Genealogy's Star

Is FamilySearch.org AI Really AI?

2026-09-08T07:28:39.382-07:00

There are a few places on the FamilySearch.org website that appear to be AI driven. I am not certain which of these features are seen by which users. The AI research Assistant appears on my startup page. From time-to-time, I check to see what the suggested ancestors "who could grow" my family tree work out. Here is an example using the first "ancestor" on the list I found at the time of this post, "William Thurston 1809-1871 2Z3W-1N2."

It turns out that William Thurston 2Z3W-1N2 is a cousin (not technically an ancestor) who is connected to me by a link going back to 1702 through one of my less reliable lines. The record the hint takes me to is a birth record for "Infant Thurston."

If I put aside my concerns about the accuracy of the line leading up to and then down to William Thurston, I could easily attach this record to the Family Tree. It could be a baby that died in infancy. The record does show a William Thurston with a "no name" female born in March of 1848. This list of birth is not chronologically organized. There are birth from various years including dates back to 1828. the date in question is this number.

The number to me that it could be 1848. By the way, a quick check with Google Gemini about this document shows that it is not the complete document. Gemini seems to think the document says 1847 and that the "no name" children died in infancy. This raises the question of whether this child was stillborn (by our current definition) or lived for a period of time. The entry on the FamilySearch Family Tree for William Thurston already shows a child born in 1848 who shows no death date.

Now lets look at the William Thurston family. William, his wife, Eleanor Chase, and all the children were born in Bristol, Massachusetts. The suggested record is from Massachusetts. Birth Records 1847 | Barnstable. Birth Records 1848. Well, Barnstable and Bristol are almost at opposite ends of Massachusetts.

The record set of the vital records is actually the following.

Assuming that this one child was somehow born in a different county from the one where the family lived, there is already a child in the family born in 1848 who has a source showing she was two years old in the 1850 US Census.

So, without going through my whole family line, I would still not add this child to a family that already had a child born the same year and shows up in the 1850 US Census as a two year old.

Now what did the FamilySearch AI want me to do? If I ignored the fact that the suggested child was said to belong to William Thurston 2Z3W-1N2 who I might be related to and if I fail to check and simply believe what is offered by the AI Research Assistant, I find this is the family that the child would be attached to.

This is not even a valid record and who I am not at all related to.

So much for the AI Assistant.

US Copyright Law vs. AI

2026-09-07T07:11:48.836-07:00

There are three areas of U.S. law in which prediction of outcomes of litigation is difficult to impossible. The three areas include tax law, immigration law, and copyright law. The few contradictory statutory provisions in each of these three areas change on almost a daily basis. Most U.S. citizens have a vague, if not inaccurate, understanding of U.S. copyright law. For individuals and not major corporations, the rule that all copyright claims must be filed in the federal courts makes pursuing a claim of breach of copyright almost unavailable to the average citizen.

Let me explain this in simpler terms. Basic copyright law, as it is expressed in the statutory base, is easily obtained from the U.S. Copyright Office, which is part of the Library of Congress. See https://www.copyright.gov/ You can see the entire statutory law from this website by selecting a menu item to look at the law itself. See https://www.copyright.gov/title17/ To even begin to understand copyright law, you would need to study and understand all the provisions of Title 17 of the United States Code. However, the rest of the bulk of the U.S. copyright law is not readily available. Most of what is considered to be "the law" in the United States is court-made law by judges' decisions in specific lawsuits. To resolve any particular questions regarding copyright coverage and application, you would have to do extensive research on all U.S. court cases decided since the inception of copyright around the time of the U.S. Civil War.

If you want to get started with the existing court decisions, you can do a quick search on it, a program from Google called scholar.google.com. https://scholar.google.com/schhp?hl=en&as_sdt=3,45. Do a general search on copyright law, and you will find a response that shows about 180,000 cases. Major copyright cases can cost the litigants tens of thousands, if not hundreds of thousands, or even millions of dollars.

What about the current status of copyright and AI? Here is the report from the copyright office in a part of the website called "Copyright and Artificial Intelligence." See https://www.copyright.gov/ai/. Here is a quote from the website about the current status of AI and Copyright law.

Since launching an initiative in early 2023, the Copyright Office has been examining the copyright law and policy issues raised by artificial intelligence (AI), including the scope of copyright in AI-generated works and the use of copyrighted materials in AI training. After hosting public listening sessions and webinars, the Office published a notice of inquiry in the Federal Register in August 2023, which received over 10,000 comments by December 2023.

If you know anything about copyright or AI, you will realize that the U.S. government's efforts to address the issues cannot possibly keep up with the changes occurring on an almost weekly basis. Here is a screenshot of the Copyright Office's most recent announcements.

Here is the Copy Office's summary of the most recent release "Issue No. 1060 - January 29, 2025."

Today, the U.S. Copyright Office is releasing Part 2 of its Report on the legal and policy issues related to copyright and artificial intelligence (AI). This Part of the Report addresses the copyrightability of outputs created using generative AI. The Office affirms that existing principles of copyright law are flexible enough to apply to this new technology, as they have applied to technological innovations in the past. It concludes that the outputs of generative AI can be protected by copyright only where a human author has determined sufficient expressive elements. This can include situations where a human-authored work is perceptible in an AI output, or a human makes creative arrangements or modifications of the output, but not the mere provision of prompts. The Office confirms that the use of AI to assist in the process of creation or the inclusion of AI-generated material in a larger human-generated work does not bar copyrightability. It also finds that the case has not been made for changes to existing law to provide additional protection for AI-generated outputs.

Just ask yourself, "What has changed in AI from January of 2025?"

As an experienced trial attorney, if you ask me about a specific AI versus copyright issue, at the time of this post, I could not answer the question without spending a considerable amount of time on research. Even if I did the research, my conclusion would be my personal opinion. My review would have to include all the recent decisions made about copyright law. For example Thaler v. Perlmutter (D.C. Cir. 2025; U.S. Supreme Court, cert. denied March 2, 2026) See https://www.supremecourt.gov/search.aspx?filename=/docket/docketfiles/html/public/25-449.html

What more can I say?

Privacy and online family history

2026-09-01T05:03:48.538-07:00

I may use AI, specifically Google Gemini, to assist me in research, composition, and editing of my blog post; however, the ideas and content of this blog are my own.

One of the first things you should know about genealogy is that dead people do not have any rights to privacy. One legal example is the case of Mrs. Alline E. Cordell, Plaintiff-appellant, v. Detective Publications, Inc., Defendant-appellee, 419 F.2d 989 (6th Cir. 1969). That case held the following:

The courts are virtually unanimous in ruling that incidental mention of a plaintiff in a suit brought for the invasion of a relative's privacy is not an actionable tort. Coverstone v. Davies, 38 Cal. 2d 315, 323-324, 239 P.2d 876 (1952); Metter v. Los Angeles Examiner, 35 Cal. App. 2d 304, 310-311, 95 P.2d 491 (1939); James v. Screen Gems, Inc., 174 Cal. App. 2d 650, 653, 344 P.2d 799 (1959); Bradley v. Cowles Magazines, Inc., 26 Ill.App. 2d 331, 333, 336, 168 N.E.2d 64 (1960); Carlson v. Dell Pub. Co., 65 Ill.App.2d 209, 213 N.E.2d 39 (1965); Kelley v. Post Pub. Co., 327 Mass. 275, 278, 98 N.E.2d 286 (1951); see Milner v. Red River Valley Pub. Co., 249 S.W.2d 227 (Tex.Civ.App.1952). Cf. Annerino v. Dell Pub. Co., 17 Ill.App.2d 205, 210, 149 N.E.2d 761 (1958) (case not dismissed because defendant published a photograph of the plaintiff, wife of the decedent against whom the complaint focused its attack).

The Court also states:

As the District Court found, this is a case of an alleged unauthorized public disclosure of private matters. Certain limitations have almost universally been imposed on this kind of tort. Prosser, Torts, § 112, at 843 (3d ed. 1964). The cause of action is regarded as purely personal, and only the person about whom facts have unauthorizedly been published may recover. Consequently, the right lapses with the death of the person who enjoyed it, and one cannot recover for this kind of invasion of the privacy of a relative, no matter how close the relationship.

Over the years, certain exceptions to the no privacy after death rulings have been made for commercial, medical, and digital estates. Here are three examples of statutory exceptions carving out parts of the general rule:

Codification of Post-Mortem Rights of Publicity (Property vs. Privacy)

At the time of Cordell, appropriation was still viewed primarily as a personal privacy tort that expired at death. Beginning in the late 1970s and accelerating through the 1980s and 1990s:

The right to control the commercial use of one's name, likeness, image, and voice was decoupled from personal privacy and reclassified as a transferable property right.
Over 25 states enacted dedicated post-mortem right-of-publicity statutes (such as California's Astaire Celebrity Image Protection Act and Tennessee's Personal Rights Protection Act) that allow heirs and estates to enforce commercial likeness protections for decades after death.

Statutory Medical Privacy (HIPAA Omnibus Rule)

Prior to federal health privacy rules, hospital records of the deceased were governed haphazardly by state law.

Under the 2013 HIPAA Omnibus Rule (45 C.F.R. § 164.502(f)), a decedent’s protected health information (PHI) remains subject to federal privacy protections for 50 years following their death.

Digital Estate Laws (RUFADAA)

Because the common law in 1969 did not envision digital identities, cloud data, or online accounts, the Uniform Law Commission introduced the Revised Uniform Fiduciary Access to Digital Assets Act (RUFADAA) in 2015.

Adopted in almost every U.S. jurisdiction, RUFADAA prevents estates and fiduciaries from automatically accessing private digital communications (emails, direct messages) unless the decedent explicitly granted consent in an estate planning instrument or via an online legacy setting.

What this means for genealogists is that an unhealthy attitude toward privacy risks work being permanently lost upon their death. Even more importantly, excessive privacy restrictions often mean researchers spend years duplicating work that others have already completed. We share our genealogy for the simple reason that we share a common ancestry with all our cousins.

Understanding FamilySearch.org's User-Owned Trees

2026-08-31T05:25:00.343-07:00

https://www.familysearch.org/en/search/genealogies/

This blog post is basically a transcription of my online class for Sunday, 30 August 2026 from the BYU Library Family History Center. The narrative audio transcription has been edited by Google Gemini. The video recording of the class will be available on the BYU Library Family History Center YouTube Channel.

A major structural shift is taking place at FamilySearch.org and while it has been quietly in development for several years, its full implementation fundamentally changes how individual researchers and genealogical societies can manage their lineage data. For years, the centerpiece of the platform has been the shared FamilySearch Family Tree; single, massive, collaborative universal tree. The collaborative model has clear strengths, but every seasoned researcher knows the immense frustration of returning to a painstakingly documented ancestral line only to find that an inexperienced user has merged individuals incorrectly, altered vital conclusions, or stripped away hard-earned documentation. While safeguards continue to improve on the shared tree, FamilySearch has now rolled out a long-awaited alternative: User-Owned Trees.

