Elevenses with Lisa Episode 26 Video and Show Notes
Live show air date: September 24, 2020
Join me for Elevenses with Lisa, the online video series where we take a break, visit and learn about genealogy and family history.
Newspaper Navigator is a new free online tool for finding images and photos in old newspapers at Chronicling America. It doesn’t work the way the Library of Congress website works, so in this episode I show you how to navigate the Newspaper Navigator. It’s a fun session that will have you finding new newspaper gems in no time!
About LOC Chronicling America
Chronicling America is sponsored jointly by the National Endowment for the Humanities and the Library of Congress. It features free digitized historic newspapers spanning 1789-1963.
Newspapers Contain Imagery such as:
Photos
Drawings
Maps
Cartoons
Advertisements
You may not find the newspaper that you need for your research in the Chronicling America digitized collection. In those cases, turn to the US Newspaper Directory. It catalogs newspapers published 1690-present. Click the US Newspaper Directory button on the Chronicling America website to search. The catalog will tell you where known copies of the paper can be accessed.
Uses of Newspaper Images
Most of the old newspapers featured in Chronicling America include images. And because these old images are in the public domain, they are an ideal complement to family histories.
If you are very fortunate you may find photos or images of your ancestors, their homes, or other things specifically about your family.
Newspaper images are also a wonderful source when you need a photo or image to represent an important idea or item when telling your family’s story, whether in a blog post, article, book, video, PowerPoint presentation or other medium. Example of this would include a photograph of a blacksmith shop in the 1890s in the area where your ancestor worked as a blacksmith, or an advertisement for a Sears home kit just like the one your grandfather built.
Chronicling America’s Newspaper Navigator
The Newspaper Navigatordataset currently consists of 1.5 million pieces of extracted visual content from 16,358,041 historic newspaper pages in Chronicling America.
The visual content was identified using an object detection model trained on annotations of World War 1-era Chronicling America pages, made by staff and volunteers.
This “visual content recognition model” detects the following types of content:
Photograph
Illustration
Map
Comics/Cartoon
Editorial Cartoon
Headline
Advertisement
It also includes text corresponding to the imagery, identified by Optical Character Recognition (OCR).
Searching the Newspaper Navigator
You can search all images with captions. The results will be returned in a Gallery view featuring up to 100 images per page. This results format makes it very easy to quicky browse the images.
You can also switch to List view which lists the images along with the text retrieved by OCR.
How to Find Images Faster in Old Newspapers
Run a search in Newspaper Navigator of the word baseball and then run the same search in Chronicling America. A comparison of the results highlights the between Chronicling America and Newspaper Navigator when it comes to finding images in old newspapers.
The search results returned by the Newspaper Navigator are solely focused on photos and images. This means you have a fraction of the number results to review. Another big advantage of Newspaper Navigator over Chronicling America is the size of the image. Newspaper Navigator gives you just the large image to review, while Chronicling America shows you a thumbnail of the entire page with images so small that you must click and load the page to analyze them.
Images appear much smaller at Chronicling America and require you to click through to the page for closer examination.
Start by running a keyword search. (example: Blacksmith). On the results you can filter the results by Location and Years. Because the search currently doesn’t support Boolean operators or other types of search operators, you may need to run a few different versions of the same search to get a complete picture of the potential results. We’ll talk more about search strategies in just a moment.
Once you find an image you want, click to open it. The pop-up box offers these four buttons:
Download Image – Downloads a high-resolution copy to your hard drive.
Cite this – Generates a source citation that is automatically copied to your computer clipboard. Then you can simply paste it as needed. You can also cite the dataset by including the image URL, plus a citation to the website such as “from the Library of Congress, Newspaper Navigatordataset: Extracted Visual Content from Chronicling America.” According to the website, all images are in the public domain and free to use. Learn more about Rights and Reproductions at https://chroniclingamerica.loc.gov/about/.
Learn about this newspaper – Takes you to the Chronicling America catalog listing for the newspaper from which the image comes.
View Full Issue – Takes you to the complete newspaper issue at the Chronicling America website.
Click the buttons to select the options
My Collection at Newspaper Navigator
You can gather and save collections of the newspaper images you find using Newspaper Navigator. Start by running a search. On the results page click to select the desired images, then click the Save button. This will generate a URL for that collection and copy it to your clipboard. Since Newspaper Navigator doesn’t currently allow you to log in and return to your past searches during different sessions, I suggest pasting the URL into a research log for future reference.
