More conceptually, this may be of interest to you and your premise.

At a certain level, artifacts and historical records are a source of bias. Early cultures are often referred to by their pottery style because that's the sort of artifact that survived best; there are many sources that are untranslated or simply not available easily in other languages; military and espionage intelligence is classified until decades after the fact.

Additionally, historians have to decide what information is relevant to their topic and what isn't. A perfectly unbiased source would have to answer any mundane questions like "what did this Egyptian peasant eat for breakfast on this day?" or else be unable to apply to every facet of history. These decisions are important to shaping the narrative a historian wants to discuss, so an unbiased source would most likely read as a list of events with no attempt to discuss them besides discussions already in the record, which would themselves be presented as equally valid points of data.

As an example, and without directly referencing any events within the 20 year rule, you can compare international coverage of events related to political issues, such as a forest fire. Often one side of the aisle will refer to issues that support their biases such as local governments' responses while the other will stick to dry statistics that a local reporter would be able to fill with local perspectives.