The Evolution from CET to User Trees

During its testing and beta phases over the past four to five years, this initiative was developed under the working title of "Controlled Edit Trees," or CET. If you search online discussions or older forums, you will still see the CET acronym used frequently. However, the official designation adopted by FamilySearch is simply User Trees. The defining principle behind a User Tree is ownership and control. Unlike the open, shared Family Tree where any registered user can modify deceased individuals, a User Tree belongs entirely to the creator or designated managing entity. You decide who has permission to view, edit, or contribute. In practical terms, it functions much like hosting an individual pedigree database on desktop software such as RootsMagic, or maintaining an individual tree on platforms like Ancestry or MyHeritage. This is not only a welcome development for individual genealogists seeking to safeguard complex research; it also provides genealogical societies, lineage organizations, and family associations a stable environment to publish and preserve curated lineages without the risk of arbitrary public alterations.

Beyond the Static Files of the Past

Having multiple trees hosted on FamilySearch is not a new concept. Long before the universal Family Tree, the platform hosted millions of records across static compilations accessible under the FamilySearch Genealogies Collections such as the Ancestral File and the Pedigree Resource File (PRF) served as digital repositories for decades. The Ancestral File represented early static lineage submissions to the Genealogical Society of Utah, while the Pedigree Resource File allowed researchers to upload GEDCOM files. The major limitation of these legacy collections was their immobility: once submitted, they were frozen snapshots in time. Correcting an error required compiling and uploading an entirely new GEDCOM, leaving old, outdated submissions perpetually preserved in search results alongside corrected ones. User Trees modernize this infrastructure into a dynamic, manageable system. Rather than uploading static dead-end snapshots, you can now curate, correct, and update your lineage in place while retaining complete administrative authority.

Integration with Full-Text Search and Modern Discovery

This shift coincides with a much broader transformation of FamilySearch's underlying search engine. With the expansion of AI-driven Full-Text Search and features tested through FamilySearch Labs, the way documents and trees are indexed is moving away from rigid, pre-extracted summary indexes toward deep textual search across entire record repositories. Under new aggregated search tools, records and lineages uploaded across user trees and genealogical databases are indexed so that researchers across the globe can still discover connections to deceased individuals without compromising the integrity of your master tree. Deceased ancestral profiles remain discoverable, yet the authority to modify the underlying data rests strictly in your hands.

Accessing and Managing Your User Tree

Because FamilySearch has designed this feature to coexist alongside the shared tree rather than replace it, you will not find a prominent banner on the main landing page. Accessing the interface requires navigating through the search hierarchy:

1. Log in to FamilySearch.og and select the Search menu from the top navigation bar.
2. Click on Genealogies.
3. On the Genealogies landing page, scroll past the search fields down into the lower section of the page, where you will find the prompt to manage and upload your individual family tree.

From there, you can import a standard GEDCOM file exported from your primary genealogy software or build your line directly within the interface.

A Note on Virtual Classes and Moving Forward

This discussion formed the focus of my final live Sunday class presentation for the BYU Library Family History Center. The library is refocusing its instructional resources toward in-person student services, meaning live webinar broadcasts and virtual Sunday classes are concluding. Past class recordings, handouts, and webinars remain fully archived and accessible through the BYU Library Family History Center learning portal and the library’s YouTube channel. I will continue sharing regular research strategies, evidentiary analyses, and technological updates directly here on Genealogy's Star and across our ongoing educational channels.

If you have held off on integrating your work with FamilySearch out of concern over unauthorized merges or overwritten evidence, take a serious look at User Trees. They offer the archival reach of FamilySearch while keeping evidentiary control firmly where it belongs: with the researcher.

Should You Use AI to Repair or Enhance Old Photos

2026-08-28T08:00:09.087-07:00

One of the strange quirks of fate is that I have inherited thousands of old photos. The majority are unidentified. I guess my relatives decided they couldn't identify the people either, and I was a willing recipient of the whole pile. About 4,500 of the original photos came from my great-grandmother, Margaret Godfrey Jarvis Overson. She was a professional photographer, as was her father. Her collection went to the University of Arizona Special Collections Library in Tucson, Arizona (see

Overson Family Photograph Collection

). There are also about 1,000 of these photos online on FamilySearch.org. (see Margaret Godfrey Jarvis Memories).

The old photos are all in the public domain.

I am also a professional photographer, though not as active as I have been in the past. See my Walking Arizona blog, where I have just over 5,000 photos online.

Now what about using AI to repair or enhance some of these old photos? You can see that the photo above has several problems. The photo is torn and has a significant crack or line. Parts of the photo are obscured, and the boy's faces are washed out with little detail. The first issue is AI itself. AI does not repair or enhance the photos. It takes the original photo and regenerates or generates a similar copy, only with the additional information that it's missing from the original photograph. Here is a simple example of Google Gemini's Nano Banana's enhancement effort for this photo with a simple prompt to repair and enhance.

There are a few differences between this "enhanced" photo including parts of the photo that were missing have been restored, and the photo has been cropped slightly, apparently across the bottom. Most people would agree that it is an improvement over the original photo.

However, for historical purposes, it is not the original photo. What happens if I use a very developed, specific prompt giving instructions to Gemini about the actions it should take with regards to this old photo?

Here is the prompt that I am going to use:

Act as a master photographic conservator and senior digital restoration specialist specializing in archival preservation and Adobe Photoshop. Your objective is to analyze a historical photograph and generate an actionable, surgical digital restoration plan that prioritizes ethical conservation over cosmetic recreation.

### Core Guiding Principle

Adhere strictly to the Principle of Minimal Intervention: repair only what is compromised to stabilize the image, leaving 100% of authentic original detail, paper texture, and natural film grain intact. Never recommend global blurs, aggressive noise reduction, automated AI over-smoothing, or destructive layer merges.

---
### Step-by-Step Assessment & Repair Workflow
1. Medium & Baseline Analysis
- Identify or estimate the photographic process (e.g., Daguerreotype, Tintype, Albumen, Gelatin Silver, Cyanotype) and approximate era.
- Catalog critical intact elements (e.g., facial feature micro-details, authentic paper grain, midtone density) that must be strictly protected from alteration.
2. Damage Inventory & Categorization
Map the damage across three distinct categories:
- Structural Defects: Cracks, tears, missing emulsion/substrate, fold creases, punctures.
- Chemical & Surface Deterioration: Silver mirroring, foxing/mold, water stains, fading, chemical discoloration.
- Optical / Scanning Artifacts: Dust, scratch lines, scanner glare, sensor noise.
3. Non-Destructive Photoshop Layer Architecture
Provide an explicit, bottom-to-top Photoshop workflow organized into dedicated layer groups:
- Group 1: Base & Registration (Locked original capture layer, working duplicate, reference guides).
- Group 2: Structural Repairs (Sample All Layers healing/cloning, custom texture patching, brush modes).
- Group 3: Texture & Surface Cleanup (Frequency Separation: High/Low parameters, localized Clone Stamp / Healing Brush settings, blend modes).
- Group 4: Density & Contrast Correction (Targeted Curves adjustment layers, localized Black & White/Channel Mixer adjustments, Luminosity masks).
4. Surgical Tool Specification Table
Present a structured table mapping each damaged area to its precise repair method:
| Location / Feature | Damage Type | Photoshop Tool / Technique | Layer Settings & Blending Mode | Preservation Constraint |
|---|---|---|---|---|
| [e.g., Subject's cheek] | [Scratch] | [Spot Healing (Proximity Match)] | [Dedicated Blank Layer, Sample All Layers] | [Avoid adjacent pore texture loss] |
5. Quality Control & Verification Protocol
- Specify a "Difference Mode" or toggle-check inspection routine to verify that zero alterations have spilled into undamaged zones.

The advantage of using the prompt is that more of the original photo is preserved, and there are no apparent additions other than the original photo's missing areas

But I suspect that those people who are using AI to enhance photos would be expecting something completely different from of this photo "restoration." My long, specific prompt would not do the changes. It explained in great detail the reason why the changes could not be made for historical reasons. if I go back to ignoring the specialized prompt, and simply give some additional details to the photo, outlining the changes needed to the first copy of the photo you might get something like this:

My prompt consisted of the phrase, "Take out the large crack, take out the dark spots all over the photo, and enhance the background." I think that most people would agree that this was a better enhancement of the photo, but any connection to the historical, original photo has been almost completely severed. the background has been completely changed. Details have been added, especially to the floor and the faces. my guess is that as long as this technology is available, it will be used by most people, including most genealogists, to enhance their photos, especially very damaged photos like the one I use here as an example.

The most important issue is ensuring that the original photograph is preserved—both as a physical artifact and as an unedited digital representation. Whenever an enhanced version is shared online, it should be clearly marked as AI-enhanced, with the original copy provided alongside it.

As a quick note, none of the photos I put up on FamilySearch have been enhanced or changed in any way by AI.

Why I Use Google Gemini to Help with Writing my Blog Posts

2026-08-26T20:06:20.767-07:00

What I have learned over the past years is that I have a tendency to skip words and misspell a lot of words. Even when I proofread my documents, for some built-in reason I cannot see the errors. For many years, I have been relying on spell checkers and grammar checkers to correct what I say. However, with Google Gemini, I can take this assistance to a much higher level. Because my typing ability is degenerating over time, I also have moved to dictating with voice transcription almost everything I write. However, this causes a problem because the way I speak is more colloquial and not as polished as my writing should be. Here is an example. I dictated the above sentence using Wispr Flow. Now I am going to ask my Gemini assistant to review my paragraph for spelling and grammar and accuracy. Here is the product of my prompt.

Over the decades I have spent researching, writing, and teaching, the physical mechanics of how we get our thoughts onto the page have changed dramatically. We have moved from typewriters and carbon paper to word processors, and now to advanced digital workspaces. Along the way, our physical limitations and habits inevitably catch up with us. Over the years, I have noticed a persistent tendency in my own work to skip words or misspell terms, and even during a careful self-proofreading pass, the mind often sees what it intended to write rather than what is actually on the screen. For a long time, standard spell checkers and basic grammar tools offered a helpful safety net, but they were limited to rigid dictionary lookups and simplistic syntax rules that frequently missed the real context of a thought.
As typing becomes more physically demanding over time, adapting our workflow is essential rather than optional. Moving toward voice dictation tools like Wispr Flow has transformed how quickly and easily I can capture ideas, but speaking naturally introduces a distinct set of challenges. Spoken language is inherently colloquial, full of conversational shortcuts, loose phrasing, and informal cadence that does not always translate cleanly into the polished, structured prose required for published research, case studies, or instructional articles. Dictation solves the mechanical problem of typing, but it leaves the editing and refinement stage wide open.
This is precisely where modern artificial intelligence tools like Google Gemini provide practical value. Rather than functioning merely as an aggressive red pen, an AI assistant can be used as a sophisticated editorial partner that bridges the gap between raw, dictated speech and coherent written analysis. When you feed a spoken draft into a language model and ask for an evaluation of spelling, grammar, clarity, and structural flow, it evaluates the passage in full context. It identifies dropped words, straightens out winding colloquial sentences, and flags ambiguous phrasing without stripping away the writer’s original meaning.
However, utilizing an AI assistant in your writing workflow requires the same discipline we apply to documentary evidence: trust, but verify. These tools are assistants, not substitutes for critical thinking. When reviewing the output generated from a prompt, you must read the suggestions critically, ensuring that the tool has preserved your authentic perspective and has not introduced unintended distortions or hallucinations. Use voice transcription to get your thoughts down freely without physical strain, leverage AI to refine and polish the mechanics, and always give the final piece a rigorous manual review before sharing your work.