Train My AI Navigators at Newspaper Navigator
A unique feature of the Chronicling America Newspaper Navigator is the ability to “train” the site to search for you. It does this through machine learning.
Elevenses with Lisa Episode 26
How to Train My AI Navigator:
Run a search
Click to select desired images
Click Save to save the collection of images
Click Train My AI Navigators
Newspaper Navigator will deliver a new set of images based on your selected images. On that page, select additional images that you want by clicking toward the top (+) of the image.
Click unwanted images by clicking toward the bottom (-) of the image.
Click to select the images you don’t want the AI Newspaper Navigator to find.
Click Train My AI Navigator again
Continue adding and subtracting images as needed to further train the system
Type a name for this training session in the Name My AI Navigator The saved AI Navigator name will appear in the Select an AI Navigator column
Click Save to generate a URL for this training session and paste into your research log.
Click + New AI Navigator to create a new training session spring boarding from the first
Click Clear & Restart to start a new search
Newspaper Navigator Search Strategies
Newspaper Navigator doesn’t, as of this writing, support Boolean Operators or offer an advanced search field. Here are some strategies that can help you have more success in searching the site:
Don’t use search operators, use variations
Even a space between initials can make a difference.
Each variation has the potential to deliver a different result in newspaper images.
Search Locations
Considering how many variations there can be to a name, when searching for ancestors try searching first on the name of their town or location. If there are still quite a few results, you can then filter to only newspapers from their state. I search the town name first because an article may appear in a newspaper from a different state. In the case of my search for McMinnville, I received a small, manageable results list. Had it been large and included both McMinnville, TN and McMinnville, OR, filtering to just Oregon would be helpful.
Test your search theories
Analyze your results and try variations based on what you are learning about what Newspaper Navigator is focusing on.
Search for word strings
In testing my search theories, I learned that Newspaper Navigator did not do well with multiple words that do not appear right next to each other. Therefore, I tried to find word strings that pertained to my family that I could search for such as the name of a business: Consolidation Coal Company.
Search for Photos
Another interesting search you can run is the word Photo. On the results page filter to the state and years that apply to your research.
Use List View to Find on Page
When dealing with a large number of results, List View can help speed up the review process. List View also displays the text generated by OCR. While not perfect, it can be helpful. Use your computer’s Find on Page feature (control + F on a PC, Command + F on Mac) and type in a keyword such as a surname. This will take you instantly to all occurrences of that word in the text on the page. Click the next page and run it again.
Find images quickly by word search in the List View
Learn More About Machine Learning
In the menu click Data Archaeology to learn more about machine learning and the Newspaper Navigator project.
If you’re enjoying the show, you can help others benefit from it too by leaving a comment below. Your comments…
helps me understand what matters to you.
helps others gather new ideas and encourages them to give the show a try.
tells Google / YouTube that this show is interesting and worth sending other people to through the search results.
provides great potential content for future episodes.
Clicking the red subscribe on our Genealogy Gems YouTube and then giving this video a thumbs up below the video
By leaving a comment (what you enjoyed, questions you have or what you’d like to see in the future) below this video after the show’s over or at the bottom of the show notes page if you’re watching on my Genealogy Gems website.
Answers to Your Live Chat Questions
One of the advantages of tuning into the live broadcast of each Elevenses with Lisa show is participating in the Live Chat and asking your questions.
Bert asks: Are some newspapers only available for a fee on websites such as Ancestry Lisa’s Answer: Yes, several genealogy websites have exclusive collections of digitized old newspapers. You can usually search or browse the site for free to determine if they have newspapers from the location and time frame that you need before you make a purchase. Here are some of my favorites that I’ve had good success with:
We are compensated if you make a purchase after using our links above (at no additional cost to you.) Thank you for supporting this free show by doing so!
Christine asks: (What is the ) newspaper navigator date range? Lisa’s answer: Here’s a break down of the dates:
Chronicling America covers 1789 – 1963 (digitized newspapers) Newspaper Navigator covers 1900 – 1963 (photos in digitized newspapers) U.S. Newspaper Directory at Chronicling America covers 1690 – present (catalog, only some are digitized and those are part of Chronicling America.)
