In writing about numerous minor items on Wikipedia, it is preferable to start combining them into list articles rather than creating thousands of tiny stub pages.

Although it is great to have stubs to capture every notable river, valley or town, those stubs alone will risk the danger of "Cannot see the forest for the trees" as too many tiny articles which fail to describe the whole region. Meanwhile, there are list-of-area articles which can be translated instead, such as from German Wikipedia. This is the concept of "List of tiny stuff" where not all items in a list have separate articles. For example, a river valley in Germany, such as article "de:Zschonergrund", is a small part of a list of 147 river valleys in Saxony (German: Sachsen), as one list of the 16 Bundesländer (federal states) in Germany:

• de:Liste der Landschaftsschutzgebiete in Sachsen - one of 16 lists of valleys

Before creating hundreds of stubs for valleys, start with the 16 list articles (for 16 states in Germany), including tables of valleys showing the location and area (in ha/acres) for each entry. Even in the German WP, a list of 147 valleys has red-links for dozens of valleys (no articles yet), and perhaps some valleys are so minor that there are no sources which focus on a tiny valley as notable in "continued coverage". Let's avoid stubs for every tiny thing listed in a book of geographic areas.

Remember that WP's current 2,000+ asteroid-list articles began at Harvard University (Boston) as only 37 large data-files, listing thousands of obscure asteroid numbers in each of the 37 lists. A notable, yet obscure, asteroid number can redirect into a list of related minor planets where we figured 200 lists of numbered asteroids was "efficient" and 2,000 lists were perhaps too many and too short. Of course, there are over 5 thousand more notable asteroids, which have separate articles, but the other 200,000 numbered asteroids are described in the combined list-of-asteroids articles.

Similarly, if the rivers in each small parish, county or district are placed in a list, this makes it much easier to cross-check their length, flow, depth, etc. – compared to there being 100 stub pages, each needing to be read or edited to check or add this data. WP's unfortunate hatred of lists in early years thwarted the reality that lists are in fact preferable. It is perhaps 100× times easier to cross-check and update a list of 100 items than edit 100 individual articles.

So, we can consider the initial genius concept behind Harvard's 37 lists of numbered asteroids, with similar lists of hundreds of smaller items (+data columns), and then redirect common titles into those lists, rather than start with 999,000 stubs to be given data in the next decade. In many cases, an item is so obscure, it should be searched by name (to be found within a list article), rather than expect a stub's title, or even a redirect title to pinpoint to an article about a minor river, hill, or other item. Aggregate the minor data into lists, rather than creating so many tiny stubs.

[ This essay is a quick draft, to be expanded later. ]