Yes, this is what was written from what I dictated in my first paragraph. Here is a copy of the prompt that I used.

# Role & Persona
You are an expert genealogy educator, researcher, and tech columnist writing in the authentic, authoritative, and practical first-person narrative voice of James Tanner from *Genealogy’s Star* (https://genealogysstar.blogspot.com/).

# Core Philosophy & Tone
* **Voice:** Practical, direct, experienced, tech-forward, and encouraging yet realistic. Avoid hype, fluff, marketing jargon, and sensationalism.
* **The Core Bridge:** Emphasize sound evidentiary methodology (original source records, careful citation, critical evaluation) while embracing modern digital and AI tools as aids—never substitutes—for sound research.
* **Format & Style:**
* Write in an essay/narrative style with smooth paragraph transitions.
* Avoid fragmented listicles, excessive bold topic headers, or artificial corporate section titles.
* Write with the perspective of someone who has spent decades in libraries, archives, and digital repositories, understanding both the pitfalls beginners face and the nuances advanced researchers care about.

# Task
Write a comprehensive, engaging blog post based on the source link(s) or summary provided below.

# Article Architecture
1. **Contextual Lead:** Open with a thoughtful reflection on how the technological or methodological landscape in genealogy is evolving.
2. **The Core Topic/Tool:** Introduce the specific tool, dataset, platform update, or methodology found in the source material.
3. **Practical Application & Analysis:** Discuss how researchers should practically use it, highlighting both its strengths and where it requires caution (e.g., verifying against original documents, avoiding common assumptions).
4. **Conclusion/Takeaway:** Conclude with practical advice on what the reader should do next in their own research.

# Input Data

[PASTE YOUR LINKS, NOTES, OR SOURCE TEXT HERE]

This is a third-generation prompt, meaning that I had an initial draft. I had Gemini critique it and rewrite it, and then critique it and rewrite it again. you might notice, but it does right in my style. reviewing the rewrite from Gemini, I find very little to criticize or change. this is, of course, the time necessary to review it, read it carefully, and make any changes I think are necessary.

in any event, I use the following as a notice that what I have written has been reviewed or altered by Gemini.

The content of this blog post has been reviewed for accuracy and content by Google Gemini.

Legacy Family Tree Webinars introduces Learning Paths

2026-08-25T07:09:21.130-07:00

Read the full announcement on the Legacy Family Tree Blog

Over the years, I have given a number of presentations for Legacy Family Tree Webinars. They have a family tree webinars section with an extensive number of high-quality presentations. It has increased its global influence by becoming part of MyHeritage.com.

With a library now exceeding 2,700 genealogy webinars, finding a specific entry point or structured educational sequence on Legacy Family Tree Webinars has presented a logistical challenge for many researchers. To address the sheer volume of available content, the platform has introduced a new organizational tool called Learning Paths.

Designed to function as short, focused courses, Learning Paths consolidate three to six related webinars into a logically sequenced curriculum. The initial rollout includes over 100 pre-created paths organized by subject matter. The curated sequences guide researchers from foundational concepts—such as beginning family history and understanding DNA estimates—to advanced applications like resolving conflicting evidence or incorporating artificial intelligence into a research workflow.

To streamline navigation, the paths are categorized by experience level, methodology, historical record types, and specific geographic regions, including various U.S. states, Canadian provinces, and European countries.

Beyond the pre-constructed options, the platform also offers a custom building tool. The "Build Your Own Learning Path" feature allows users to select any webinars from the broader catalog and organize them into personalized study plans. A highly practical function of this tool is its shareability. Genealogists can distribute their custom-curated lists to local genealogical societies, special interest groups, or family members, facilitating collaborative learning or guiding relatives through introductory topics.

The new feature includes an integrated progress tracker to help users manage their viewing history and resume their studies efficiently. Upon completing a path, the system generates a printable certificate of completion.

Enrolling in the paths, utilizing the custom builder, tracking progress, and generating certificates are available at no cost to all users. However, access to the video content itself continues to follow the site's existing model: webinars designated as free remain accessible to the public, while viewing any members-only videos within a path requires an active Webinar Membership.

See this presentation also: Introducing Learning Paths at Legacy Family Tree Webinars

This post was created, in part, with assistance from Google Gemini.

Introduction to the FamilySearch Help Center

2026-08-24T10:26:21.897-07:00

https://youtu.be/StO6PCVtEpw?si=mL1IINrl1MlqDKr6

Navigating a platform as expansive as FamilySearch can feel daunting, but the built-in Help Center is designed to guide researchers through every stage of their family history journey. In this instructional presentation from the BYU Library Family History Center, I provide a clear, comprehensive walkthrough of the Help Center’s vast features and resources. Viewers will discover how to access self-paced learning modules, search thousands of instructional articles, and utilize tailored research pathways designed for both beginners and seasoned genealogists.

Beyond basic search tools, this video explores deeper support layers, including the Learning Center’s country-specific research guides, technical skill lessons, and the interactive FamilySearch Community. Tanner also outlines key support avenues—such as specialized helper tools for consultants and direct contact options with FamilySearch specialists—giving you the skills and confidence to solve tricky genealogical problems and make continuous progress on your family tree.

The BYU Library Family History Center classes will be discontinued after August 30, 2026. The videos here on YouTube and on youtube.com will remain active for an unspecified period of time. For more rare information, see the link below.

https://familyhistory.lib.byu.edu/online-learning-update

You can view my own YouTube channel using the link on Genealogy's Star Blog, this blog.

As noted on the video itself, some of the images in this video were AI-generated. Additionally, I use AI for drafting, spell-checking, and conciseness for some parts of my blog.

BYU Library Family History Center Sunday Classes and Webinars Discontinued

2026-08-24T07:56:32.138-07:00

https://familyhistory.lib.byu.edu/online-learning-update

Yes, this says exactly what it means. The Sunday classes and webinars during the week are now canceled. I suggest you click on the above link and read the entire statement. The statement contains links to the following:

Kathryn Grant: Goldenrod Genealogy (https://www.goldenrodgenealogy.com)
James Tanner: Genealogy's Star (https://genealogysstar.blogspot.com)

When you read this blog, you should now see a link to my YouTube channel. https://www.youtube.com/@jamestanner45

If you need to talk to me directly, I suggest we use the BYU Library Family History Center Virtual Help Desk. This gives us unlimited time to speak in a breakout room and also for me to talk to a number of people at the same time. You can make an appointment by contacting the library or contacting me directly at genealogyarizona@gmail.com This address has been genealogyarizona@gmail.com for many years, and I have passed out thousands of cards with the email address.

Please click on the library's explanation link above and check out my YouTube Channel.

Thanks for you patience in this change.

MyHeritage adds Y-DNA to trace your paternal line back

2026-08-24T07:39:52.239-07:00

https://blog.myheritage.com/2026/08/new-on-myheritage-trace-your-paternal-line-back-thousands-of-years-with-y-dna/

For years, genetic genealogy platforms have largely separated autosomal testing—used for finding cousins across recent generations—from direct-line Y-DNA testing, which traces direct paternal lineages deep into human history. MyHeritage has bridged this divide by integrating baseline Y-DNA haplogroup reports directly into its existing testing database. Existing male MyHeritage DNA customers and eligible raw data uploads can generate baseline Y-DNA results without submitting a new sample. However, access to the reports is only included at no additional charge for users holding an active MyHeritage Complete or Omni plan. The MyHeritage Complete plan presently has a promotional first year price of $199 and then an annual renewal price of $299. The Omni Plan's promotional first year price is $239 and then $399 thereafter.

The Omni Plan is MyHeritage's all-inclusive bundle. It adds access to external properties owned by MyHeritage, specifically historical newspapers via OldNews.com, collaborative tree tools on Geni Pro, and the entire library of 2,000+ educational sessions on FamilyTreeWebinars.com.

Because the analysis relies on the limited Y-chromosome single-nucleotide polymorphisms (SNPs) captured during standard microarray autosomal testing, the resulting haplogroups represent broad, ancestral branches—often dating back to the Middle Ages or earlier. For researchers seeking high-resolution matching and sub-clade identification within recent genealogical timeframes (the past several hundred years), MyHeritage offers an integrated pathway to upgrade directly to FamilyTreeDNA’s Big Y-700 next-generation sequencing test. The FamilyTreeDNA Big Y-700 test is presently $399. These prices may change.

The newly introduced feature extracts Y-chromosome markers already present in the raw data of biological male test-takers, eliminating the need to submit a new sample. Developed in partnership with FamilyTreeDNA, the analysis assigns qualifying users a foundational Y haplogroup and places their lineage on the broader human paternal phylogenetic tree.

Once activated through account settings, the tool provides five core reports:

Haplogroup Story: An overview of the lineage’s classification, accompanied by historical context, a branch diagram, and badges outlining key lineage characteristics.
Migration Map: A geographical model plotting estimated ancestral migration paths originating from early human populations in Africa, including divergence points for parent and sibling branches.
Notable and Ancient Connections: Side-by-side historical comparisons linking a user’s lineage to notable historical figures and ancient DNA samples recovered from archaeological excavation sites.
Country Frequency: A distribution breakdown based on comparative testing databases and academic studies, illustrating where specific haplogroups are most frequently reported worldwide.

These feature may change from the first introduced features.

Regarding data handling, MyHeritage confirmed that the underlying computational models are licensed from FamilyTreeDNA but executed entirely within the MyHeritage infrastructure; user raw genetic data is not transferred to third-party databases during processing.

Access to the Y-DNA reporting suite is available immediately to male test-takers with active MyHeritage Complete or Omni subscriptions. Female users wishing to explore their direct paternal lineage will need to test a qualifying direct male relative, such as a father, brother, or paternal uncle.

This blog post was written with the assistance of Google Gemini. The post has been checked for accuracy.

The Changing World of AI -- Tokens and Users

2026-08-24T06:56:10.178-07:00

If you are wondering how to compare the various AI websites that have sprouted up over the last four years or so, there is a common way to do this. The most common comparison involves the amount of information, and to some extent the amount of detail and helpfulness of AI websites' responses. The comparisons are based on the total amount of information the AI chatbots have available. Tokens are the basic units of data (pieces of words, characters, or syllables) that the model processes at one time for input (your prompt) and output (its answer). Roughly, 1,000 tokens equal about 750 words. See What are tokens and how to count them?

The number of tokens available makes a substantial difference across three main areas: the context window (input/working memory), the output limit (generation size), and cost/speed. The following explanations came from Google Gemini.

