Best practices for managing sources, citations and media files?

RootsMagic
1

Hi everyone -

The more I get familiar with RootsMagic, the more I realize how much I don’t know. One of the things that has caused me the most rework is first causing data problems in my sources, citations and media files, then spending hours fixing them. That’s a long story that I won’t burden you with, but here’s my question:

For those of you who have been using RM a long time (or even a short time and it’s working great for you), what works best for you in how you set up your sources, citations and media files? For example, do you group up your sources (i.e. 1940 census) and then have a bunch of citations (John Smith in the 1940 census)? Or do you have one citation per source? How about media files? How you do handle those? At the source level or the citation level?

Please hit me up with your suggestions for what you have found works well for you. I’d love to get some suggestions so I get this right the next time and don’t have to keep spending time reworking them. Suggestions of what not to do are also very welcome. I already learned that you don’t use the automated merge function unless you make sure that your citations are unique. That’s what I’m fixing now.

Thanks in advance for educating me!

P.S. I have read the help files in this area and watched all their YouTube videos. I also looked through the posts in the Community. I have a copy of “Evidence Explained” on its way. I’m on RM10 and have about 1000 people in my database.

1 Like
2

You might get a wide range of answers. This also gets into how one desires lump vs split. (Which is where / how the info is split).

I will be rebuilding sources/citations later this year. This response with be high level without getting into the details of how to execute the different variations. My first thought is to do:

Census by location (Avoiding the topic about how places may changed over 150 or so year).
CENSUS Source by place “Boston, Suffolk, Massachusetts, United States”: (each location would be a separate source

  • In Citation have Year and other details (each decade would be its own citation)
  • in Citation Link – link to people & citations

This method allows you to see multiple decades grouped for that one location depending on how you execute.

Could do 1940 Census , 1930 Census etc (for Source)

  • In Citation have place and other details
  • in Citation Link – link to people & citations

Kevin

2 Likes
3

Before one makes important decisions about sourcing (lumping or splitting), I suggest asking yourself a few questions.

  1. What are the goals of your research and for whom is it intended?
  2. Do you want your documentation very specific or general just so that people can tell where you looked and delve in further for themselves? Making documentation simple leaves more time and energy for actual research. I believe documentation should serve the research.
  3. How do you want your printed reports to look?
  4. Does the intended recipient of your report want or need extreme specificity?

I have chosen to use general, short/concise sourcing such as 1850 US Federal Census, Find a Grave, Obituary, Birth certificate, etc. I intend the majority of people to access my information in online shared databases and written reports, that I choose to email to people or leave in a family family in a local library.

If the citations are too long and specific, they make some reports look quite cluttered and confused.

The primary reason I never plan to use RM10 as my primary database entry program and stick with RM7 is that the sourcing is simpler to use. I don’t try to force the program to write complicated source or citation descriptions. Creativity and unique sentences can be used in the general notes.

That being said I do not yet add media to my RM7 database. Family Photos I want to share I attach to my smaller Ancestry DNA database.

1 Like
4

Whoo boy is that an understatement! :laughing:

@MardeeB First, welcome!

I made a pretty detailed couple of comments about how I structure US & State censuses recently on this post, so I won’t re-hash that.

If I could offer a single Cliff Notes nugget of wisdom … decide on what you consider sources versus repositories. That decision will affect a lot.

I for example, consider Ancestry, Family Search, MyHeritage, etc to be repositories for 95% of their “records”. They didn’t originally create the document / record … they only store it like a library or warehouse. The exceptions have to do with specific indices they’ve actually created themselves. That viewpoint, in turn, means that any records / citations / “sources” I get from those gets re-worked.

1 Like
5

[quote=“KimberlyGreen, post:4, topic:13533”]
Whoo boy is that an understatement! :laughing:
[/quote] :grinning_face:

The other think to consider is where you was sending the RM info – via GEDCOM and treeshare for example. Using Source Template vs free form etc etc. IF only using in RM and for the near future then less to considered.

I would venture to share that in many cases for the same fact type – 7 out of 10 will do things differently from one another – and that is not necessarily wrong – but goes to show – there are many ways to do things.

6

Oh, that’s a very good idea. Right now, I have them all grouped by year but I will definitely put that on the list of things to consider.

Thank you for taking time to reply!

7

Those are great questions and I really appreciate you helping me think through it in this way. The reality is that I may be overthinking this and I’m definitely spending too much time on cleaning up source and citation data, which is taking away time from research. Funny enough, I swore when I retired last year that I wasn’t going to spend time on a computer unless I was absolutely forced to, and here I am updating hundreds of records.

