
Here is a diagram of the general idea: click for fullscreen view
Hypothes.is is an attempt to implement what is described in the standard. For the paranoids out there, it does try to load a Google Analytics tracker, but if you have an ad blocker you should be covered.
The UI is pretty straightforward and non-intrusive. It offers in-line comments and highlighting. A perfect fit. We’ll see how it fares down the road.