18 June 2019

3 Simple Rules for Managing Your Digital Genealogy Documents

My on-the-job organization skills help a lot with my genealogy research.

How did you begin this all-consuming hobby called genealogy? You may have started your simple family tree for a school project, or your kid's school project.

At some point, we each decided to get more serious about genealogy. We started looking for documents on genealogy websites. We tried to find our parents or grandparents on a census sheet, and we downloaded the images to attach to our family tree. Then we branched out. We found our grandparents' brothers and sisters living with their spouses and kids. Then we looked for ship manifests with the names of our immigrant ancestors.

It becomes addictive before you know it. Time passes, and we have a decent collection of facts and images.

Follow 3 simple rules and you'll always know where to find any digital genealogy file.
Follow 3 simple rules and you'll always know where to find any digital genealogy file.

But what do we do with all those census and ship manifest images?

I started downloading census images in 2002 when I got my Ancestry.com subscription. Seventeen years later, I have 732 census sheet images in one folder. I have 414 ship manifest images in another folder.

You may have a lot more than that. My census collection is small compared to the size of my family tree. That's because I don't have a single blood relative who was in a census-taking country before 1900.

It wasn't long before I realized I had a problem. I had to figure out how to organize the census sheets, ship manifests, and everything else I was finding. I wanted the alphabetical organization of files in folders to make it easy to find any one image.

I had to name and organize the files in a way that would always make sense to me.

GeneaLOGICAL™ Organization

These are my 3 basic rules for taming my collection of genealogy document images.
  1. Have one main FamilyTree folder. Mine backs up to a cloud automatically.
There are a limited number of genealogy document types. Filing by type makes great sense.
There are a limited number of genealogy document types. Filing by type makes great sense.
  1. Have a sub-folder for each type of document:
    • census forms
    • certificates (birth, marriage, and death documents)
    • city directories
    • draft cards (registration cards for World War I and II)
    • immigration (ship manifests)
    • military records (different than draft registration cards, these detail a person's military service)
    • naturalization (declaration of intent, petition for citizenship, and actual citizenship)
    • passports (applications)
    • photos (including grave marker photos)
There's no doubt what's where and which file is which. All it takes is a simple naming pattern.
There's no doubt what's where and which file is which. All it takes is a simple naming pattern.
  1. Name each file for the main person, last name first, and include the year. For example, in my immigration folder I have:
    • IamarinoPasquale1902.jpg. That's my great grandfather who came to America once and stayed.
    • IamarinoPietro1920-p1.jpg and IamarinoPietro1920-p2.jpg. That's my grandfather who came to America at a time when ship manifests covered two pages.
    • IamarinoPietro1958.jpg. That's Grandpa when he was a widower and went to visit his mother in Italy for the first and only time.
    In my census folder I have images named for the head of household:
    • IamarinoPasquale1910.jpg
    • IamarinoPasquale1915.jpg
    • IamarinoPasquale1920.jpg
    That's clear, right? Now here's an exception because my grandfather had a cousin with his same name:
    • IamarinoPeterLucy1930.jpg. That's Grandpa and Grandma Lucy.
    • IamarinoPeterMarie1930.jpg. That's Grandpa's cousin and his wife Marie.
    For Birth, Marriage, and Death records, it's Lastname Firstname Event Year. For example:
    • ZeollaPasqualeBirth1821.jpg. The 1821 birth record of Pasquale Zeolla.
    • MarinoFrancescoDeath1844.jpg. The 1844 death record of Francesco Marino.
    • IamarinoAngeloAntonioPozzutoAnnaelenaMarriage1817. Marriage records get the groom's name and the bride's name for clarity. Thank goodness for long file names. This is the 1817 marriage record of Angelo Antonio Iamarino and Annaelena Pozzuto
I've seen countless debates about family tree file storage. They all look too complicated to me, and not helpful at all. Some people suggest keeping the document images in folders separated by family.

If I had a separate folder for each family or each last name, I'd have an insane amount of folders. And where would I put my great grandmother before she married my great grandfather? In the Caruso folder or the Iamarino folder?

My file names would have to be much longer, too. How would I know the CarusoGiuseppe1900.jpg was a ship manifest and not a census form?

Say I need to look at my great grandmother's brother Giuseppe's 1905 New York State Census record. I know exactly where to find it. It's in the census forms folder, and it's named CarusoGiuseppe1905.jpg.

