GSU on FamilySearch

TheCount, when records go through all these iterations, citation does get complicated.  When I first read through your citation and compared it to the material filmed at the link, I made these alterations:

        1. Oswego County, N.Y., "Marriage Records, Vol. 1, years 1908-1913" [as identified on filmer's target; volume label not photographed], p. 298, entry no. 1489, Anderson-Gardner, married 15 October 1910; imaged in "New York, County Marriages, 1847-1848; 1908-1936," FamilySearch (https://familysearch.org/ark:/61903/3:1:S3HT-6FX9-7V3? : viewed 22 February 2021) image 170; imaged from GSU microfilm 872941, citing "Oswego County Courthouse Clerks Office,” 6 August 1971.

Explanations:

1. If we copy what is on the target and put quotation marks, without saying that we are quoting the target, users of our citation (and we, at a later date after our recollection has gone cold), will assume that the register itself carries that exact title—but we don’t know that. It’s better to be specific here and say that we are taking the wording from the target.
2. Your second layer said “imaged at ‘New York, County Marriages, …,’ FamilySearch …”.  EE would say that it is imaged in a database that is at a website.
3. The number of layers separated by semicolons can be reduced if GSU’s statement of provenance,  i.e., “Oswego County Courthouse Clerks Office” is included in the source-of-the-source layer.
4. In that source-of-the-source layer, we also might note that it comes from GSU microfilm made in 1971 rather than being a digital image that FamilySearch did in, say, 2019. Noting that point is an alert that the filming quality is not as good as what it would be today.

Then, after I worked through these issues, I noticed something else. Your citation to the online image follow this form:

“Database,” Website (URL : date) > image number.

However, at FamilySearch, a database is almost always subdivided and there will be many parts that carry the same image number. The FamilySearch header to the image shows a path (identifying those parts and sub-parts) that is not included in the formulation above. When we add that path, we get this:

         1. Oswego County, N.Y., "Marriage Records, Vol. 1, years 1908-1913" [as identified on filmer's target; volume label not photographed], p. 298, entry no. 1489, Anderson-Gardner, married 15 October 1910; imaged in "New York, County Marriages, 1847-1848; 1908-1936," FamilySearch (https://familysearch.org/ark:/61903/3:1:S3HT-6FX9-7V3? : viewed 22 February 2021) > Oswego > Marriage records, 1908–13, vol. 1 > image 170; imaged from GSU microfilm 872941, citing "Oswego County Courthouse Clerks Office,” 6 August 1971.

This, obviously, creates redundancy. The path unnecessarily duplicates the target data we added in layer 1.

All things considered, EE would make the FamilySearch database the master source, which would shorten the citation considerably and simplify it as well. The following reduces the word count from 70 to 47 and pares the citation down to just two layers.

         1. "New York, County Marriages, 1847–1848; 1908–1936," FamilySearch (https://familysearch.org/ark:/61903/3:1:S3HT-6FX9-7V3? : viewed 22 February 2021) > Oswego > Marriage records, 1908–13, vol. 1 > image 170, entry 1489, Anderson-Gardner, married 15 October 1910; imaged from GSU microfilm 872941, citing "Oswego County Courthouse Clerks Office,” 6 August 1971.

TheCount, sometimes it helps to try it both ways and see which is the simplest.