🔗 Why Feeds Fun normalizes tags — and how

Read this post if

• You want to use Feeds Fun rules more efficiently.
• You're looking for a deeper understanding of how tags work in Feeds Fun.
• You self-host Feeds Fun and want to configure tag normalization.
• You are interested in tagging and approaches to it.

Tags are the real source of power in Feeds Fun. They allow you to rank, filter, and sort news how you want, saving your time and energy.

However, tagging is not as simple as it seems. There is no strict definition of what a tag is, nor is there a standard for naming them.

We may agree that a tag is a short string that describes the text or part of it. But…

• How short should it be? Is this-text-is-about-modern-politics-in-central-asia a tag or not?
• What about nowadays-politics and modern-politics — are they different tags or forms of the same tag?
• Which one of the tags book-review, book-reviews, books-review, books-reviews should we use?

There are numerous questions like that, and no clear answers.

The situation is even more complicated because Feeds Fun uses multiple independent sources of tags, called "tag processors", each of which may have its own rules and conventions for tags. Some tags even come from the feed itself and, therefore, may take any form. You can find more about tag processors in our README.