After 17 years, I haven't had a split-second of regret about my file-naming and filing system. A place for everything, and everything in its place.

Is your filing method driving you crazy? Are you wasting time trying to find the right image? Are you ready for a genealogy file do-over?

A word of warning. I use Family Tree Maker, and I let it copy all images into a single media folder. That way, anything I do to my images' names or locations has no effect on my family tree.

Your setup may be different. Don't dive into a big file naming project before testing what it will do to your genealogy program. Let's get organized!

14 June 2019

Searching for My Only Missing 3rd Great Grandparent

What can I do to find my closest unidentified ancestor?

It bothers me that I'm missing the name of only 1 of my 32 3rd great grandparents. It's amazing to have that column of my grandparent chart all filled in. But one of the spaces says "mother of Maria Luigia Muollo"—my 2nd great grandmother.

Come along with me as I try to find her.

My Muollo family comes from a tiny little hamlet called Pastene—part of a small town in southern Italy. Because the town is so small, it seemed to fly under the radar. It didn't start collecting civil records in 1809 like the rest of Italy.

All I can access on the indispensable Antenati website are:
  • 1861–1915 birth records
  • 1931–1942 death records
  • 1931–1942 marriage records
Those gaps in the record collection are enough to make me cry. I even hired a pair of Italian genealogists to go to the local church for records. What they found was wonderful. But they also discovered a huge lack of records.

I know that my 2nd great grandmother's father was Antonio Muollo. He was born in about 1818. I don't know when he died.

I've been piecing together families from the records I have. I've found only one other Muollo who's father is an Antonio. Maria Saveria Muollo was born in 1859.

Could she be my 2nd great grandmother's younger sister? There's a 16-year age difference between them. That's common in Italian families of the 19th century. And my Antonio was only about 41 when Maria Saveria was born. I know that Maria Saveria married Orazio Sarracino.

This is a TINY town. My 2nd great grandparents were Maria Luigia Muollo and Giuseppe Sarracino. If this theory pans out, the 2 Muollo sisters each married a Sarracino.

How can I prove Maria Saveria's relationship to my 2nd great grandmother, Maria Luigia?

The first step is always to gather as many documents and pieces of evidence as you can about the person. Let's see what I can learn about Maria Saveria Muollo.

To find more clues about Maria Saveria, I gathered the birth records for her children.
To find more clues about Maria Saveria, I gathered the birth records for her children.

Birth Records

I have the Pastene birth records for Maria Saveria's 9 children with Orazio Sarracino:
  • Maria Giuseppa, 1880
  • Maria Assunta, 1884
  • Antonio, 1886 (who must have died before 1892)
  • Antonia, 1889
  • Antonio, 1892
  • Francesco, 1894
  • Carmine, 1896
  • Rosaria, 1899
  • Maria Luigia, 1903
The last birth record, dated 24 November 1903, says Orazio was not present. He was domiciliato all'estero—living abroad. Sure enough, I found Orazio's 1903 ship manifest.

Immigration Records

Orazio and his daughter Maria Assunta left for New York THREE DAYS before his 9th child was born. I already can't stand this guy. Ships left for New York often. He had to leave when his wife was 9 months pregnant? Sheesh.

Orazio never sent for his wife to join him. Maybe he planned to return to Italy. Maybe he did return to Italy. Orazio died between 1910 and 1920. I don't know where.

In 1920, Maria Saveria and her 2 youngest daughters boarded a ship. They went to the Bronx to join her son Francesco. I like to think of her saying, "Who needs you? I'm going to America anyway."

This lady made that difficult voyage to America at 61 years of age.
This lady made that difficult voyage to America at 61 years of age.

Census Records

Back in the 1910 U.S. census, Orazio and his son Antonio are living in the Bronx. They're in the home of Orazio's eldest daughter Maria Giuseppa and her husband.

In the 1915 New York State Census, Antonio Sarracino is still living in the Bronx. He is married and has one child. His father is not with him.

In the 1940 U.S. census, Maria Saveria is 80 years old. She's living with her daughter Rosaria and Rosaria's family. It's amazing to find her at age 80. I would guess she didn't live a whole lot longer.

Death Records

One death record shows up on Ancestry.com that's promising. If it's Maria Saveria, she died in the Bronx in 1944 at age 84. The same record on FamilySearch.org offers more information. It says that her husband was Orazio Sarracino and her father was Antonio Muollo. It's her, alright.