Context Window (Working Memory): The maximum input token limit determines how much information an AI can analyze at once. A small window (e.g., 4k–8k tokens) limits you to brief queries, while a massive window (e.g., 128k to 1M+ tokens) allows you to feed in entire books, hundreds of pages of legal transcripts, extensive codebases, or complex multi-document archives without the AI "forgetting" earlier context.
Output Generation Limits: The maximum output token capacity dictates how long a single response can be. If an AI has a 4,000-token output limit, it cannot generate a complete 30-page document in a single pass; it will cut off mid-sentence once that threshold is reached.
Cost and Billing (APIs): For developers and API users, token volume directly drives pricing. Every token sent (prompt) and received (completion) is metered. A larger token budget means higher processing costs.

Back to Me:

What this means to a user is that the more complicated the requests that are made to the AI, the greater the cost. But the AI websites have a tiered subscription cost based on the amount of information processed or tokens used.

A comparison of the value of the AI website subscription could be made using a cost/token breakdown. However, consumer subscriptions are usually based on a flat monthly rate with usage caps. But comparisons can be made through the cost versus rate caps. High-end users work through an Application Programming Interface or API and are charged based on input and output tokens. Here is a table that Gemini generated showing the rate caps, reset windows, and soft limits of the most popular websites.

Provider / Plan ($20/mo)	Standard Chat Rate Cap	Reasoning / Deep Thought Limits	Context Window on Web	What Happens When Capped?
ChatGPT Plus (OpenAI)	~150–160 messages every 3 rolling hours	Strict separate allowances for dedicated reasoning modes (e.g., weekly/daily limits)	~32k to 128k tokens depending on model	Automatically throttles down to a lighter "mini" model until the 3-hour window resets.
Gemini Advanced (Google)	High soft cap (rarely hits a hard message wall in regular chat)	Dynamic limits based on server load when using extended thinking/analysis	1M to 2M tokens (ingests massive documents/books directly in the web UI)	Temporary slowdown / cooldown notice; drops to standard Flash models if server capacity is tight.
Claude Pro (Anthropic)	Dynamic 5-hour rolling pool (~45 messages for long threads; more for short chats)	"Thinking Mode" burns through your 5-hour budget significantly faster	200k to 1M tokens	Hard block until the 5-hour rolling reset window expires.
Microsoft Copilot Pro	~100–300 turns/day (integrated across 365 and web)	"Think Deeper" runs at slower speeds during peak hours	Typically capped around 32k–128k tokens	Slows to standard processing speed or caps conversation turn depth.

See for example: Claude pricing in 2026: every plan, API rate, and what it actually costs, and Cost of building and deploying AI models in Agent Platform.

Again, from Gemini:

Core Mechanics That Dictate Limits

Conversation Depth Depletion (Claude): Anthropic calculates limits based on total token volume sent. Because Claude resends the entire conversation history with every new message, long conversations or large file uploads consume your 5-hour quota much faster than starting fresh chats.

Fixed Time Windows (ChatGPT): OpenAI uses rolling 3-hour windows for general chat, making it predictable for burst work, but caps reasoning models with much stricter weekly or daily quotas.

Context Capacity (Gemini): Gemini offers the largest context window for consumers on the web (up to 2 million tokens), allowing entire archives or hours of video to be uploaded in a single prompt without immediately triggering rate limits.

Back to Me:

If you are a casual user of any of these websites using the free versions, it is possible that you have not encountered any data limitations. But a genealogist, for example, can run into an immediate token limit by asking one of the AI websites to transcribe or explain a long, difficult document. What is disturbing is that the lack of a token buffer can cause the AI chatbot to hallucinate or make up information. See Why Language Models Hallucinate.

If you doubt the accuracy of any of the AI output, I suggest you take time to learn how to ask questions and generate prompts. See

How to Improve Any Prompt

Another suggestion is that you spend time using your AI to explain the terms and conclusions of this blog post.

A Major Update of Keep or throw away: the challenge of genealogical documents and photos

2026-08-15T19:13:16.772-07:00

Over the past few weeks, we have spent a considerable amount of time working through tens of thousands of documents and photos and trying to reduce the number of documents and photos we save to a manageable one or two boxes for my and the same number for my wife. This post is a sequel to the earlier post called "Keep or Throw Away: The Challenge of Genealogical Documents and Photos." The key to this reduction is that thousands of the original documents were scanned and now reside on three large (16 - 22 TB) hard drives. One of those essentially duplicate hard drives in addition to my main computer are also backed up to the internet on https://www.backblaze.com/. Some of the documents, such as original journals, will also be offered to the BYU Harold B. Lee Library Special Collections.

There are still some major decisions that need to be made. The main issue deals with original photos and original handwritten letters. All the photos can be digitized, and all of the handwritten letters can also be digitized. The issue is: which of these two categories of oath should be preserved as physical documents?

Even when we are able to make the decisions and have decided on the original physical documents to be preserved, including letters, photos, certificates, and other historical artifacts, we will still need to work through the thousands and thousands of digital files. I certainly realize that most genealogists are not confronted with either the diversity or quantity of documents that were given to me over the last, going on, 45 years.

One of the main issues is that, over the time period that we have been accumulating these documents, I have had virtually no contact from any of my vast number of relatives asking if I have the documents and if they are available for review. One tactic my wife has used is to take piles of the documents to family gatherings and offer to let anyone at the gathering take some of the original documents. Of course, all the documents offered have already been digitized also.

Over the next few weeks, I intend to highlight and explain some of the interesting documents I have "discovered" while working through the piles. Here is an initial. example.

this is a graduation certificate for my grandfather, and it is about three feet square.

Stay tuned for further developments.

Computer Generated Family Trees Ignore the Heart and Soul of Genealogy

2026-08-08T16:25:31.428-07:00

https://www.familysearch.org/en/search/genealogies/submissions?f.collectionId=10000130

Just because you can do something does not make it a good idea. There is no question that a computer using an AI program could examine a record and create a family tree from the record. In fact, here is an example of an AI prompt using Google Gemini to do exactly that.

Here is the record.

Sahuaripa (Sonora). Registro Civil

Sonoyta (Sonora). Registro Civil

Tepache (Sonora). Registro Civil

Here is a suggested prompt generated using Google Gemini.

Act as an expert Mexican genealogist and master paleographer specializing in early colonial Spanish records. You strictly adhere to the Genealogical Proof Standard (GPS) and execute a reasonably exhaustive search methodology across Mexican archives (including Catholic parish registers, información matrimonial, civil records, notary protocols, land grants, and Inquisition/governmental administrative files).

Paleographical & Technical Directives:

Script & Hand Analysis: Identify and adapt to specific colonial hands—including Cortesana, Procesal, Procesal encadenada, and transitional hands (16th–18th centuries).
Contractions & Abbreviations: Systematically resolve historical abbreviations, suspension marks, and sigla. Expand abbreviated names, words, and titles in brackets [ ] within transcriptions while preserving original spellings in literal passes.
Orthography & Phonetics: Maintain historical spelling variants (e.g., interchangeability of b/v/u, c/z/ç/s, i/y/j, silent or added h) while accurately interpreting intended proper nouns and surnames.

Transcription Conventions for Damaged or Defective Manuscripts:

Apply standard diplomatic-paleographic bracket conventions to denote document damage, ink bleed-through (traspaso de tinta), fading, or torn margins:

[word]: Expansion of an abbreviation or scribal contraction (e.g., Fran[cis]co).
[text?]: Conjectural or probable reading of faded, obscured, or damaged text.
[...]: Illegible or destroyed text. Indicate approximate length when possible (e.g., [... 2-3 words]).
[sic]: Confirmed scribal error or unusual spelling in the original manuscript.
<text>: Text inserted by the scribe above the line (interlineado) or in the margin.
[strikethrough] or [crossed out: text]: Text canceled or erased by the original writer.
[damage: reason]: Explicit note explaining physical obstruction (e.g., [damage: hole in folio], [damage: heavy ink bleed], or [damage: torn margin]).

Tasks:

Dual Transcription & Translation:
- Literal Diplomatic Transcription: Provide a verbatim transcription strictly preserving line breaks, original spelling, unexpanded abbreviations, and the bracketed damage conventions above.
- Normalized Text & English Translation: Provide a modernized, expanded version alongside a precise, context-aware English translation that accounts for missing text inferred through formulaic record patterns.
Data & Context Extraction: Extract all explicit and implicit genealogical facts for every individual mentioned, including full names, relationships, dates, jurisdictions (doctrina, jurisdicción, parroquia), witnesses, godparents (padrinos/compadres), occupations, and social/ethnic status descriptors (casta, vecindad, originario de, titles such as Don/Doña).
Structured Relationship Table: Construct a clear, structured table/spreadsheet detailing the family connections established by the document with the following columns:
- Individual's Name (Literal & Normalized)
- Role / Position in Record (e.g., Subject, Father, Paternal Grandmother, Godparent, Witness, Notary)
- Relationship to Subject
- Stated/Estimated Event Date & Location
- Casta / Status / Origin
- Paleographic & Condition Notes (e.g., reading uncertainty due to bleed-through, reconstructed from formulaic context)
Genealogical & Evidence Analysis: Evaluate source reliability (primary vs. secondary), evidence type (direct vs. indirect), and information quality under the GPS. Flag name variations, conflicting evidence, or missing linkages caused by document damage, and suggest specific archival next steps (e.g., checking duplicate parish books or local notary registers).

Here is the analysis after applying the prompt:

Document Analysis: Civil Registration Death Records (Tepache, Sonora, 1941)

These entries come from a pre-printed form register of the Civil Registry (Registro Civil) in the Municipality of Tepache, State of Sonora, Mexico, dated July 1941. The script is a 20th-century modern cursive hand with standard abbreviations and minor printed form overlays.

1. Document 1: Folio 8, Acta No. 16 – Defunción de la Niña Margarita Ayón

A. Literal Diplomatic Transcription

Plaintext

Oficina del Registro Civil en Mpio Tepache, Sonora
Acta No. 16 dieciseis
Defunción de la Niña Margarita Ayón

En Mpio Tepache Sonora a las 12 Doce
horas del día 21 Veintiuno del mes de Julio del
año de mil novecientos 41 cuarenta y uno compareció en esta Oficina el
señor Ramón Ayón Y expuso que a las 17 diez
i siete horas del día de ayer 20 Veinti
un[o] [crossed out: de] [damage: printed line break overlap]
de Dolor en su domicilio de su casa
de Refugio Velarde, su hija [crossed out: Bru] Margarita Ayón
rita Ayón de raza [crossed out: g m c s] Y nacionalidad Mexi
cana de 37 días años de edad, de estado civil _____
_____ de ocupación _____ Le sobrevive
su cónyuge _____ La enfermedad causa de la defunción se
comprueba con certificado médico expedido por el señor Dr. _____
_____ Se mandó inhumar el cadáver en la fosa número _____
_____ de la clasificación de 2ª clase del panteón
de este Pueblo Fueron testigos de este registro los
señores Jesús Montaño y Manuel Enriquez
_____ el primero de 32 treinta idos años de edad, de ocupación
de Agricultor y domiciliado en este Pueblo
y el segundo de 58 cincuentaiocho años de edad, de ocupación de Agri
cultor y domiciliado en este Propio lugar quienes
manifestaron tener con el difunto el parentesco de _____
    Leída esta acta, el compareciente y testigos fueron conformes con su contenido y firmaron los
que supieron hacerlo

[Signatures]
COMPARECIENTE: Ramón Ayón
TESTIGO: Jesús Montaño
TESTIGO: Manuel Enriquez
Doy fe
EL OFICIAL DEL REGISTRO CIVIL
José G. Blanco

B. Normalized Text & English Translation

Normalized Spanish Text

Oficina del Registro Civil en Municipio Tepache, Sonora.
Acta Número 16 (dieciséis). Defunción de la niña Margarita Ayón.

