Remove Not Similar Channels, but Repetitive Chains of Origin
Reduce the flow based on the origin of facts, while preserving diversity through independent grounds and methods of interpretation.
Two retellings of the same press release are not two independent sources. However, two analyses of that release may be necessary if they help reveal different limitations, consequences, or objections. Therefore, you should reduce not similar channels, but repetitive chains of information origin.
An information bubble does not arise from a small number of subscriptions per se. It arises when all remaining sources obtain facts from the same place and explain them in the same way. For a review, two separate metrics are needed: where the fact comes from and what new information the publication adds.
Conduct a Review Based on Chains of Origin
1. Choose one topic that is currently on your work agenda. Do not try to analyze the entire feed at once. For each similar publication, record three things: the original source of the fact, the role of the publication, and its unique addition.
The original source may be an initial document, statement, research paper, or message from a participant in the event. If a channel links to another post, which in turn links to a press release, the chain does not start with the channel. If the origin is not specified, do not consider such a post an independent confirmation until you find where the assertion came from.
The role is also important. One material may report the original fact, another may quickly retell it, and a third may analyze the consequences. Do not compare them solely based on matching phrasing: they may have different objectives.
2. Collapse fact duplicates within the same chain. If you need to establish exactly what was announced or published, keep the primary material. When it is inconvenient to work with the primary source, keep one intermediary that clearly leads to the original and does not distort the subject of the message. Other retellings of the same news do not create additional verification: they only increase the number of notifications.
Do not remove a publication simply because it relies on the same fact. First, answer the question: what does it contain besides a retelling? A valid reason to stay could be an analysis of assumptions, comparison with other primary materials, professional objection, or another evaluation criterion. The phrase “this is important” without new grounds is a sign of duplication.
3. Select alternatives not by channel name, but by independent line of reasoning. Two authors discussing the same original document do not confirm the event for each other. However, they can provide different viewpoints if one relies on additional materials while the other checks the conclusion from a different practical angle.
After the review, each remaining subscription should have a clear function: it either leads to the original fact or provides a separate interpretation with its own grounds. If the function cannot be stated in one sentence, the source has not yet earned a place in your working set.
This approach aligns with recommendations from the UBC Library guide: seek the best coverage of a topic, verify statements in other sources, and consider the material from different perspectives. The National Library of Australia separately emphasizes the value of different perspectives when comparing them with primary materials.
The result of the review is not a minimal number of channels. It is a set in which repetitive messages no longer masquerade as multiple confirmations, while different viewpoints remain visible and verifiable.