But it also has the most important piece of information I'd been hoping to find: her mother's name. There it is.

Is Giuseppina Torrico my 3rd great grandmother? Is she the only 3rd great grandparent whose name I'm missing?

A check of the vital records in Pastene shows no one with that last name. Was it transcribed incorrectly because it was hard to read? What names do I have from the town that are close?

I can't accept a stranger's transcription blindly. That is not quite the right name.
I can't accept a stranger's transcription blindly. That is not quite the right name.

My best guess, without seeing the actual death certificate, is that her last name was Errico. Maybe the E was all fancy and hard to read. Errico is a fairly common name in Pastene. There used to be a neighborhood with that name in the 1860s.

After all that research, I'm left with a theory. A stronger theory than I had before. I'm still missing marriage and death certificates from Pastene, so I can't prove anything.

But perhaps—just perhaps—Maria Saveria Muollo is the younger sister of my 2nd great grandmother. And perhaps Giuseppina Errico is my missing 3rd great grandmother.

It's a good enough theory to record in my family tree.

11 June 2019

Drawing Inspiration from the Genealogy Pros

Do high-profile genealogists inspire you to do better family tree research?

If this sounds like I'm writing a grade school assignment, stick with me.

The genealogy professional I find most inspiring is Megan Smolenyak Smolenyak. She's a key researcher behind my favorite genealogy shows, "Who Do You Think You Are?" and "Finding Your Roots". She's an author and the former Chief Family Historian at Ancestry.com.

Megan works to identify the living descendants of deceased military personnel. She helps get the remains of these service members back to their homes.

Ms. Smolenyak also solves the mysteries of unclaimed bodies in the morgue. Finding living descendants is hard! You can watch a well-produced video to get a feel for Megan's work.

Megan's work with fallen military personnel reminded me of a website in my Favorites list. It's a database of Italian soldiers who died in World War I. There's another site for soldiers who died in World War II.

Long ago I searched for any Italian soldier with my maiden name who died in the first world war. I found only one: Alfonso Iamarino, born in my grandfather's hometown of Colle Sannita on 15 Feb 1892.

Alfonso's birth record and military record led to a ton of names, and our relationship.
Alfonso's birth record and military record led to a ton of names, and our relationship.

Finding living people is hard. But I should be able to identify this soldier's ancestors and siblings. My goal was to see where Alfonso fits in my family tree.

The fallen soldiers website tells me Alfonso's birth date and his father's name. The collection of the town's vital records sitting on my computer will make it easy to find his relatives.

But how will I find his relationship to me? Step by step.

Initial Facts

In 1892 when Alfonso was born, his father Pasquale was 30 years old. His mother Orsola Marino's age is not stated.

Since Pasquale was 30, I looked for and found his birth record in 1862. It even says at the bottom that he married Concetta Orsola Marino on 21 Dec 1889. I'll be sure to search for Alfonso's siblings later. But I want to go up his tree first.

Pasquale Iamarino's parents were Nicola Iamarino and Concetta Zeolla. The 1862 birth record said his parents were both 40 years old. But I couldn't find Nicola's birth record.

Don't know when a couple got married? Work your way back to their first baby.
Don't know when a couple got married? Work your way back to their first baby.

Searching Sideways

Since I couldn't find Nicola's birth record around 1822, I needed more information. If Nicola and Concetta were 40 in 1862, they should have several older children.

I checked the index of births for each year, going backwards from 1861. Nicola and Concetta's ages were so inconsistent! I found these babies:
  • Michele Arcangelo Iamarino, born 5 Apr 1859
  • Francesco Saverio Iamarino, born 19 Jan 1857
  • Giuseppantonio Iamarino, born 19 Feb 1852
  • Antonio Iamarino, born 13 Nov 1850
  • Angelantonio Iamarino, born 2 Apr 1849
I couldn't find any babies born before Angelantonio. It was time to search for Nicola and Concetta's marriage record.

I found them quickly. They married on 21 Feb 1848. The marriage records should include their birth records and their parents' names.

I found Concetta's 1824 birth record. But Nicola was different. Instead of the usual birth record, there's a 2-and-a-half-page document in hard-to-read handwriting. After staring at it for a while I was able to read it.