En Municipio Tepache, Sonora, a las 12 (doce) horas del día 21 (veintiuno) del mes de Julio del año de mil novecientos 41 (cuarenta y uno), compareció en esta Oficina el señor Ramón Ayón, y expuso que a las 17 (diecisiete) horas del día de ayer, 20 (veintiuno) [sic: 20/veinte], falleció a causa de dolor en su domicilio de su casa [hija] de Refugio Velarde, su hija Margarita Ayón, de nacionalidad mexicana, de 37 días de edad, de estado civil [soltera], de ocupación [ninguna]. Le sobrevive su cónyuge [ninguno]. La enfermedad causa de la defunción se comprueba con certificado médico expedido por el señor Doctor [no consta]. Se mandó inhumar el cadáver en la fosa número [sin número] de la clasificación de segunda clase del panteón de este Pueblo.

Fueron testigos de este registro los señores Jesús Montaño y Manuel Enríquez, el primero de 32 (treinta y dos) años de edad, de ocupación agricultor y domiciliado en este Pueblo, y el segundo de 58 (cincuenta y ocho) años de edad, de ocupación agricultor y domiciliado en este propio lugar, quienes manifestaron tener con el difunto el parentesco de [ninguno].

Leída esta acta, el compareciente y testigos fueron conformes con su contenido y firmaron los que supieron hacerlo.

[Firmas]: Compareciente: Ramón Ayón. Testigo: Jesús Montaño. Testigo: Manuel Enríquez. Doy fe, El Oficial del Registro Civil: José G. Blanco.

English Translation

Civil Registry Office in the Municipality of Tepache, Sonora.
Record No. 16. Death of the Child Margarita Ayón.

In the Municipality of Tepache, Sonora, at 12:00 PM on the 21st day of the month of July in the year 1941, appeared in this Office Mr. Ramón Ayón, and stated that at 5:00 PM (17:00 hours) yesterday, the 20th, died of pain/colic at her home address, [child] of Refugio Velarde, his daughter Margarita Ayón, of Mexican nationality, 37 days of age, unmarried, with no occupation. No surviving spouse. The illness causing the death is certified by a medical certificate issued by Dr. [blank]. The burial of the body was ordered in an unnumbered grave of the 2nd-class category in the cemetery of this Town.

Witnesses to this registration were Messrs. Jesús Montaño and Manuel Enríquez, the first 32 years of age, by occupation a farmer, residing in this Town; and the second 58 years of age, by occupation a farmer, residing in this same place, who declared having no relationship to the deceased.

Having read this record, the declarant and witnesses agreed with its content and those who knew how signed.

[Signatures]: Declarant: Ramón Ayón. Witness: Jesús Montaño. Witness: Manuel Enríquez. Attested, The Civil Registry Official: José G. Blanco.

2. Document 2: Folio 9, Acta No. 17 – Defunción de la Niña Antonia Blanco

A. Literal Diplomatic Transcription

Plaintext

Oficina del Registro Civil en Mpio Tepache, Sonora
Acta No. 17 diecisiete
Defunción de la Niña Antonia Blanco

En Mpio Tepache, Sonora a las 18 dieciocho
horas del día 26 Veintiseis del mes de Julio del
año de mil novecientos 41 cuarenta y uno compareció en esta Oficina el
señor Loreto Blanco Y expuso que a las 17 diez
i siete horas del día de ayer 25 Veinti
cinco fallecio a causa
de Vaso y Dolor en su domicilio de este Pueblo
Su hija legítima la niña Mar
garita [crossed out: Ayón] de raza [crossed out: g m c s] Y nacionalidad Mexi
cana de 40 días de años de edad, de estado civil _____
_____ de ocupación _____ Le sobrevive
su cónyuge _____ La enfermedad causa de la defunción se
comprueba con certificado médico expedido por el señor Dr. _____
_____ Se mandó inhumar el cadáver en la fosa número _____
_____ de la clasificación de 2ª clase del panteón
de este Pueblo Fueron testigos de este registro los
señores Juan Moreno y Francisco Montaño
_____ el primero de 27 Veintisiete años de edad, de ocupación
de Labrador y domiciliado en este Pueblo
y el segundo de 31 Treinta i uno años de edad, de ocupación de Agri
cultor y domiciliado en este lugar quienes
manifestaron tener con el difunto el parentesco de _____
    Leída esta acta, el compareciente y testigos fueron conformes con su contenido y firmaron los
que supieron hacerlo.

[Signatures]
COMPARECIENTE: Loreto Blanco
TESTIGO: Juan Moreno
TESTIGO: Francisco Montaño
Doy fe
EL OFICIAL DEL REGISTRO CIVIL
José G. Blanco

B. Normalized Text & English Translation

Normalized Spanish Text

Oficina del Registro Civil en Municipio Tepache, Sonora.
Acta Número 17 (diecisiete). Defunción de la niña Antonia Blanco.

En Municipio Tepache, Sonora, a las 18 (dieciocho) horas del día 26 (veintiséis) del mes de Julio del año de mil novecientos 41 (cuarenta y uno), compareció en esta Oficina el señor Loreto Blanco, y expuso que a las 17 (diecisiete) horas del día de ayer, 25 (veinticinco), falleció a causa de vaso y dolor en su domicilio de este Pueblo, su hija legítima la niña Antonia Blanco [erróneamente asentado como "Margarita" y corregido en el título del acta], de nacionalidad mexicana, de 40 días de edad, de estado civil [soltera], de ocupación [ninguna]. Le sobrevive su cónyuge [ninguno]. La enfermedad causa de la defunción se comprueba con certificado médico expedido por el señor Doctor [no consta]. Se mandó inhumar el cadáver en la fosa número [sin número] de la clasificación de segunda clase del panteón de este Pueblo.

Fueron testigos de este registro los señores Juan Moreno y Francisco Montaño, el primero de 27 (veintisiete) años de edad, de ocupación labrador y domiciliado en este Pueblo, y el segundo de 31 (treinta y uno) años de edad, de ocupación agricultor y domiciliado en este lugar, quienes manifestaron tener con el difunto el parentesco de [ninguno].

Leída esta acta, el compareciente y testigos fueron conformes con su contenido y firmaron los que supieron hacerlo.

[Firmas]: Compareciente: Loreto Blanco. Testigo: Juan Moreno. Testigo: Francisco Montaño. Doy fe, El Oficial del Registro Civil: José G. Blanco.

English Translation

Civil Registry Office in the Municipality of Tepache, Sonora.
Record No. 17. Death of the Child Antonia Blanco.

In the Municipality of Tepache, Sonora, at 6:00 PM (18:00 hours) on the 26th day of the month of July in the year 1941, appeared in this Office Mr. Loreto Blanco, and stated that at 5:00 PM (17:00 hours) yesterday, the 25th, died of "vaso y dolor" (spleen/abdominal ailment) at her home address in this Town, his legitimate daughter, the child Antonia Blanco [erroneously filled as "Margarita" in the body text before cross-out; title establishes Antonia Blanco], of Mexican nationality, 40 days of age, unmarried, with no occupation. No surviving spouse. Cause of death verified by medical certificate issued by Dr. [blank]. The burial was ordered in a 2nd-class grave in the cemetery of this Town.

Witnesses were Messrs. Juan Moreno and Francisco Montaño, the first 27 years of age, worker/farmer, residing in this Town; the second 31 years of age, farmer, residing in this place, who stated they have no relationship to the deceased.

Having read this record, the declarant and witnesses agreed with its content and signed.

[Signatures]: Declarant: Loreto Blanco. Witness: Juan Moreno. Witness: Francisco Montaño. Attested, Civil Registry Official: José G. Blanco.

3. Structured Relationship & Extract Table

Individual's Name (Literal / Normalized)	Role / Position in Record	Relationship to Subject	Stated/Estimated Event Date & Location	Casta / Status / Origin	Paleographic & Condition Notes
Margarita Ayón	Deceased Subject	Self	Died 20 Jul 1941 in Tepache, Sonora	Infant / Mexican / 37 days old	Name written across line split (`Bru- / rita Ayón` crossed out and corrected to `Margarita Ayón`).
Ramón Ayón	Declarant / Informant	Father	Recorded 21 Jul 1941 in Tepache, Sonora	Adult / Resident of Tepache	Father of deceased infant; signed the record.
Refugio Velarde	Mentioned Household Member	Mother (Probable)	Resident of Tepache, Sonora	Adult / Resident	Form states: "en su domicilio de su casa de Refugio Velarde, su hija..." indicating mother or female head of household.
Jesús Montaño	Witness	None stated	21 Jul 1941, Tepache	Age 32 / Agricultor / Local resident	Signed as witness.
Manuel Enriquez / Manuel Enríquez	Witness	None stated	21 Jul 1941, Tepache	Age 58 / Agricultor / Local resident	Signed as witness.
José G. Blanco	Civil Registrar	Official (Oficial del Registro Civil)	21 Jul 1941 & 26 Jul 1941, Tepache	Official	Signed both records.
Antonia Blanco	Deceased Subject	Self	Died 25 Jul 1941 in Tepache, Sonora	Infant / Mexican / 40 days old	Record title lists Antonia Blanco; scribe accidentally wrote Margarita Ayón on line 9 due to copying error from prior record, then crossed it out.
Loreto Blanco	Declarant / Informant	Father	Recorded 26 Jul 1941 in Tepache, Sonora	Adult / Resident of Tepache	Father of deceased infant; signed the record.
Juan Moreno	Witness	None stated	26 Jul 1941, Tepache	Age 27 / Labrador / Local resident	Signed as witness.
Francisco Montaño	Witness	None stated	26 Jul 1941, Tepache	Age 31 / Agricultor / Local resident	Signed as witness.

4. Genealogical & Evidence Analysis (GPS Standard)

Source Reliability & Information Quality

Source: Primary official civil registry register (Registro Civil de Tepache, Sonora). Original register pages 8 and 9.
Evidence Type: Direct evidence for the dates, ages, causes of death, and paternal relationships of both deceased infants.
Informant Accuracy: High for basic parental and event details (fathers Ramón Ayón and Loreto Blanco personally reported their daughters' deaths within 24 hours).

Critical Observations & Scribal Errors

Acta 16 (Margarita Ayón):
- Scribal Error: The clerk crossed out pre-printed/handwritten entries regarding raza (race/casta), reflecting post-revolutionary legal transitions removing race categories from civil registration.
- Maternal Identity: Refugio Velarde is listed as the head of household or mother ("en su domicilio de su casa de Refugio Velarde, su hija...").
Acta 17 (Antonia Blanco):
- Copying Slip (Lapsus Calami): In line 8-9 of Acta 17, the clerk mistakenly began writing the subject's name from the adjacent record (Acta 16) as "Margarita [Ayón]" before striking it out. The header clearly establishes the child's correct name as Antonia Blanco.
- Cause of Death Terminology: "Vaso y Dolor" refers to a traditional diagnosis for abdominal/splenetic disease or severe infant colic.

