Wikipedia:Reliability of GNIS data/Robert M. Rennick Manuscript Collection

One source of GNIS data is the Robert M. Rennick Manuscript Collection at Morehead State University, with the following annotation in the citation field of the record: "Rennick, Robert M. Geographic Names Field Research . 31-Dec-1973. Conducted over a span of approximately 40 years, and annotated on USGS 1:24,000 scale topographic maps which date from 1938 to 1973. This data was also used in several published works by the author that were published during and after the field research was conducted."

Source
Robert Morris Rennick (1932-06-28 – 2010-08-05) was a retired professor, member of the American Name Society, chairman of the Kentucky Committee on Geographic Names, and member of the GNIS Phase II committee; who did a lot of primary research himself. As an employee of the Kentucky Division of Licensing and Regulations from 1972 until 1997 he travelled over many parts of the state collecting information on names as he went. Xyr manuscript collection and books at the University comprise six major things as far as Wikipedia needs are concerned:
 * a set of annotated maps, marked up by Rennick xyrself, incomplete list:
 * a place-name card collection (sometimes with more than 1 entry for a given place documenting different primary sources)
 * county-by-county reports on place names for all of the counties digitized from the card collection, incomplete list:
 * A to Z alphabetized reports of place names digitized from the card collection, incomplete list:
 * a bunch of recorded oral histories (primary sources)
 * around 50 books on county post offices and county histories, based upon the primary source maps and oral histories with explanatory footnotes, incomplete list:
 * A to Z alphabetized reports of place names digitized from the card collection, incomplete list:
 * a bunch of recorded oral histories (primary sources)
 * around 50 books on county post offices and county histories, based upon the primary source maps and oral histories with explanatory footnotes, incomplete list:
 * a bunch of recorded oral histories (primary sources)
 * around 50 books on county post offices and county histories, based upon the primary source maps and oral histories with explanatory footnotes, incomplete list:
 * a bunch of recorded oral histories (primary sources)
 * around 50 books on county post offices and county histories, based upon the primary source maps and oral histories with explanatory footnotes, incomplete list:
 * around 50 books on county post offices and county histories, based upon the primary source maps and oral histories with explanatory footnotes, incomplete list:

Morehead State University provides the manuscript collection on-line with a full index at https://scholarworks.moreheadstate.edu/rennick_ms_collection/. The post offices and county histories books are in its Federal Writers' Project collection at https://scholarworks.moreheadstate.edu/kentucky_county_histories/. The maps are indexed by USGS quardangle at https://scholarworks.moreheadstate.edu/rennick_maps_all/. The oral histories are at https://scholarworks.moreheadstate.edu/rennick_oh_collection/.

You might also come across xyr books through a Google Books search for a GNIS record name, albeit that the manuscript collection comprises a lot more than what went into the book. Here is an incomplete list:

Example entries
The seventh entry, out of eight, for Eriline/Erline post office in Clay County, Kentucky in reads:

As can be seen, all names are underlined and post offices and suchlike are identified as such. Sources are given at the ends of entries. This is of particular note in the entry for Catts Fork of Blaine Creek, Lawrence County, Kentucky, in, which follows two entries for "Cat" and two for "Cat Creek" giving tales about wildcats. The Catts Fork of Blaine Creek entry in turn gives a lengthy story (elided here for brevity) about a William S. Catt (quotation style changed from typewriter for clarity):

In several such cases, such dubious information is called out in another work. For example: then states that "Seven post offices were located on Cat Creek, the 12&frac12; mile long stream that joins Blaine one mile below Fallsburg. Neither of the two traditional accounts of the derivation of this stream's name have been confirmed. It was either named for area wild cats (more likely panthers) or for a Virginia salt dealer named William S. Catt."

Problems
Unfortunately, little to no care has been taken in transcribing this into the GNIS, despite the attention to detail in the original. For examples:
 * Even though Rennick calls the account unconfirmed, it is clearly folkloric, and the account itself calls the place a "rockhouse" (i.e. a cave, pretty much), the GNIS has a "Populated Place" record for . Hence the mass GNIS importers gave us an "unincorporated community" article for a cave that someone camped in in a folk tale.
 * Because it was an underlined name, the GNIS has a "Populated Place" record for listing "Big Creek" as the relevant map.  (Interestingly, it is not hand-marked on .)  Hence the mass GNIS importers gave us an "unincorporated community" article for a (clearly not a community) creek mouth that is, even according to Rennick's own "RED BIRD RIVER" entry in, across from the the erstwhile site of Red Bird River Petroglyphs and the murder of Red Bird on the Red Bird River.