I hadn’t even thought about the issue of reports. I need to run a few and see how that is impacted. I don’t know how much I will use reports other than for my own purposes because what I’ll be sharing with the family will probably be in a more narrative “story telling” type format.

Thank you again for your insights and please let me know if you have any other insights.

8

Thank you!

About repositories, that’s a good point. I think I was so hung up on figuring out how to structure sources/citations/media files that I blew right past the question about repositories. I agree with your thought process though.

And you’re right, I will get a wide range of responses, but that’s what I was hoping for - just some more things to think about so I can make better decisions moving forward.

I used to teach insurance people how to use their work systems, and while it’s great that the system is so flexible, that comes with a complexity that can be a challenge. Yes, I can do it any way I want, but which way do I want to do it and what are the repercussions of that?

I think I am definitely going to simplify the data that I’m tracking just so I don’t have to keep redoing the data to be consistent, and I’m probably also going to stop retrieving sources from Ancestry and FamilySearch. It’s really convenient but it creates such a mess.

Thanks for taking time to reply and please let me know if you think of anything else that might help. Thanks again!

Mardee

1 Like
9

I’m a sample size of one and I’m probably in the minority, but what finally clicked for me after years of struggling with this issue is that 1 image file = 1 source = 1 citation. Among other things, that makes me into a source splitter. And by the way, it seems to me that the 1 image file = 1 source = 1 citation model is pretty much the way it works if you do your work online in something like Ancestry or FamilySearch. So it’s not like it’s some radical concept that I just invented out of thin air.

My source list in RM is enormously long because I put all the sourcing data into the source of RM’ source and citation system. The citation part of RM’s source and citation system serves primarily just to produce the footnote sentence. My footnote sentences are much simpler than those from Evidence Explained or other authorities, but it seems to me that my footnotes still provide all the information required for me or anybody else to find the document again. Plus I do download all images or otherwise create them on my own, for example by making my own photographs for grave markers rather than by relying on Find-A-Grave. And I do link all of my images into the citations and I link some of my images into facts.

But my long source list is not a problem, especially in RM8/9/10 because I can start typing in the source list and find any source and citation very quickly. For example,
I can start typing something like the following into the search box in the source list

census, US, 1900, TN, Loudon County, ......

and almost immediately I can find a census source/citation that I’m looking for. As soon as get to 1900 in my typing, I have filtered to all U.S. censuses source/citations for 1900. As soon as get to TN in my typing, I have filtered down to all the 1900 census entries for Tennessee. Remember, that this is the source name. A full comparison of the source name with the footnote would be something like the following.

Source name: census, US, 1900, TN, Loudon County, Tennessee, Enumeration District 103, Dist. 9, page 84b

Footnote sentence: U.S. Federal Census: 1900, Loudon County, Tennessee, Enumeration District 103, Dist. 9, page 84b, ancestry.com (1900 U.S. Census) viewed on 5 January 2016.

As you can see, I cite to the census page, not to the census family as suggested by Evidence Explained. That means that sometimes several families on the same census page use the same census citation. That also means that a family that spans two census pages will have two citations, one for each citation. But that’s really no different that citing from a compiled family history where John Doe is found on page 197 and his brother William Doe is also found on page 197. You wouldn’t make two different citations, one for each brother. You would just have a single citation for page 197 and use it for both brothers.

Most of my other citations are organized by person or by couple rather than geographically. Here are some samples.

Source name: obit: Abner, Thurman Keith; Knoxville News-Sentinel [Knoxville, Tennessee], 3 May 1997

Footnote sentence: Obituary: Abner, Thurman Keith; Knoxville News-Sentinel [Knoxville, Tennessee], 3 May 1997, page C7, viewed at newspapers.com on 2 February 2022.

Source name: death certificate: Bryan, Harley; Certificate 229, Kodak, Sevier County, Tennessee, 1915

Footnote sentence: Death Certificate: Bryan, Harley; Certificate 229; Kodak, Sevier County, Tennessee, 1915; ancestry.com (Tennessee, Death Records, 1908-1958); viewed 20 January 2015.

Source name: marriage: Bryan, Emmert and Ruby Cox, Jefferson County, Tennessee 1921

Footnote sentence: Marriage Record: Jefferson County, Tennessee, 1921; bound marriage application book page 50; Emmit Bryant and Ruby Cox, ancestry.com (Tennessee State Marriages, 1780-2002), viewed 20 December 2014.