Archival & Research Recommendations

Parish Registers: Search the local Catholic church registers (Parroquia de San Joaquín de Tepache) for the baptismal records of Margarita Ayón (born ~June 1941) and Antonia Blanco (born ~June 1941) to confirm mothers' full names, maternal grandparents, and godparents.
Marriage Records: Search civil and parish marriage registers for Ramón Ayón & Refugio Velarde, as well as Loreto Blanco and his wife (not named in Acta 17), between 1925 and 1941 in Tepache, Moctezuma, or surrounding Sonoran municipalities.

Do you happen to read Spanish? Can you look at the original record and tell whether or not the information given by Google Gemini is correct? The only way that I could possibly rely on this transcription and analysis would be if I could actually read the document for myself, which I can do. The information provided is extensive and very convincing, and it is obviously more convenient. Absent the knowledge about the record and the ability to translate the information from Spanish into English, it would be impossible for an individual to determine whether or not all the information provided by Gemini was accurate. There is no trade-off.

Real Genealogy Threatened by Illiteracy

2026-07-31T07:17:26.658-07:00

Genealogy is a literacy intensive pursuit. Lack of literacy skills essentially closes the door to any real involvement in family history. Lack of literacy may also contribute to lack of interest. One simple example. Many users of the FamilySearch.org website are frustrated with apparent unsupported changes to individual entries. The common reaction is to tell the offending contributor to "read or look at the sources." What if the contributor cannot read at higher than a sixth grade level? Will telling them to read the sources help to stop the changes? What if the sources are all in handwritten script? Even some of my college level grandchildren cannot read handwritten documents. I have often written about the need for some type of qualification before making changes to existing entries in the FamilySearch Family Tree but to no avail.

Here is my favorite example of unrestrained changes being made to an individual entry without a modicum of effort to read the extensive sources available on this person's Profile Page.

Here is a current screen shot of the changes made to just this one individual.

I count 12 changes in the last week and this has gone on for years. How much genealogical time and effort have been wasted by these frivolous and unneeded changes? By the way Francis Cooke LZ2F-MM7 has 35 very valuable sources. How much of this constant waste of time is the result of people who cannot read?

Interestingly, my own grandfather, who was a newspaperman and editor, currently has his entry marked as read only in the Family Tree and no one can make changes except FamilySearch whose last changes were in 2024. This also means I can't add any additional memories.

I can think of a number of ways that these sort of changes could be avoided. Here are a few sources about illiteracy in America.

Online Fake Genealogy -- Surname scams

2026-07-25T11:21:42.718-07:00

To legally and legitimately claim an authentic Tanner coat of arms, a researcher must trace their direct, unbroken male line back to a specific ancestor who was officially granted or recognized as bearing those arms by an official heraldic authority. Because there is no authentic, traceable lineage of my particular Tanner family past the most remote ancestor who lived and died in Rhode Island, there is absolutely no way to connect my family or my family name with a coat of arms. Scams aimed at connecting a family line to royalty, nobility, and some sort of coat of arms are fairly common today. Sometimes there is a tendency to use pre-paid pedigrees and faulty research to establish the claim.

Coats of arms are primarily granted by official heraldic authorities, such as the College of Arms in England or the Court of the Lord Lyon in Scotland, through a formal petition process resulting in Letters Patent. Applicants often need to demonstrate merit, such as public service or professional achievement, and pay fees to have a unique design created and registered. See Granting of Arms.

Genealogy scams exploit your family history to steal money or personal data. Watch out for phishing inheritance emails demanding "processing fees", hacked messages or fake websites selling generic surname books, and data theft where scammers map your living relatives for identity fraud.

What is crucial to involvement with genealogy is being aware that many of these "fake" genealogies appear on the FamilySearch.org Family Tree and other online genealogy family tree programs. It only takes me a few minutes on my family lines on the FamilySearch Family Tree to find lines going back to the 1200s and earlier many of which have no sources.

You can begin to see the issues when you look at the descendancy of one of these individuals.

If you take the time to click up and down in this particular type of entry, you will see that majority of these names have either one source or zero sources. So, even if you personally do not fall for the fake pedigree scam, you may still have a lot of scammed names in you acestral lines as they are shown by FamilySearch.

Keep or throw away: the challenge of genealogical documents and photos

2026-07-09T10:23:34.794-07:00

This may not be obvious, but it is a generated image and it does bear a significant relationship to the recent status of the boxes in my basement. The past few weeks, various members of my family have come and "assisted" us in working through the piles of boxes in the basement and making the hard decisions about preservation or simply throwing them away. One decision involved incorporating our greater family at a recent reunion, providing boxes of documents, and allowing them to make choices about what they wanted to keep. The real issue is what to keep.

Over the years, I have evolved a strategy for record retention. Obviously, the number of records affects the application of the strategy. If you have one box of records and a few photos, this decision to keep or throw away is probably already made. What you have is what you want to keep. On the other hand, if you are like me, the recipient of various ancestral collections of documents and records and photos, then the decision becomes much more complex.

Before getting started with making decisions about record retention, it is a good idea to look on FamilySearch.org to see whether any of the memories available are records that you are holding in your own collection and thereby duplicating. Photos fall into a different category, and I will discuss them after writing about documents.

The different retention levels I have evolved are as follows:

Records with obvious historic value. This includes handwritten old journals, diaries, and letters between members of your family in the distant past. The distant past for some may vary from others. In my case, it is great-grandparents and beyond. The questions to ask at this first level include whether or not the documents are unique or one of a kind and would have not been preserved by any other method or repository. Assuming that this type of document is not readily available online, then keeping these copies and sharing them is very important.
The second level is a little less inclusive. It contains official documents: birth, death, marriage, and official certificates from employment or military service. The dividing line here is the issue of who is responsible for being the archive for certain types of records. If the records of any type are easily obtained by searching on the Internet, then perhaps it is not up to an individual to preserve those types of records.
The third level is more readily identifiable. This includes greeting cards, birthday cards, postcards, and other items that have no actual content other than a signature. It also includes newspaper articles, whole newspapers, and a variety of documents from your ancestors' involvement in the community, such as:
- dances
- parties
- vacation trips with their friends
- involvement in civic service
- and so forth
The fourth level is really, really easy to identify. This is everything else: news advertisements, receipts for payments, ledger books without any identification other than numbers and so forth.

You may want to divide up your levels of retention differently than I have done here because of your own unique family situation. The secondary issue is also important, and that is the question of whether or not an archive would be interested in keeping and maintaining the records.

I mentioned that photographs fall into a different category. Each photograph is essentially a one-of-a-kind document. However, you may fall into the trap that I have extensively suffered, and that is that my ancestors were either professional photographers or felt a duty to take a picture of everything they saw. I have literally hundreds of photos of each of my family members. The ultimate question is: How many photographs do you want to preserve of your parents or siblings?

One solution for part of the problem of making the decisions is to digitize everything. Once it's digitized, it is no longer taking up space in your basement, and you may have a greater incentive to dispose of the paper documents. At this stage, it is important to carefully distinguish between unique historical documents, such as an original journal and its digital copy. In this case, the original can have an historical value not shared by other types of documents. The issue of how many photos to save of your parents and grandparents and siblings may be resolved by digitizing all the records, all the photographs, and then disseminating the digital images out to family members.

Finally, you may want to investigate whether or not some of your most important documents, such as journals and diaries, may have enough historical interest that a university special collections library or other repository may have an interest in preserving the document itself and making it available to your family members. This is particularly true of FamilySearch.org. Original diaries and journals sometimes have enough interest to be preserved directly by FamilySearch.

In conclusion, I just want to make sure that you understand that we spent countless hours digitizing and organizing the records and only finally came to a conclusion to spend an extraordinary amount of time during a few weeks to attempt to control the rest of the content on paper sitting in the basement

Outstanding Repair of Old Damaged Photographs by Google Gemini

2026-06-26T09:51:34.672-07:00

The basic issue with the AI repair of old damaged photos in the concept that the AI recreates the image in the form of a copy of the original, while adding the details that are impacted by the damage. The old Photoshop technique was to work pixel by pixel and repair the photo using the information in the adjacent pixels. With a photo like the one above, that process, just for the crack and the missing segments of the outline of the image, would have taken hours. Giving Gemini a prompt like this and brief instructions about what is required will dramatically produce the resultant photograph in about 15 seconds.

Act as a master photo conservator and expert digital restoration artist specializing in Adobe Photoshop. Your goal is to evaluate the damage to a historical photograph and create a precise, surgical repair plan that prioritizes preservation over total recreation.
Follow this strict multi-step process:
1. Baseline Assessment: Describe the original, undamaged elements of the photo (e.g., intact subject features, paper grain, original contrast). This defines what must NOT be altered.
2. Damage Inventory: Map out the specific types of damage found in the image, classifying them into:
- Structural (tears, cracks, missing sections, creases)
- Surface/Chemical (stains, fading, silver mirroring, mold spots)
3. Surgical Repair Plan: For each identified area of damage, specify the exact, minimal tool or technique required (e.g., Clone Stamp for localized texture replication, Content-Aware Fill for clean edge tears, Frequency Separation to remove stains without destroying skin or paper texture).
4. Preservation Safeguards: Explain how you will monitor and isolate changes (e.g., using dedicated non-destructive layers, precise masking, and blending modes) to ensure that the original, unblemished portions of the photograph remain completely untouched.
Adhere strictly to the rule of minimal intervention: make only the changes necessary to stabilize and repair the damage, preserving 100% of the authentic historical texture and detail where no damage exists. Do not apply global smoothing or destructive filters.

Here is another example of a repair from Google Gemini:

Granted, the photo itself is not a good photo but the repair improves the image. I think there is an argument for being careful about using the digitized images without disclosure, but in this case, the repaired image is clearly marked with the Google Gemini logo in the lower right-hand corner.

It is also absolutely important to keep the original photo, even though it is damaged assuming that it is a one-of-a-kind photograph.

Brick Walls Come in all Sizes -- How big is yours?

2026-06-19T07:08:42.893-07:00

I have always been in disagreement with the common family history and genealogy analogy to brick walls. Often, the term is applied to research situations that have appeared to reach a dead end. However, in real life, a brick wall is easily scaled or, even from a practical standpoint, can be knocked down quite easily using the proper equipment. But rather to continue to denigrate the use of the term "brick walls" in the genealogical context. I think it would be helpful to examine the real-life end-of-line situations that are common to every single effort to extend a pedigree.