It says, in effect, "Oops! We can't find Nicola's birth record in the register. It isn't written there, and we don't know why. But we do know he was born in September 1819."

Family Tree Maker's color coding keeps me from overlooking these important relationships.
Family Tree Maker's color coding keeps me from overlooking these important relationships.

Making the Link

The unusual birth record for Nicola tells me who his parents were, as does his marriage record.

And that's where I found a lucky surprise. Nicola's parents were Angelo Iamarino and Anna Elena Pozzuto. Those names were familiar to me because I'd been looking at them about an hour earlier.

It turns out Nicola's sister Liberantonia was married the same day as he was. Their documents are listed one after the other in the 1848 marriage register. An hour before I found Nicola's marriage record, I discovered that Liberantonia's grandparents were my 5th great grandparents.

That makes Libera my 1st cousin 5 times removed. Her brother Nicola is also my 1st cousin 5 times removed.

Suddenly I realized Alfonso Iamarino, the only Iamarino to die in World War I, is my cousin. Alfonso's 2nd great grandparents are my 5th great grandparents. Poor Alfonso is my 3rd cousin 3 times removed.

Hey. I like this. I'm ready to choose another fallen soldier from my ancestral towns and figure out how we're related. For those of you who are of Italian descent, be sure to bookmark these sites:
How do your genealogy heroes inspire you?

07 June 2019

Let Family Tree Analyzer Find Your Duplicates Duplicates

The duplicates in your family tree aren't as easy to find as that.

I've written about the free program Family Tree Analyzer many times. It's the work of programmer and genealogy fan Alexander Bissett. See the bottom of this article for links to other articles about this software.

Today I've installed the newest version of FTA, and I'm eager to find something else to explore.

I've looked at the many options on the Errors/Fixes tab before. But this time I'm focusing only on the Duplicate Fact and Possible Duplicate Fact options.

Be sure to try the Possible Duplicate Facts option in Family Tree Analyzer.
Be sure to try the Possible Duplicate Facts option in Family Tree Analyzer.

Hopefully your list of duplicate facts won't be too long. I have 7 duplicate facts and 65 possible duplicate facts in a tree of 21,001 people. That's a reasonable amount. I can look at each one and fix the error.

As I whittle down my list of duplicate fact errors, I'm finding they fall into these categories:
  • Just plain forgetting that you already entered that fact.
  • Accidentally choosing the wrong fact type, like Marriage instead of Marriage License.
  • Adding the wrong date to a fact. This often happens to me with the 1940 U.S. census. It shows you someone's address on a date in 1940, but it also says if they were in the "same house" or "same place" in 1935. Sometimes I may paste in the 1940 date again instead of typing 1935.
  • Attaching a fact to the wrong person. I have a married couple in 1800s Italy with the respectful titles of Don and Donna. I accidentally gave both titles to the husband!

The types of duplicates formed a distinct pattern.
The types of duplicates formed a distinct pattern.

And then there are some results that are not errors. Family Tree Analyzer does call them "Possible Duplicate Facts" after all. Here's where I'm seeing that happen:
  • Duplicate marriage banns. In Italy, a couple might post marriage banns in the bride's hometown and the groom's hometown—on the same date. That looks like a mistake, but it isn't.
  • The same type of fact with no date. There are cases where I entered 2 addresses for people, but I didn't add a date. I need to go back to the source and pin down a date.
I'm very satisfied with this exercise. These are the types of errors you'd never find on your own. It's great that Family Tree Analyzer can be another set of eyes for you.

Find out what else Family Tree Analyzer can do for you.

04 June 2019

Create a Genealogy Task List to Keep Track of Loose Ends

My 12-year-old task list is a big help to my genealogy research today.

As soon as I realize I can't find a document for a person in my family tree, I make a note. I want to document that fact that I searched for this record, but I couldn't find it.

If I add a note like this to someone in Family Tree Maker, I also add a bookmark to the person. The bookmark in the index of names is a sign that there's something about this person I haven't figured out.

An even better option is to add a task to this person in Family Tree Maker. Your task list is the first thing you see each time you open your family tree file.

Document those unfinished tasks and family mysteries. Some day you may find the answer.
Document those unfinished tasks and family mysteries. Some day you may find the answer.

If you don't use Family Tree Maker, look for similar options in your genealogy software. If all else fails, you can keep a text file of all the little family history mysteries you'd like to solve one day. You might want to list them alphabetically by the person's name, and add the date you recorded the note.