I do give myself a little flexibility on my rule for myself that 1 image file = 1 source = 1 citation. For example, a single courthouse marriage record may include more than one image, such as a Marriage Application and a Marriage License and a Marriage Return. In the real world, the License and the Return are usually the front and back of the same piece of paper. So use the three images collectively to create one citation and one source, and I do not use graphics software to combine the three images into a single image file.

My model is a very different model than the traditional source + citation model where the source might be the U.S. Federal Census as a whole, and the citation is every thing else. Or the source might be the U.S. Federal Census for 1900 as a whole, and the citation is everything else. Etc. In other words, you can put more of the sourcing data in the source and less in the citation, and then you can effectively reuse the the source data by using the same source with many different citations. I don’t reuse sources in that manner at all. But I do “reuse” my sources by making a copy of them as a way to start a new source.

I also transcribe the text from the images into the Research Note for the citation. And in some cases but not all cases, I also copy paste the same text into the note for a fact. For example, I have an Obituary fact and I copy and paste the obituary note from the Research Note into the note for the Obituary fact. I can include or not include the Obituary fact in my printed reports. I used to put the obituary text into the note for the Death fact, but doing it that way didn’t provide me with the same flexibility to include the obituary text or not the obituary text in my reports.

You can see a pretty good example my sources and citations and images at Sample RM data on the Web I created this page with a piece of software called GedSite, not with RM. But 100% of the data came from RM. About 99% of the time I spend “working in GedSite” is actually spent working in RM.

1 Like
10

Lol, I would say that makes you a source mincer.

I can see a certain utility in it. You don’t have to worry about how to group your citations together, since every source & citation are independent.

For me, though, I’m a bit more compusilve about clustering similar things together by more than just naming convention.

@MardeeB Jerry’s post brings another important foundational consideration to the conversation. You’ll probably want to decide your approach to source organizational granularity.

Some folks like Jerry are “splitters”, meaning their sources are split to some granular degree. (Jerry’s at a far end of that mindset.) Others group almost everything together, colloquially called “lumpers”. For example “Ancestry is the source of all, and individual items from Ancestry are just citations under it.”.

I split sources, but not to the degree that Jerry does.

11

Curiously, even though I’m on the far splitting edge of the source lumping/splitting spectrum, I consider regularly if maybe I should move more to the middle of said spectrum.

The impetus for me becoming such an extreme splitter in the first place was that no version of RM prior to RM8 supported reusable citations. So suppose I had memorized and pasted a single citation lots of different places. A good example might be a citation to an obituary where the obituary mentioned a large number of family members. And suppose I needed to make some change to said citation. I would have to spend a bunch of time chasing down all the copies of the citation and making the same change over and over again. And hoping I made the change correctly each time. And hoping I didn’t forget one I needed to change.

So why didn’t I switch over to the middle of the source lumping/splitting spectrum when RM8 first came out? Well, first of all, I didn’t give up on RM7 until the tail end of RM9, shortly before RM10 came out. I just found too many things in RM8 and RM9 that were much harder to do than they were in RM7. It took me a long time to work out those issues. Also, it took me a long time to be able to make printed reports for family reunions from RM8 and RM9. And even now, RM10 loses italics in my printed reports. I’m trying to figure out what to do about that problem.

But I have given a lot of thought to moving to the middle of the lumping/splitting spectrum. The thing that’s holding me back the most is that I would have to change a lot of my footnote sentences to be sure that all the Master Source variables are to the left of all the the Source Details variables in my footnote sentences. For example, here is my footnote sentence for my Death Record source template.

<[DeathRecordType]|Death Record>: [DeceasedName:Reverse];< Certificate [DeathCertificateNumber];> [DeathPlace], [DeathYear]<; [WhereViewed]; viewed [WhenViewed]>.

A sample footnote created by this template would be something like the following:

Death Certificate: Ailey, Henry Houston; Certificate 62-27211; Jefferson County, Tennessee, 1962; www.familysearch.com; viewed 11 February 2023.

Here was my thinking when I developed this footnote. How do I know when and where Henry Houston Ailey died? Well, my best source is Henry Houston Ailey’s death certificate. And truly, that’s all you really need to know. His death fact is going to say he died in 1962 in Jefferson County, Tennessee. So you go find his 1962 death certificate from Jefferson County. You don’t really need my citation to find it, once I tell you that the death certificate was my source. It’s easy to find when you know to look for it.

But I do include all the information you don’t really need, like the death certificate number and that it was in 1962. Before they were online, each year of Tennessee’s death certificate was a reel of microfilm for each year and for each county. And I tell you that I found it at FamilySearch, and I could just as well have found it at Ancestry. But as you read the footnote sentence let to right, each successive item is less important than the the previous item.