First of all, a short mention of the time-wasting efforts to extend a pedigree back to Adam or some royal ancestor. See my posts for "Are you related to royalty?" and "Back to Back to Adam" You might also want to review my post called "The end of an ancestral line: A Significant Genealogical Challenge"

The most common example of a brick wall involves a fixation with finding only birth, marriage, and death (BMD) records and relying almost exclusively on census records. This issue is particularly evident from the content of the sources cited on the FamilySearch.org Family Tree. The goal of extending a pedigree back to the early 1800s can usually be accomplished using records that fall into these two categories. It is also apparent from examining hundreds, if not thousands, of entries in the FamilySearch.org Family Tree that most of the research done by users ends when a person is identified using these records. Unfortunately, reliance on BMD and census records ends in the early 1800s, and any further research relies on less easily obtained record sets.

The next category of brick walls is also extremely common. This category refers to an unrealistic fixation on the identity of a particular ancestor. The genealogist or family historian focuses an extraordinarily large amount of attention to one person rather than extending their research to a community of people. This is usually called FAN or Cluster Research. In the vast majority of cases, this ancestor is an immigrant from one area of the world to another. See "Immigration: The Greatest Genealogical Challenge (three parts)" Over the years, I have probably written hundreds of articles about the difficulty of determining the origin of an immigrant. See "How does an indentured servant, redemptioner, or enslaved ancestor affect your research?"

The next brick wall category is also extremely common on the FamilySearch.org website. I use the FamilySearch website as an example because it is a huge mixture of readily accessed individual family trees, whereas trying to use a website such as Ancestry.com for the same purpose would be impossible. This common brick wall is difficult to visualize because the line seems to extend forever past the actual end. My challenge to anyone willing to spend the time has been vindicated over and over again. I simply tell people that if I look at their part of the FamilySearch Family Tree, I can easily determine or find a person whose ancestry goes back generations, including royalty, when there are not adequate sources provided to establish the actual relationship. This issue is not just limited to people with extensive pedigrees such as those with Utah pioneer ancestry. It is also sad that these genealogical researchers may spend an extraordinary amount of time doing descendancy research from ancestors they are not really related to.

Even though the analogy breaks down at this point, the next brick wall involves reality. Historical records documenting individuals begin to disappear in the 1600s and may completely disappear by the mid-1500s. In short, the ability to extend a line past 1600 is tremendously more difficult than any more recent time period. When attempts are made to do this, it is common for the lines to rely solely upon same name, same person conclusions. This is often the most common leap into the fantasy of royal or noble ancestry. I certainly do not want to denigrate the efforts of extremely cautious and talented researchers to extend their family lines into actual connections with royalty or nobility, but I find so few of these are accurate on the FamilySearch Family Tree as to make them extraordinary.

The last, but probably not really the last, category of brick walls is just due to inertia, a lack of interest and effort to extend a pedigree line.

Take heed: involving yourself in genealogical research is an educational challenge. No matter how much you already know, almost every newly discovered ancestor becomes his or her own need to broaden your educational background. Get busy and keep learning.

AI Comes to The Family History Guide

2026-06-19T06:16:48.501-07:00

The Family Histoy Guide is diving into the world of AI assisted learning. Each of the categories on the startup page lead to an AI experience that will help you quicking navigate the website and find the help you need. Of course the old method of navigating all the family history information on the website is still available through the menu bar. The website continues to provide in depth learning for all aspects of family history and genealogy. Here is a short summary of the content of the website with some assistance from Google Gemini.

The Family History Guide is a free, comprehensive learning and research platform designed to simplify genealogical research. Its vast library of resources is structured into several core categories:

1. Dedicated Learning Paths

The platform offers structured, step-by-step tutorial sequences ("Projects, Goals, and Choices") with integrated articles and instructional videos for the four major genealogy platforms:

FamilySearch
Ancestry
MyHeritage
Findmypast

2. Research Skills & Knowledgebases

Research Skills: Guided paths ranging from basic research strategies to advanced principles like the Genealogical Proof Standard.
Country Research: Over 60 dedicated country pages providing historical context, specialized research strategies, and localized records.
Ethnic Research: Dedicated sections focused on African American, Asian, Basque, Hispanic, Jewish, and Native American genealogy.
QUIKLinks: Direct, "just-in-time" links that launch you directly into specific record search screens across various databases.
The Vault & Knowledgebases: A massive secondary repository organizing hundreds of deep-dive articles and videos by topic.

3. Training & Tracking Tools

Trainers Resources: Lesson plans, presentation slide decks, 2-minute demos, and materials designed for instructors, consultants, or family history centers.
Online Tracker: A interactive tracking tool (available online or as downloadable documents) that lets users monitor, save, and measure their learning progress.
Quizzes & Exercises: Practical activities to test your understanding of various research methods.

4. Specialized & Community Resources

Activities: Hundreds of interactive, family-friendly genealogy activities separated by audience (Families, Individuals, Youth, and Children/Kids Corner).
Computer Basics: Fundamental tutorials covering Mac and Windows operating systems, desktop applications, and basic web navigation for beginners.
Faith-Based Material: Special tracking sections and custom guidance mapped specifically to Latter-day Saint (LDS) family history processes, alongside resources for other faith traditions.

Is the FamilySearch AI Assistant Really an AI Assistant?

2026-06-18T18:00:42.502-07:00

An AI assistant is designed to work with you in real time—answering questions, finding data, and keeping you firmly in the driver's seat—whereas an AI agent is built to execute workflows for you behind the scenes. The FamilySearch Research Assistant does not answer questions and keep you in the driver's seat. In the image above, the FamilySearch AI Assistant gives me 5 possible "new ancestors" Let's see how these selections actually work.

#1 William Inkerman Baxter 1885–Deceased GRPH-L4K

Clicking on the entry I get the following:

When I view the record, I see the following:

At this point, I have no idea how I am related to this person. So let's look at his profile before attaching the record.

The England and Wales Census 1921 record shows up as a record hint. So no real new information is being added by the AI Assistant. The entry cited is not for William Inkerman Baxter but for someone namedd William Rikerman Baxter who was born about the same time as William Inderman and in the same county. However, all the sources for William Inkerman have his name as Inkerman. I can't view the original record for the 1921 Census on FamilySearch. All FamilySearch has is an index.

There is no real reason to go on looking at this particular situation. Of course, a FamilySearch User could believe that the entries produced by the AI Assistant are somehow pre-validated and just follow the suggestion and add the two names to the family and leave it at that.

You might argue that some of the entries are correct and can easily be verified, however, my experience is that the suggestions made to me lead to complex problems. Here is the next name in the list.

In this case, Hosea Haywood GQK3-48C appears in a family of nine children. The husband and the wife are born in four different places: Whitwick, Leicestershire, England, Ashby de la Zouch, Leicestershire, England, Burnley, Lancashire, England, Coalville, Leicestershire, England, The wife, Clara Jane Buckley 9KKQ-GL5 has a death in Nottinghamshire, England.

Here is a Google Map showing the location of each of these towns and one county.

The 1901 England and Wales Census show Hosea Haywood age nine in Coalville. According to a quick search on Findmypast.com, there are 11,463 records for Hosea Haywood born in 1892 in England. I guess this is enough of caution for all AI apps: Always verify anything that they find or say and doubly so, with the FamilySearch.org AI Assistant.

How to Improve any Prompt

2026-06-10T12:55:06.273-07:00

A fundamental part of mastering communication with AI chatbots (subsequently AI) involves learning how to structure and write effective prompts. To develop this ability it is important to understand that the AI can help you draft an initial prompt on a specific subject and also refine and improve your and its own efforts. This means that you direct the AI to critique its own work.

For the purposes of this article, I am using Google Gems and Google Gemini. You can write your initial draft right in the Google prompt box or anywhere convenient. Here is an example of what you might write.

To illustrate this, we can start out with an initial idea for a prompt. You tell the AI how to act and what you want it to do. Here is an example. Obviously, this initial prompt could have been better written. We want to get Gemini to improve this prompt by analyzing it and making it more useful. This gives Gemini a task.

Here is the revised prompt:

The Upgraded Prompt
Act as a master genealogist adhering strictly to the Genealogical Proof Standard (GPS). Your task is to extract, evaluate, and contextualize census data for a specific family based on the transcription or raw data I provide.
When I provide the census data, execute the following four steps:

1. Data Extraction: Create a clear, structured table extracting every individual in the household. Include columns for verbatim name, age, calculated birth year, birthplace, occupation, relationship to the head of household, and any other notable columns (e.g., immigration year, naturalization status).

2. Evidence Analysis: Flag any internal inconsistencies or historical anomalies. Point out unlikely age gaps between children, discrepancies in expected relationships, or historical context specific to that census year's instructions (e.g., how the 1880 US Census differs from 1870 regarding relationships).

3. GPS Alignment: Actively look for clues that point to other records. Identify what crucial family information is hinted at but remains missing or ambiguous in this specific record.

4. Research Planning: To help me fulfill the "reasonably exhaustive search" requirement, recommend three specific, targeted record collections (e.g., specific probate jurisdictions, land deed registers, or vital records) that are most likely to resolve the ambiguities identified in Step 3.
Here is the census record data: [INSERT TRANSCRIPTION OR RECORD TEXT HERE]

Now, review this suggested prompt and add or change anything you think would help. You can do this revision by copying the prompt into Microsoft Word or Google Docs Or any other word processing program. Perhaps you can only guess what might make it better.

The next step is important. This is where you get the AI to critique itself. Copy the upgraded prompt and start a new chat. Copy the AI-generated chat into that new chat. It is important that you start a new chat and not continue with the old chat because that would confuse Gemini. Give Gemini in the new chat the same instruction about reviewing and revising but remember to put the review instructions first.

Act as a master genealogist adhering strictly to the Genealogical Proof Standard (GPS). Your task is to extract, evaluate, and contextualize census data for a specific family based on the transcription or raw text I provide.
Maintain a highly rigorous, academic tone. Differentiate clearly between direct evidence, indirect evidence, and negative evidence.
When analyzing the provided data, execute the following five steps in order:
1. **Exact Extraction & Calculation (Table):** Create a structured table extracting every individual. You must strictly separate verbatim text from your own deductions. Include columns for: Verbatim Name, Reported Age, Calculated Birth Year (mark with * to indicate calculation), Birthplace, Occupation, Stated Relationship, and any Notable Flags (e.g., tick marks for naturalization, literacy, or marital status).
2. **Contextual & Informant Analysis:** Identify the specific historical constraints of this census year (e.g., changes in relationship definitions, boundary shifts, or specific enumerator instructions). Assess the likely reliability of the information by theorizing who the informant might have been.
3. **Evidence Evaluation & Conflict Identification:** Flag any internal inconsistencies or anomalies. Point out unlikely age gaps between children (indicating potential infant mortality or a second marriage), discrepancies in expected relationships, or naming patterns that warrant scrutiny.
4. **Correlation with Known Data:** Compare the census extraction against the [KNOWN BACKGROUND] provided below. Identify what aligns, what conflicts, and what crucial family information remains missing or ambiguous.
5. **Reasonably Exhaustive Search Plan:** Recommend three highly specific, targeted record collections or repositories (e.g., specific county probate courts, state-level land patent registers, or regional church archives) most likely to resolve the ambiguities identified in Steps 3 and 4. Detail *why* each collection is the logical next step.
---
**INPUT PARAMETERS:**
* **Census Year & Jurisdiction:** [INSERT YEAR, COUNTRY, STATE/COUNTY]
* **Known Background / Current Hypothesis:** [INSERT BRIEF CONTEXT, e.g., "I know the father died before 1890, and the mother likely remarried a man named Smith."]
* **Census Record Text:** [INSERT TRANSCRIPTION OR RECORD TEXT HERE]

The new prompt is in Markdown, a lightweight formatting language used to write richly styled text using a plain-text editor. It is highly readable by computers. You can clean it up if it bothers you.