Rachel asks: I have an ancestor that was in the social pages all the time in our local newspaper in the 1800’s. I thought it would make a great book or video, any ideas on how to showcase them the best? Lisa’s answer: I love both of those ideas and I cover many more in my Premium Membership video Inspiring Ways to Captivate the Non-Genealogists in Your Life. Personally I have found that short photo books and short videos that tell one story are received the best by family members. They both offer opportunities to share and highlight items from newspapers. Learn more about quickly and easily making family history videos by watching Elevenses with Lisa episode 16. And I strongly encourage Premium Members to watch these two videos:
lagomcurt asks: Are local small-town papers included in the collection? Lisa’s answer: Yes.
June asks: When you download it ask what to save as. What is your suggestion? Lisa’s answer: I think you’ll find that JPEG is currently the only option in the Save as Type drop-down menu.
Sharon asks: Does Chronicling America have foreign language newspapers in America? Lisa’s answer: Absolutely! Searching in the language will help retrieve items.
Ohio Waisenfreund newspaper at Chronicling America
Pat asks: Does it have Irish American newspapers? Lisa’s answer: Chronicling America does have Irish American newspapers. If they were published between 1900-1963 then they will be searchable by Newspaper Navigator. I would also recommend searching all newspapers (online and offline) by clicking the U.S. Newspaper Director button at Chronicling America. Then search by ethnicity (Irish) and Material Type (online.) You will find that some are linked to other websites where they can be found online. If you see an image of a newspaper on the catalog page, then you know it is available on Chronicling America in a digital format.
Search for Irish newspapers online at US Newspaper Directory
Mark asks: Can the wash out pictures be enhance with the new MyHeritage Photo with the sharping feature and colorization to make it a better final experience with images? Lisa’s answer: Yes indeed. Because the original quality will be poor and with low dots per inch (dpi) it likely won’t improve the way an original photo would. However enhancing and coloring just takes a few seconds and definitely improves the image. Even better, it often makes the print much more readable. I use it on documents too. Click here to try MyHeritage.
Newspaper photo enhanced and colorized with MyHeritage
Kathy asks: If you do a search in English, will it find the search term(s) in newspapers that were written in German? Lisa’s answer: No. You will need to search in German to pick up on any German text. However, if the image itself is similar, My AI Navigator should pick it up.
Lucinda asks: Who is in your necklace and the photo behind you, Lisa? Lisa’s Answer: It’s my maternal grandmother’s high school graduation photo.
Please Leave a Comment or Question Below
I really want to hear from you. Did you enjoy this episode? Do you have a question? Please leave it below. You can also call and leave a voice mail at (925) 272-4021 and I just may answer it on the show!
This week’s records roundup features Irish history in pictures, film and folklore; 1939 Register updates; British and Irish newspapers; UK WWI War Memorials Register, British folk music, Norfolk and Somerset parish records, Wiltshire wills and probate and Scotland...
The recent identification of the Golden State Killer through a DNA database for genealogy is just one way your DNA may be used in unexpected ways. Lisa Louise Cooke shares 5 key principles to keep in mind when considering your online DNA presence. Golden State...
The GEDCOM digital file format is essential to genealogy. My expert guest from FamilySearch explains what a GEDCOM is, how to use it, and the most recent changes. He’ll also answer some of the most common GEDCOM questions.
Show Notes
If you’ve been watching my videos for a while, then you probably know that I really recommend that you have a complete copy of your family tree on your own computer. But what if you’ve been building your family tree totally online up to this point?
The good news is that you can export your family tree as a GEDCOM file. But what exactly is a GEDCOM file?
Gordon Clarke, the GEDCOM Developer Relations Manager at the free genealogy website FamilySearch.org joins me to answer that question and provide the latest information about the GEDCOM.
What is a GEDCOM?
(00:54) Lisa: What is a GEDCOM?
(01:14) Gordon: GEDCOM is actually an acronym for:
GEnealogical Data COMmunication.
It’s a type of file with specific rules that allows digital family history products to exchange information. It’s been around so long that all the software companies can read and export it.