However, good source lumping and Evidence Explained want you to do something like the following:

Death Certificate, Tennessee, Jefferson County, 1962, www.familysearch.org

as the Source and as the left part of the footnote (and we could quibble about the order of the items in the Master Source.). And then they want you to do something like

Certificate 62-27211, Henry Houston Ailey

as the Source Details and as the right part of the footnote (and again we could quibble about the order of the items in the Source Detail). And I am sure that some authorities would want the 1962 in the right part of the footnote because it’s specific to Henry Houston Ailey, even though the Tennessee’s death certificate microfilm is organized by year and county.

But I like my footnote sentence better than Evidence Explained style and better than source lumping style. And I’m not sure I want to give up my footnote sentences, even if they are wrong according to all the authorities and according to how even modestly lumped sources and citations work.

One last thought: I don’t use Repositories at all in my sources and citations. I do agree that Web sites such as Ancestry and FamilySearch are not sources. At best they are repositories. But even then, the real repository is surely somewhere else. For census records, the real repository is NARA. How does including NARA in a footnote for a census citation help me find anything? For my Tennessee death certificates, the real repository is TSLA (Tennessee State Library and Archives) in Nashville, Tennessee (about three hours from my house). I used to go there quite regularly, despite the distance. But they have most of their stuff online now, plus much of their stuff is also at FamilySearch and Ancestry.

When I drove to TSLA, I treated them as repository. And as a repository, I recorded their address. But guess what, they have moved into a new building several miles from their old building. I don’t have repositories in my RM database anymore. But suppose I did still have repositories. Would I have to change the address of TSLA in my RM database for my old sources and citations I found at their old address? This is madness. Surely if you wanted to go physically to TSLA you would look up their current address and use it rather than using any physical address I recorded in the past.

As you an see, I do record Web sites, but not really in RM’s repository fields and not really as a major source field. I put the Web site names at the far right end of the footnote. I just think it’s madness to group my death certificates by Ancestry an by FamilySearch just because I get some of them one place and some of them the other. Which place I got them is pretty insignificant, and the the real repository is always TSLA anyway.

12

@MardeeB

I decided that today’s project would be to get a solid understanding of Sources, Citations, So thank you so much for asking the question.
I am so new to this, that I was not even sure HOW to ask -

Second, not exactly related, but a small suggestion that has helped me over the years with other (non-genealogy related) data base experiences.

What ever you decide hand write it a least once and then keep the notes in small binder for this and other “Convention Decisions”. Since my purpose is to share mostly with family - I have a list of “abbreviations used” - particularly for images and an “address book” for surnames - sometimes paper is nice -

Thank you and all for some many important things to think about. Today will be interesting

2 Likes
13

Changing course after so much time would definitely take a lot of clerical effort. Once you’ve got inertia it takes a lot to make foundational changes.

It sounds like your current method works well for you though. @MardeeB might benefit from hearing more details on why you’re considering the shift. What are the benefits you’d gain?

BTW it sounds like you have lots of ties to Tennessee. I’ve got tons of ancestral connections to DeKalb county.

14

My main motivation for considering a change to source lumping is that the RM8/9/10 user interface is very user unfriendly towards source splitting. I find the sliding panels throughout RM8/9/10 to be inherently user unfriendly, anyway. The panels are trying to put too much data into too little space so that you can’t see what you need to see when you need to see it. The sliding panels cannot be undocked into separate windows as I can do with many other software packages. And the sliding panels are even more clicky than RM7’s popup windows. The user interface moving to RM8/9/10 needed to be flatter and less clicky. And instead , RM7’s clicky popup windows were replaced by RM8/9/10’s even less flat and more clicky sliding panels.

The sliding panels are especially user unfriendly for source splitting. Everything about how sources and citations work in RM8/9/10 is predicated on being a source lumper and on maximizing your use of reusable citations. For one example, RM8/9/10 does its best to keep the Master Source panels closed and to keep the Source Details panel open. That’s the exact opposite of what I need as a source splitter. Even when I close the Source Details panels and open the Master Source panels, they won’t stay that way. RM8/9/10 quickly switches them back. For another example, in RM7 I could see my footnote sentences after entering my Master Source data, which as a source splitter was my only sourcing data. But in RM8/9/10, I cannot see my footnote sentences until I slide into the Source Details panels.

