BazQux Reader Community

Examples of using comments with bazqux sought

I use bazqux every day as an RSS reader but I don’t understand what “comments” related features are about.

I just now tried searching all my feeds for comment:true and the results were empty.

The RSS 2.0 Specification has a section about comments which provides for any RSS item to include a link to another feed holding comments on the item. It points out Weblog Comments and News Aggregators as providing good details on this feature, which in turn points to this blog as being one supporting and having active RSS 2.0 comments (such as appear on this article. So, to test out my understanding, I subscribed to its posts feed and tried again searching for comment:true, only to again find empty results. (edit: now I see perhaps this is because Blog comments disabled??)

Clearly I need an education, but I can’t find any documentation regarding what bazqux considers to be “comments” and how it works with them.

Am I missing it?

Is there some good example of how comments are supposed to work with bazqux?

For example, I use bazqux to follow new articles posted to hacker news, but as far as I can tell, there is no way to use bazqux to to stay abreast of new comments on selected articles. Is this possible somehow with the comments capability?

BazQux supports:

  • RSS/Atom comments where they’re enabled (some Wordpress/Blogspot blogs).
  • Reddit comments (via Reddit API)
  • Disqus comments (via Disqus API)
  • comments (page scrapping + json)
  • comments (page scrapping)

HN comments were supported some time ago, but HN enabled very strict rate limits, so it became impossible to update comments at a reasonable rate, and HN comments were disabled.

The site you mentioned has a few comments. comment:true will only show comments in the expanded view (first view mode, keyboard shortcut ‘1’).

If you want to track comments, there are a few options:

  • Expand the comments on the article that has them (click the ‘comments’ button at the bottom of the post if it exists; or press ‘o’). This will allow you to read current comments, but won’t allow you to check for new ones.
  • Use the expanded view (the first view mode). This will show all comments (may be too much for popular sites, but may be good for less popular ones). And it will show new unread comments, just as it does with posts.
  • Tag the post you want to monitor comments on, and set tag’s view mode to expanded. This way you can keep the blog showing only posts (any view mode except first) while monitoring comments for specific posts. This is very non-obvious and could be implemented better.

There’s a pretty interesting story behind comments.

The ability to read comments (to follow functional programming discussions in the Russian segment of LiveJournal and Blogspot back in the day) was the main idea of the reader when I started developing it.

Updating comments means that the reader doesn’t just have to update the feed and that’s it, but also fetch all the posts (they may have comment feeds that aren’t present in the blog feed) and refetch the posts’ comments occasionally. This means a much larger number of fetches and a lot of work to distribute them evenly so as not to get banned. It also means that the reader needs to be able to quickly display dynamic comment trees and mark them as read. And it means that the display should be optimised for one- or two-line messages.

So, as a side effect of solving the problems with comments, I got a reader that:

  • updates feeds fast and is not blocked by most sites, even if you subscribe to many feeds at once (which helped a lot during the GoogleReaderApocalypse)
  • is very fast (if it can display hundreds of comments, it can certainly handle a handful of posts)
  • has a very compact interface.

Turns out this is what people actually like.

The first paying customer asked me to disable comments (they were always expanded at first). A bit shocked, I asked “then why did you buy it? comments are kinda the main idea” and he said “because your reader is much faster than Google’s”.

So I shifted my focus from comments (though I still use them for Reddit) to making the reader even faster and more streamlined.


Excellent and thorough response. Thanks very much. This would likely make a good blog article (or FAQ entry, or section of documentation).