Say for example that you have a particular family tree program you’ve been working in but there are some features in another application that you like to try out. You want to try it out with a computer file that the program can read. All of the popular genealogy programs allows you to write a GEDCOM file and then you can read it in and review your information and add to it. That is what a GEDCOM is for.
It’s a specific file type that was works with most family history applications. It’s a text-based file, though it has special constraints to it. It was designed to be easily adaptable and compatible with importing and exporting. So, as long as the developers of both products adhere to GEDCOM specifications, you shouldn’t have a trouble downloading from one and uploading to the other.
Lisa: It sounds like each genealogy software database and website probably have their own proprietary file type, right? So, this is one everybody sort of agrees on that can extract the genealogy data set right. Is that right?
Gordon: Right, and there are differences between the proprietary program and GEDCOM. There are some products out there that only support GEDCOM. So that’s their proprietary format.
Why Use a GEDCOM?
(03:45) Lisa: So why should we use one a GEDCOM. When would we find ourselves wishing we had this universal file?
Gordon: Family history is more of a record keeping whether it’s photos and stories and genealogical data. People like to keep it and have control over it. So, GEDCOM is I like the word “personal”. You can personally control it. It’s just a .GED file, so any operating system can handle copying and emailing it. So, for personal control, preservation and sharing of genealogical data. It’s the most universally accepted format.
I would think for your backup purposes because it’s so universal, make sure that the program that you’re using has the ability to save your data in GEDCOM. Then you can decide whether you put it in your thumb drive or removable drive or you put it up in the cloud, you can decide how to preserve it. Think of it more as your personal file over this important information.
(05:31) Lisa: I like that idea. I’m probably not alone in that I once had somebody give me a little floppy disk and it had the whole family tree that this person had been working on. Unfortunately, it was a proprietary file, and it was a program that no longer exists. I’m helpless to be able to use it. So, a GEDCOME can really solve that issue.
Do All Family Tree Programs Support GEDCOM?
(06:00) You kind of touched on this, but I just want to just double check. Can all family tree programs and websites export the GEDCOM? Are you familiar with anything that don’t?
Gordon: I would say all of the popular programs and websites make it possible to import GEDCOM, and most of them allow for exports. There are some exceptions to that rule. If you’re going to spend your time using a program, look to see if it’s GEDCOM compatible.
To help even more so standardize the industry, the software providers commit to implementing the newest version of GEDCOM. Much of that is backward compatible. We presented those that have or will be planning to implement the newest version of GEDCOM at Rootstech. You can search at Rootstech for “GEDCOM” and see the videos of what’s been rolled out and what’s coming.
Who Owns and Controls GEDCOM?
(07:41) Lisa: Is there one particular group or authority or somebody who’s in charge of deciding what the GEDCOM is and how it works? Or is that a role that FamilySearch is playing?
Gordon: It is a role that Family Search has been playing. FamilySearch is the software development, education marketing, support arm of the department called The Family History Department of The Church of Jesus Christ of Latter Day Saints. So sometimes because of marketing reasons, people think that we’re different. Family Search is totally run by the Church of Jesus Christ of Latter Day Saints.
From a historical standpoint, the original specification was created and released in 1984. All subsequent versions have been copyrighted by the Church of Jesus Christ of Latter Day Saints.
Now, in the last three years, as a like a product manager, I took on the responsibility for working on the new version, version 7 of GEDCOM. But it’s always been an effort of FamilySearch as the outreach arm for the Family History Department.
What we did differently in this last version is we solicited all the key players and software companies. It was much more of a collaborative effort to go through the changes, things to keep, things to just get rid of. It took about two years working with many people. Now the version is what is called a public GitHub repository. As we worked toward version 7, it was to prepare it for a starting point. The decision process is still a steering committee sponsored by FamilySearch, but the input and the communication on changes is open to all software developers. You can learn about all that because it’s hosted at GEDCOM.io. So GEDCOME.info is kind of like the general public, and GEDCOM.io is more for technical software developers.
GEDCOM Features
(10:38) Lisa: What are some of the features of GEDCOM 7? What are some of the things that you consider when you’re continuing to develop the GEDCOM?
Gordon: The process that we worked on was, I think to eliminate ambiguity, there could be different software providers that would interpret the file specifications a little bit differently. We wanted to clean up the specifications so that there would be much more, not 100%, but a much better compatibility between the people that were reading it and writing it. So, we worked very tediously on eliminating the ambiguity.