So if I were a source lumper in RM8/9/10, I would create new citations by copying existing citations and modifying the copy the same way I now create new sources by copying existing sources and modifying the copy. I would take full advantage of reusable citations. And more importantly, the sliding panels for Master Source would mostly stay closed automatically, and the sliding panels for Source Details would mostly stay open automatically. That’s what I would need as a source lumper. I wouldn’t constantly be fighting with the user interface the way I do now.

It may sound like I’m changing the subject, but there is one other thing that’s holding me back from changing to being a source lumper in RM8/9/10. I mostly update FamilySearch completely manually while logged into FamilySearch rather than using RM’s FamilySearch Central tools. But I am warming up to the idea of using RM’s FamilySearch Central tools instead for updating notes in FamilySearch. Doing so works great for a source splitter because RM updates FamilySearch with the notes from the Master Source. Doing so works terribly or not at all for a source lumper because a source lumper’s source notes are in the Source Details panel. I can’t think of a solution for that problem except that the design of FamilySearch Central for uploading notes seems broken to me for source lumpers. I could be wrong, and I would love to stand corrected. But that’s the way it seems to me. It’s like the RM designers forgot about source lumpers and FamilySearch Central when they added reusable citations to RM8.

15

My conversion would by assisted by SQLite scripts. Other users have converted from source splitting to source lumping in RM8/9/10 with the assistance of SQLite scripts, so it’s not like it’s a new concept.

16

Hi @thejerrybryan and @KimberlyGreen. I really appreciate you taking the time to think through this with me. Jerry brings up a good point and one that I have definitely been struggling with - the way the user interface works and the sliding panels.

I have no experience with RM7 so I don’t know how the sources worked before, but simpler would be helpful. I think I started with RM 8. I was an FTM user before that and gave it up because I kept losing data. (Also, to be honest, I was so new to this that I don’t really have an opinion about how that system worked).

The way that the source/citation panels flow in these more recent versions do seem very clicky. I prefer a keyboard and hotkeys, so every time I have to reach for the mouse, it slows me down. The amount of mousing/clicking required to update these sources is driving me just a little crazy because it’s just so slow. Right now, I’d kill for a spreadsheet-style interface that allows in-line updating. It would save me so much time.

I agree that the design of the user interface is definitely aimed towards the needs of source lumpers, but I’d note that the search features aren’t. To my knowledge, there’s no place that I can search and return a citation without knowing the original source. If I remember that I have the document (or can even see the media file) but don’t remember where I got it, it’s going to take a lot of digging to find it because you can’t search directly on citations and you can’t tell from the user interface where that media file is linked.

I’m also with Jerry on downloading the media files. I don’t want to have to keep ongoing subscriptions to all these sites just to be able to see my source documents. That said, it’s a lot of work keeping up all the captions on the media files, not to mention that there’s a whole separate issue about naming convention for the files themselves.

I’m still in a bit of a puzzle about the best approach, but it’s slowly starting to crystalize. Please feel free to keep the conversation going because I’m learning a lot. Also, maybe at some point, I need to break down and have my husband (a software developer) teach me how to use SQL Lite so I can pull and/or update some of this data directly. I have done some basic SQL in the past as a business analyst, but never anything that had to run from the command line and I find that kind of intimidating. I was much more comfortable in SQL Server where I could watch the file run and see the results as they were generated.

Thank you!!

17

I’m so glad you’re benefitting, and thank you for the suggestion. Paper can be nice, and the physical act of writing it down helps you retain it, so I may do that. I think I may also draw some of this stuff out on paper before spending more hours updating it into a particular structure. At the moment, I’m spending more time on shifting source and citation data around than I am on actual research!

Please feel free to share anything you come up with as you dig into this as well. I still have so much to learn.

Thanks!

18

In case you’re not aware, the forum has a nice compliment of SQL gurus. Besides Jerry, two additional forum member SQL gurus are @TomH and @RichardOtter, with both having developed tools and such to compliment RM. Check their user profiles for links. (Their stuff has saved me one than one headache and they’re just all-around nice dudes.)

RM does have something called the “Command palette”. Some of the commands have keyboard shortcuts. Learn more here. The palette can be viewed by clicking this icon:

RootsMagic_ZbWMZvMnJC
RootsMagic_ZbWMZvMnJC490×199 9.09 KB

You’re welcome :hugs:

1 Like
19

My Parents and oldest brother are listed in the 1950 Census not only same page but same building (3 floor tenant) 2nd and 3rd floor apartments along with my cousins.

20

That’s neat. I have found several related families on the same census page too, only in my case, living on nearby farms because they were in rural Iowa.