Likes, upvotes, replies, friending. What if it’s all just linking? In fact, what if linking is actually more meaningful!

They can be dated.

Now let’s turn to categories. A small directory doesn’t need a full-blown hierachy—the hierarchy shouldn’t dwarf the collection.

A structure for a flat directory (in one file or one file per entry) can be:

---
Link Title
url://something/something
*topic/subtopic format time-depth
Markdown-formatted *description* goes here.

• topic/subtopic is a two-level ad-hoc categorization similar to a tag. A blog may cover multiple categories, but I’m not sure if I’ll tackle that. I’m actually thinking this answers the question, “Why do I visit this site? What is it giving me?” So a category might be supernatural/ghosts if I go there to get my fix of ghosts; or, it could be writing/essays for a blog I visit to get a variety of longform. An asterisk would indicate that the blog is a current feature among this topic (and this designation will change periodically.)
• format could be: ‘blog’, ‘podcast’, ‘homepage’, a single ‘pdf’ or ‘image’, etc.
• time-depth indicates the length one can expect to spend at this link. It could be an image that only requires a single second. It could be a decade worth of blog entries that is practically limitless.

The other items: author, url and description—these are simply metadata that would be collected.
The directory would then allow discovery by any of these angles. You could go down by topic or you could view by ‘time depth’.