Let's look at one of my tasks and what I can do to solve it.

Teofilo Iamarino was my great grandfather's brother. The task I recorded for him 7 years ago is, "Did he stay in the US? Check for 1910, 1920 census." Let's see what I do know about my 2nd great uncle Teofilo:

  • I have his Italian birth record from 1876.
  • I have his Italian marriage record from 1896.
  • I have his February 1909 ship manifest. He arrived in Boston, but headed to New York with my great grandfather and other relatives. My great grandfather came to America a handful of times, worked, and went back to Italy.
  • I have Teofilo back in Italy for the October 1912 birth of his son. He has to have been in Italy since at least January 1912.
  • I have his 1918 U.S. World War I draft registration card when he lived in Detroit, Michigan. The card says his nearest relative is his wife in Italy. I'm a little surprised that he had to register. He was a non-declarant alien who had no plans of becoming a U.S. citizen.
  • I have him in Italy for the July 1923 birth of his daughter. He has to have been in Italy since at least October 1922.
Each document you find may be a clue that leads you to the next.
Each document you find may be a clue that leads you to the next.

That timeline of known events tells me:

  • He may have been in New York for the 1910 census.
  • He may have been in Detroit for the 1920 census.

I'm not sure if I ever looked for Teofilo in Michigan, so I can check the U.S. census and include his Detroit address.

Nothing.

When I take out the Michigan address and hope to find Teofilo anywhere in 1910, I also get no results.

I can do a broad search for any documents I don't already have. I can leave off his last name and search for any Teofilo born in Italy in 1876.

Hurray, I found something new! I found Teofilo and 3 other men from his hometown arriving in New York in September 1913. According to the documents, he was in Italy in 1912. But he returned to New York for more work, leaving his wife to care for their young children.

In 1913 Teofilo returned to his brother Giuseppe's home in the Bronx. That's the same Giuseppe my great grandfather often returned to. It's the same Giuseppe my dad lived with as a toddler.

But wait! There's more. On the ship manifest, when asked if he had been in the U.S. before, Teofilo says yes, he was in New York from 1909–1911. That bit of information goes straight to the question I've been trying to answer: Was Teofilo in the United States for the 1910 census? If this ship manifest is right, then yes, Teofilo should have been in New York for the 1910 census.

I don't know why he's in Detroit, but now I know when he was in New York.
I don't know why he's in Detroit, but now I know when he was in New York.

Teofilo's brother Giuseppe has no boarders living with him in the 1910 census. It's only Giuseppe, his wife and daughter. Is Teofilo in another household nearby?

I can return to Giuseppe's 1910 census online because I always note exactly where I found an image. I can go page-by-page looking for Teofilo.

I love paging through a Bronx neighborhood in the census. I recognize most of the last names. They're names from my ancestral hometowns, or they're names of my parents' friends and neighbors.

Unfortunately, Teofilo was not in that 56-page census collection. But I can look at other collections that are in the same general neighborhood. The important thing is, I now have reason to believe Teofilo is there.

Before I look for more documents, I want to check out the other townsmen with Teofilo on that ship in 1913. One of them, Francesco Pozzuto, is a good fit with someone in my family tree. He may be the father of a woman who is a critical DNA match. She and her son are DNA matches to both my father and my mother. I'm adding a note and a bookmark to him right now.

This example shows the value of creating these tasks or bookmarks. Years can go by. Without that note, I wouldn't have realized I needed to do more research on Teofilo Iamarino. There are literally 21,000 other people in my family tree.

But today I felt like looking at my task list in Family Tree Maker. And I wound up discovering a ship manifest for Teofilo with a very important clue.

Genealogy is a long game. Leave detailed notes for future-you whenever you can.

31 May 2019

How to Make Your Sources Clear and Accurate

NOTE: There is a way to make your sources indisputable. Please see "Taming a Tangle of Source Citations."

Don't practice smash-and-grab genealogy. Make this method an unbreakable habit.

Seventeen years ago. That's when I graduated from scribbling in a notebook to building a family tree on my computer. I've had Family Tree Maker and an Ancestry.com subscription since 2002.

From the start, I never liked the option of saving a fact or a document to my family tree on Ancestry.com. I hated the ridiculously long source citations it added to my tree. I wanted my sources to be clear and easy to understand.

