When to layer a database with images

So, reading over the QuickLesson 25: ARKs, PALs, Paths & Waypoints (Citing Online Providers of Digital Images), I got a little bit confused on when to layer a citation.   Two of the examples are:

Oklahoma County Marriage Records,” database with images, FamilySearch (https://familysearch.org/ark:/61903/3:1:9Q97-Y3Q7-Z6Y?owc=waypoints&cc=1709399 : accessed 1 January 1017) > 1313685 (004532716) > image 479 of 711.

and

“Alien Crew List,” S.S. Arrino (Lota, Chile, to San Francisco), arriving 7 October 1913, p. 11, C. S. Dendy; imaged in “California Passenger and Crew Lists, 1882–1959,” database with images, Ancestry(http://www.ancestry.com : accessed 1 January 2017) > M1416 – San Francisco, 1905–1954 > 012 > image 656 of 1016.

Why is the second layered when the first is not?  I thought when we were accessing images like this we cite the digital image, not the original, like on page 166 in the Corporate Records Online Images QuickCheck.

Don't get me wrong, I kind of prefer the second (although that would mean I need to go back and change a TON of citations), because that way if a provider goes out of business, my citations are still useful, but I'm so confused....

Submitted byrraymondon Tue, 01/02/2018 - 15:59

Cynwrigte,

FamilySearch citations are suggestions only and for various reasons are not always complete. I would recommend that you add the additional layer for the original document. As for the order of citation layers, FamilySearch always cites the online item first, for consistency and because of limitations in our system. You are welcome to change this as you desire in your own records.

Robert Raymond
FamilySearch

Submitted byrraymondon Tue, 01/02/2018 - 16:08

In reply to by rraymond

Oops. It looks like I wasn't clear. I should say, "Citations on the FamilySearch website are suggestions..." I didn't intend to comment on Elizabeth's QuickLesson.

Submitted byEEon Wed, 01/03/2018 - 19:07

Good question, Cynwrgte!  This goes back to the decades-old debate between database users who considered themselves to be "lumpers"or "splitters."

Some researchers—when they use many different items from a single database (or any other source), prefer to focus the citation upon the larger item (the database, the record collection, etc.) and then add each individual record's details in the "specific item" field. Others prefer to start the citation with the specific record and then, in a second layer, identify the larger item—in this case the database and provider that published the database-images.

To use the first database above as an example, if you pull 20 or 30 different records from that database and you cite each one individually using the second format above, then you would have 20 or 30 different "master sources" in your Source List.  By using the database as the lead element, you would have only one "master source" to deal with.

EE illustrates both approaches—both in the manual and at this website.  If you have a digital edition of EE, query for the term "lead element" and you'll find dozens of examples demonstrating the differences that result when you focus your citation on one element rather than another.

Submitted byRobyn_62on Wed, 01/03/2018 - 22:32

I still don't really understand the difference between a "splitter" and a "lumper." But maybe what I do in my reference notes is perhaps over the top?

Churchill Parish (Oxfordshire, England), "Register of Baptisms in the Parish of Churchill in the County of Oxford [1857 to 1904]," p. 6, entry no. 46, James Albert Penson, baptised 22 May 1859; digital images, "Oxfordshire, England, Church of England Births and Baptisms, 1813-1915,"  Ancestry  (http://search.ancestry.com.au/search/db.aspx?dbid=61057 : accessed 1 November 2016) > Churchill > 1857-1904 > image 11 of 61; citing "Reference Number: PAR66/1/R2/2," Oxfordshire History Centre, Oxford, England.

 

 

Robyn62,

I could be explaining this badly, but a splitter vs. a lumper has more to do with how your citations are organized in your software vs. how your reference note actually looks.  For example...and these are extremes...a "lumper" may have one master source for the 1940 U.S. Census and all "details" about the individual or household are in the "details" part of the citation.  A "splitter" on the other hand may have hundreds (or more) master sources for the 1940 U.S. Census because they might have a master source for each enumeration district, or in some cases, for each page of the census they access. So they have more data in the "master" part of the citation.  Others of us (well, me anyway) are somewhere in the middle. I have a master source in my database for each county...so I have one master source for "1930 U.S. Census - Ohio - Franklin County" and another master source for "1930 U.S. Census - Ohio - Pickaway County". 

Submitted byRobyn_62on Thu, 01/11/2018 - 00:07

In reply to by niteowl1851

Dear niteowl1851,

Your reply makes perfect sense, and thank you for taking the time to explain this to me.

Like you, I think I am in the "middle".

All the best,

Robyn

Submitted byniteowl1851on Thu, 01/04/2018 - 09:27

Cynwrigte,

I know it is mentioned in a couple places, but one place where the option of citing the original record vs. the image is discussed is pg 433 (3rd ed.), which is chapter 9. "Whether you cite first the record or the publication is a situational choice--one usually guided by (a) whether you have just one or many images to cite from the medium and (b) whether the source of the image is fully identifiable."

For two personal examples, I have multiple citations for the "Ohio Deaths, 1908-1953" database on FamilySearch. So, for citations from that database, I lead with the database rather than the "death certificate". 

Alternately, I also use "Ohio, County Marriages, 1789-2013" database at FamilySearch. However, because marriages are recorded at a county level, and because I want those entries broken out by county in my software, I lead with the original record instead of the database. So as an example, one starts like:  Hancock County, Ohio, Marriage Record, vol. 4:511, groom-bride, date; digital image, "Ohio, County Marriages, 1789-2013," FamilySearch (url : accessed date).

So, layering allows some flexibility while still capturing the necessary information.

Submitted byc0r8g30on Thu, 01/25/2018 - 21:16

Going back to the example Cynwrigte posted in the inital inquiry:

“Alien Crew List,” S.S. Arrino (Lota, Chile, to San Francisco), arriving 7 October 1913, p. 11, C. S. Dendy; imaged in “California Passenger and Crew Lists, 1882–1959,” database with images, Ancestry(http://www.ancestry.com : accessed 1 January 2017) > M1416 – San Francisco, 1905–1954 > 012 > image 656 of 1016.

Should we add do the research and add a third layer, that of

; citing National Archives Microfilm Publicatiom M1416, Crew Lists of Vessels Arriving at San Francisco, California, 12/28/1905 - 10/30/1954, Microfilm Roll #12.

c0r8g30, "do the research" is always a great motto. But, if Ancestry does not actually cite "National Archives Microfilm Publication M1416, Crew Lists of Vessels Arriving at San Francisco, California, 12/28/1905 - 10/30/1954, Microfilm Roll #12," then our citation to what we saw at Ancestry would not say "citing ....."  If Ancestry did not cite its own source, then Ancestry is not citing anything. If we did additional research in an effort to identify the exact microfilm publication, its title, and specific roll--and all those details come from our own research--then it would be appropriate to add a sentence to the citation to say

Ancestry appears to have imaged "National Archives Microfilm Publication M1416, Crew Lists of Vessels Arriving at San Francisco, California, 12/28/1905 - 10/30/1954, Microfilm Roll #12."

Sometimes, the source from which a provider has taken an image appears to be obvious. Sometimes, when we jump on what seems to be obvious and assume that's what was used, we end up with egg on our face because there can be two separate sets of records with small but significant differences and we jumped on the wrong one. Ergo, we always need to be explicit as to whether the provider actually cites XYZ or whether it's our assumption that the provider used XYZ.