I would think that the biggest thing is, it’s become more of a storage format of photos, and records and data. Let me read something, “FamilySearch GEDCOM version 7 incorporates the added ability to include photos, and other files when users download a FamilySearch GEDCOM 7 file from a supportive family tree product.”
Your local photos can be bundled in a special file that we called GEDZIP. It’s a GEDCOM file that is a zip package. That means that anybody that unzips that package will get the GEDCOM file and all the external files associated with it and have everything be readable. It’s a packaging technique to put everything together, which really adds to this idea of a personal preservation and sharing. Now you can package everything together and preserve it and share everything that’s important to you with others.
In addition to this zipped packaging capability, notes have been expanded for more versatile use and styling of text. When you add notes, whether it’s a relationship or a location, you can actually stylize those notes now and use bold and italic.
Many tools and sample files were created to help with self-testing. It’s based upon the Apache license, which is more of a technical slant on things, but to software developers, that means it’s an open software license. There’s a public GitHub repository that you go to github.com/familysearch so that you can request and watch ongoing changes in a more of a public environment, though Family Search is still the stewards and has the final say on decisions.
So that’s what’s new. It’s more open to the public. It’s been cleaned up with some important new features.
But backward compatibility for 90% of the GEDCOMs that are out there (and the last one was 5.5.1) is still possible. But it won’t go back to 3.0, 2.0. That’s where that’s where some of the incompatibilities are, is because people are using versions that are 20 years old. And things have changed a lot in the last 25 years. We have a clean, fresh start, and a new community working on continuous improvements. But there won’t be changes because the standards shouldn’t change much. This new version 7 is going to be pretty much the same for a while everybody gets on board.
Do GEDCOMs Include Image Files of Attached Records?
(15:21) Lisa: You mentioned photographs. Would that include image files? Would that include if we downloaded an image of a genealogical record which might be a .JPEG file? Would those come along with the GEDCOM?
Gordon: Yes, absolutely. All the elements of GEDCOM have definition of how to use them. And what’s called the multimedia link, the multimedia link means you can link to local files, JPEGs, PDFs, you know, whatever they are. And if you don’t want to put it all together, you can link to files that are in the cloud, and it will remember where they are. If you package them together in a GEDZIP file, and then you unpackage it, you’ll be able to access the local image files and the local records there.
So, this idea of putting it all together, I mean, bandwidth is much better than it used to be. But still, for people that have hundreds of thousands of images. This is not the best format for that. So they can work out a strategy taking into account the cloud service they use, and which photos they will keep locally on their computer. So, they can keep track of everything, both in the cloud and on their local drive. And that can all be referenced in this new version of GEDCOM.
Is There Data Loss When Exporting a GEDCOM?
(16:59) Lisa: Excellent.
So one of the questions I’ve heard from people is that they are concerned about loss data loss. If they’re importing or exporting, maybe going back and forth, is there a chance that you’re going to lose things or even introduce an error of some type?
Gordon: This is kind of the issue of the work on version 7. One of the biggest issues is not only new features, but to get a new standard to kind of clean the slate. If you get stuff into the new GEDCOM version 7 the likelihood of data losses is greatly reduced. So, we’re encouraging the adoption and use of GEDCOM 7 because it’s less likely to cause any data loss or errors.
Family Search and industry experts have worked for two years to remove ambiguities, simplify the definitions and samples in order to eliminate the possibility of data loss and errors when transferring between programs. In the long run, not only does it include more media, but the whole goal is to improve the consistency, the compatibility and minimize or even eliminate data loss. So, what you will start being seeing is the question “is GEDCOM 7 compatible?” Because GEDCOM 7, when we were working on something that was 20 years old, is going to be more compatible in the future. We have a body to watch out for it. Your data will migrate to the new version without data loss. But looking at down the road, staying with the version 7 or higher will assure a sure better preservation of what you have.
Learn More About GEDCOM at Rootstech
(19:17) Lisa: I think you mentioned or alluded to that there were some announcements at Rootstech 2022.
Gordon: Yes, go to the sessions and type in “GEDCOM” and you will get three opportunities. One is a session called GEDCOM 7 Launched and Rolling Strong. Another session will be FamilySearch GEDCOM 7 What’s Next? And the answer is teamwork.