Because I run a tight ship, I have 275 sources in my Family Tree Maker file. And I have 20,963 people at the moment.

Here's how I keep my sources neat but thorough and retraceable.

Once I found the citation detail and citation text on Ancestry.com, it became too easy not to do.
Once I found the citation detail and citation text on Ancestry.com, it became too easy. I had to do it.

One Name to Rule Them All

When I started building my family tree, most of my sources were census pages and ship manifests. I didn't know what other people were doing. I only knew I wanted clarity. So my census source titles are as simple as can be:
  • 1900 U.S. Federal Census
  • 1910 U.S. Federal Census
  • 1920 U.S. Federal Census
  • 1930 U.S. Federal Census
  • 1940 U.S. Federal Census
If my tree has that the source of a person's address as the "1930 U.S. Federal Census," there's no mistaking where it came from. It doesn't need to say, "United States of America, Bureau of the Census. Fifteenth Census of the United States, 1930. Washington, D.C.: National Archives and Records Administration, 1930. T626, 2,667 rolls."

But a simple source title does not give us everything we need in our family tree. Behind the scenes, we need all the information. We need an absolute way to get back to the source. We need a link to the proof of each fact in our tree.

Fill in these details for the main source, not for each fact or document.
Fill in these details for the main source, not for each fact or document.

More Details Under the Hood

Family Tree Maker has a Sources tab where you can see and control all your sources. I've gone down the list of source titles looking for a few things:
  • Are people attached to this source? If not, I can delete the source. Maybe it belonged to people I've removed from my family tree.
  • Are there duplicate or very similar titles? If I decide to merge a couple of sources, I have to update each person with a fact linked to the source I want to merge.
  • Does each source have a clear title, citation details, a web address, and a repository?
If a source needs more detail, I go look it up. For example, I'll look up the 1915 New York State Census on whichever website I prefer. It can be Ancestry.com, FamilySearch.org, or any of the other official sites you may use. (Yes, it may be in a book, too.)

Instead of going to a particular document or record, go to the top-level page for that collection. On Ancestry.com, that page has the source citation detail and text, the exact name of the collection, and its URL.

These extra details make your family tree research more reliable.

Add enough details to each image to allow anyone to find it for themselves.
Add enough details to each image to allow anyone to find it for themselves.

Specific Micro-Details Where They Count

I want each document image in my tree to show exactly where it came from. It should tell anyone who's looking at it how they can find the original.

On the image of a 1920 ship manifest, for example, I added a breadcrumb trail to the description.

If you view this immigration fact in my tree, you'll see only the source title: "New York, Passenger Lists, 1820-1957". But the image itself has a description. And that gives you everything you need to know to find the original image file:
  • Line number(s). For a ship manifest or census form, I note which line number(s) to look at.
  • Source title. This is the exact name of the collection as seen on Ancestry, FamilySearch, or wherever. This will match my source title mentioned above.
  • Location of the film. Most of our genealogy documents exist on a roll of film somewhere. If I were trying to find that roll of film in a drawer at an archive, where would I look? This is always shown beneath the source title on Ancestry. For example, Roll > T715, 1897-1957 > 2001-3000 > Roll 2853. Or Wisconsin > Milwaukee City > 14 > Draft Card S.
  • Image number. If I asked you to go to John Stefaniak on a roll of Milwaukee, Wisconsin draft registration cards, it'd still take you a while to find him. So I include the image number. For example, image 664 of 973.
  • Web address. Sure, a URL may be different today than it was 5 years ago. But maybe it won't change for decades. And if it does, the film location and image number are even more important.
If you haven't been capturing these details for your documents, you may feel overwhelmed. Don't feel that way. Make it a habit going forward. Add a chunk of this annotation work to your annual genealogy goals. I finished all 630 of my census documents, but I have to go back and fix many of my 365 ship manifests.

The name of the game when it comes to your family tree's sources is usefulness. Each source should be useful in these 3 ways:
  • Understandable. Anyone can see where your information came from in general.
  • Retraceable. Anyone can follow your breadcrumbs to get to the original document collection.
  • Specific. At the document image level, anyone should be able to see exactly where this one unique image came from.
Make this a mandatory part of your fact gathering from now on. Don't just get the image and put it in your family tree. Don't just add the facts and let the image serve as your source. Gather all the facts about your source and record them where they belong. Right away.

Doing it right, from the start, makes the whole process much easier.