One of the longstanding issues in any sort of historical discussion is sourcing. In addition to the usual OSM issues relating to ownership of data (and compatibility with the ODbL), historical mapping has sourcing issues similar to any historical work. This page is intended to provide an introduction and a guide to "doing it right".
A primary source is typically a first hand account by a participant or direct observer of an event, or other formal or official record made at the time. It may be found in official documents (e.g., The War of the Rebellion: a Compilation of the Official Records of the Union and Confederate Armies), or in private letters and diaries written at the time. Delays in writing can cause sources to be secondary rather than primary, as memories may fade or mistakes creep in, or other participants die allowing the survivors to cast blame on the dead.
Secondary sources are generally created after the fact. For example, the large body of writings by participants in the American Civil War published in the 1870s through the 1890s, while valuable, are too removed from the actual events to be considered 100% reliable unless other corroborating evidence can be found. Care must be taken to insure that apparently corroborating sources are not actually citing each other or citing a shared secondary source.
Secondary sources can still be quite valuable, though. The works produced by and in collaboration with Ezra Carman on the Battle of Antietam provide substantial detail on the battle not available anywhere else.This includes both Carman's manuscript on the battle and the maps which were made by E.B. Cope.
Secondary sources which were assembled by consulting previously published secondary sources are substantially less useful, however.