There’s two pre-recorded videos about the What’s New in GEDCOM 7 and then how the industry’s going to join together in working on it in the future. In in one of the sessions, the first one, there actually is a slide that shows all the companies that have committed to it. But all the majority of the companies have said, both in the cloud and desktop and laptop, and some have said when they’re going to release it. And one company I think, is announcing their release at Rootstech of the new GEDCOM version 7.
Future Updates and Changes to GEDCOM
(20:44) Lisa: That’s great to see. Anything I didn’t ask you or that you think people should really be aware of as they move forwarding and keeping up to date with GEDCOM 7?
Gordon: Again, with a standard, we don’t want to change too much too fast, because they wanted to get solid as a new transfer format.
I think the big areas that we’re working on for future versions is related quite a bit to internationalization. There are probably 20 different calendaring systems that are different than what we do in the U.S. To be able to respect those different calendars and to understand the translation between calendars is a big part of internationalizing GEDCOM.
The other part related to that is that there are some places in the world where how they define relationships between people is not typical to either the US or Western Europe. And so we are working on major upgrades and encourage people to come join with us. With naming conventions we may think given name, surname, but in reality, there’s other relationships that get into the name. If we even go to Africa their name is the first name may go back 10 generations, so their name is a memorization of all those names. So, improving on names is an important effort, the structure and relationships.
Another improvement is places. We think hierarchal and certain jurisdictions, but over time, and in different areas of the world, how you organize places is different. We need to address that in the GEDCOM specification.
Sources and Citations need to be upgraded for the genealogical community. And so, we certainly invite not only software developers, but genealogists to join our effort to improve sources and citations.
GEDCOM Hypothesis
One thing I’m really excited about is that we have a team that’s been working a year, and they’re probably working on it another year or two, on what we call hypothesis. This is so that you can share information without claiming it as a conclusion, and keep it separate from a conclusion. This encourages collaboration. So instead of arguing about I’m right, you’re wrong, we call it a hypothesis. Then we can have a discussion until there’s enough sources to prove it. This Hypothesis module I think is going to be really exciting. But that won’t be for a couple years or so until we actually release it.
Lisa: I think that’s a terrific idea because so often we are just battling with ourselves over what we think the answer is, and we want to track it while we’re doing it.
I’m curious: sometimes we go to a website, and you have to pick what language you speak. Perhaps if you’re searching for videos on YouTube you might say English. Is this something being considered? Is the goal no matter what that it’s only one type of file that serves every country or was there a consideration that you could select your country and then the GEDCOM would support your calendar and your geographic areas. I’m sure that was a discussion.
Gordon: Oh, absolutely. And, but what you’re talking about, just to be clear, is the specification to give all of the options and more to the software developer. The software developer can decide the language of the interface, and many of them are already doing this. So the actual presentation, if it’s Norwegian, or Danish, or whatever, it’s different according to the language that you place. What we’re looking according to your language of choice is that the orientations are names, relationships, and places jurisdictions, will be easy for the software developer to switch to by just changing that.
When we look at an international – how people look at information – it may be a different lens that they look through. So having the ability to give the software developers out of our future specs, to switch their interface, and switch around because they might be working in one part of the country because of their heritage, and then they might work in another and to be switched between it and to still have the data be the same, regardless of what national lens they’re looking through.
Lisa: It’s amazing that one little package contains so much and so much flexibility. That’s really terrific.
The Team Working on GEDCOM 7
(26:52) Gordon: I won’t drop names but in my immediate steering committee, that we meet with weekly, not only do I have three representations from within FamilySearch, but from the community, I like to call them doctors, they are doctors, they have their PhDs in computer science. Some are genealogists, they have their peers, one is even a linguistic professor. Another is an actual legal professional. It’s been wonderful to work with such experts, really, that are reasonable, and want to make things easy for the software developer. So, it’s quite a dilemma, instead of just making it right in the specification, but we’ve got to make it right and make it easier for the software developers to implement it. So that’s my thanks to all the people I’ve been able to work with.
Click to select the images you don’t want the AI Newspaper Navigator to find.
Premium Video & Handout: Getting the Scoop from Old Newspapers. (Not a Premium member yet? Learn more here.)