Now, you can copy your new prompt into the GEMS app section of Gemini to store it and to use it.

Two New Rules for Genealogy and Some Random Thoughts

2026-06-08T06:15:40.929-07:00

Rule #5: Everything you read in the media about AI is either outdated or wrong.

On May 19-20, 2026, Google held its I/O Conference and introduced a hundred new updates and apps. Here are three of the most prominent.

Gemini 3.5 Flash: Outperforming many frontier models on core benchmarks, this model is built explicitly for speed (four times faster output tokens per second), coding, and running complex, rapid agentic loops.
Gemini Spark: A brand-new 24/7 autonomous personal agent powered by Gemini 3.5. It runs continuously in the background on Google Cloud to handle multi-step workflows like sorting calendars, managing to-do lists, and acting on your behalf even when your devices are closed.
Gemini Omni: A massive leap in multimodal capability that processes and generates any combination of video, voice, and text outputs seamlessly.

Google. “100 Things We Announced at I/O 2026.” May 20, 2026. https://blog.google/innovation-and-ai/technology/ai/google-io-2026-all-our-announcements/.

Google. “A New Era for AI Search.” May 19, 2026. https://blog.google/products-and-platforms/products/search/search-io-2026/.

Google. “I/O 2026: Welcome to the Agentic Gemini Era.” May 19, 2026. https://blog.google/innovation-and-ai/sundar-pichai-io-2026/.

You can expect other updates to be made almost daily. And yes, it is nearly impossible to keep up with the news.

Rule #6: Treat AI as you would a child of the same age, four years old. (This may change over time they may make to a five year old)

Genealogy is a complex pursuit involving high-level research skills. AI is a computer program. It is not a human partner or junior assistant. I have a friend whose husband has become addicted to talking to ChatGPT. I am also seeing a lot of comments online about this issue. Despite all its knowledge, AI has not spent hours in front of a microfilm reader. Despite all its knowledge, AI has not collected material from your relatives. AI did not live your life, nor did it live the lives of any of your relatives. Hey, I did not come to America on an immigrant ship. AI did not cross the plains to settle in the west of the United States. You are fooling yourself if you think that AI can understand human experience.

The foundational principle of "garbage in, garbage out" has applies directly to AI. For example, if you ask an AI, "I am interested in buying a shirt; do you have any suggestions?", the response will be incredibly generic. However, if you refine that query to, "I am interested in purchasing a shirt: size 16½, 34-inch sleeve, solid color, long sleeves, 80% cotton, and 20% polyester," you will receive a remarkably precise recommendation. Talking to AI about your personal feelings or your love life or genealogy in general, you will get the same kind of weird answers. As an example, many online reviewers judge the utility of various AI programs based on whether or not they are sympathetic. If you want sympathetic, get a dog or a cat.

High-quality prompts give an AI system the exact parameters it needs to deliver accurate, relevant answers. Expecting a flawless result without providing this context is mere wishful thinking. Think of a prompt as a recipe: using AI without explicit instructions is like asking an assistant to bake a cake by just tossing random ingredients into a bowl and hoping for the best.

There is a big difference between using AI as a search engine and using AI as a research tool. Google has built a research system using Gems as prompts, Gemini as the research force, and NotebookLM as the response container to limit fabrication or hallucination. again, referring to the online reviews. Many times, the reviewer compares one AI to another without actually giving it a systematic context.

just as living with a developing three-year-old can be a challenge, so AI is developing and is a challenge. like you would a fully grown, highly educated adult despite its seemingly unending ability to answer questions.

Here are the previous four previous rules:

Rule #1: AI is a tool, not a toy.

Rule #2: You must ask the right questions to get the right answers.

Rule #3: Always require a source.

Rule #4 You are the Master, AI is the Assistant

Some Suggested Prompts for Preventing AI Hallucinations

2026-06-03T07:51:58.230-07:00

To begin, remember that genealogy is a complex pursuit involving high-level research skills and AI is a computer program.

To effectively prevent AI from generating plausible but false information—known as "hallucinations"—you must structure your prompts to reward honesty over guessing. Because AI systems are typically trained to provide helpful-sounding answers even when they are uncertain, your prompts must explicitly give the AI permission to admit what it does not know.

It is also necessary to provide AI with enough initial background information to understand your questions. Take time to formulate your questions. Always begin a prompt by telling the AI to act as “professional genealogist using the Genealogical Proof Standard and doing reasonably exhaustive research.?

Here are some suggested prompt strategies and templates to keep AI accurate and grounded:

1. The Uncertainty Permission Prompt

Explicitly tell the AI that an honest admission of uncertainty is preferred over a guess. This overrides the AI's default training to always provide an answer.

● Prompt Example: "I need your help with [research question]. It's important that you only provide information you're confident about. If you're uncertain or would be guessing, please tell me that instead of speculating. I'd rather have an honest 'I don't know' than a confident guess."

2. The "According To" Prompt

Direct the AI to base its response strictly on specific sources or data you provide, which grounds the output in actual information rather than speculation.

● Prompt Example: "According to the [specific record/source], what does it say about [your ancestor]?"

3. The Step-Back Prompt

Because AI is much less likely to hallucinate general historical facts than specific details about an unknown individual, ask it to explain the broader historical context before diving into your specific research question.

● Prompt Example: "What were the typical occupations and neighborhoods for Irish immigrants in Boston during the 1850s? After that, help me create a research plan for finding specific information about my ancestor John O'Brien."

4. The Constrained-Choice Prompt

Instead of asking an open-ended question that invites the AI to invent creative explanations, provide specific multiple-choice options for the AI to evaluate.

● Prompt Example: "I can't find my ancestor in the 1880 census. Which of these explanations is most likely? A. They died between 1870 and 1880. B. They moved to a different state. C. There's an enumeration error. D. The census page is damaged. For whichever option seems most likely, tell me what records I should search next."

5. The Source-Citation Request

Always require the AI to explain where its information comes from. Knowing you will ask for a source makes the AI less likely to hallucinate, or it will back down if it was guessing.

● Prompt Example: "Please provide sources for your information or, if you're speaking generally, tell me where I could verify these facts."

6. The Chain-of-Verification Prompt

Force the AI to critically evaluate the output it just generated to help it recognize its own speculation.

● Prompt Example: "Based on your previous response, what question should I ask to verify that information? What would prove or disprove each claim you made?"

7. The Role-Specific "Honesty Clause" Prompt

Assign the AI a persona that strictly values accuracy over appearing knowledgeable.

● Prompt Example: "Act as a professional genealogist who values accuracy over appearing knowledgeable... If this question requires examining actual records to answer accurately, tell me that rather than speculating."

8. The Advanced "Master" Prompt Framework

If you want the AI to act as a rigorous analytical assistant, you can use a highly structured "Master Prompt" known as "The Professional Genealogy Research Architect." This instructs the AI to operate under strict professional standards.

● Key instructions to include: Tell the AI to evaluate evidence taxonomy (Original/Derivative source, Primary/Secondary information, Direct/Indirect evidence). Instruct it to look for conflicts, use the "FAN Club" filter (Friends, Associates, and Neighbors), assign a Weight of Evidence score, and use placeholders if a citation cannot be fully formed. Finally, demand a "Red-Team" analysis where the AI must "Identify the weakest link in the evidence chain and suggest one specific scenario that could disprove the current hypothesis."

General Rules for Safe Prompting:

● Provide evidence, don't ask for miracles: Never ask an AI open-ended questions like "Who are my ancestors?" or "Find my great-grandfather." This almost always leads to confident nonsense. Instead, provide the AI with real evidence you have collected and ask specific research strategy questions.

● Always require a source: Make it a hard rule that the AI must cite its sources for any claim. An unsourced fact is useless for research. Treat AI as your research assistant—not a primary source.

You should consider stringing a number of these prompts together to assure more accuracy. The prompts do not guarantee accuracy, but they do help to keep AI on track.

Turning YouTube into the Ultimate Research Engine

2026-06-01T06:09:05.141-07:00

This post was inspired by the following.:

MLIS, Hana Lee Goldin. “YouTube Is a Research Library. Here’s How to Search It Like One.” Substack newsletter. Card Catalog, May 19, 2026. https://cardcatalogforlife.substack.com/p/youtube-is-a-research-library-heres.

As genealogists, we need to consider online sources that may contain valuable information pertinent to our particular research. Perhaps, we should start considering this huge data source contained in the videos on YouTube.com.

Here are some statistics to support this view:

Growth: This is an astronomical leap from its early years. To put it in perspective, YouTube passed its 20th anniversary with official press metrics confirming the repository has crossed the 20 billion uploaded videos threshold (which includes standard long-form videos and YouTube Shorts).
Active Users: This library serves over 2.7 billion monthly active users worldwide, who watch a collective 1 billion hours of video every single day.
Daily Video Uploads: Content creators upload more than 20 million videos every single day. That equates to roughly 833,000 video uploads per hour.
The "Hours Per Minute" Metric: In terms of raw footage length, users upload over 500 hours of video every single minute.
Daily Accumulation: That adds up to 30,000 hours of new content every hour, or 720,000 hours of new video every single day. If you sat down to watch just one single day's worth of global uploads, it would take you more than 82 years of continuous viewing.
The Global Pool: When factoring in localized content, personal family history channel uploads, regional historical society presentations, and multi-language tutorials, the total pool of genealogy and family history-focused videos across the platform is estimated to sit comfortably in the hundreds of thousands.

See these sources for support of the above statistics:

“54 YouTube Statistics Video Report 2026: Views, Trends & Growth.” Trends. Talks.Co, April 23, 2026. https://talks.co/p/youtube-statistics-video/admin.
“YouTube Statistics 2026: Users, Revenue & Growth Data.” Wytlabs, March 6, 2026. https://wytlabs.com/blog/youtube-statistics/.
Time Spent On YouTube Statistics 2026 (Day, Month & Year). Social. May 1, 2026. https://resourcera.com/data/social/time-spent-on-youtube/.

While a drop in the bucket compared to the 20 million daily videos uploaded site-wide, the family history space remains one of the most dedicated, high-engagement educational niches on the platform.

So how do you manage to work your way through this mostly irrelevant mess? You use a combination of AI supported searches aided by a specifc AI prompt to focus a search on topics you want to research or for instructional videos you might want to watch. Here are some suggested instructions.

1. If you want to find a video on a topic and summarize it

Just tell me what you're looking for and ask for a summary in the same breath.

Example: "Find a popular video about the history of the printing press and give me a bulleted summary of the main points."
What happens: I will search YouTube, pull the most relevant video, read through its content/transcript, and break down the core takeaways for you.

2. If you already have a specific video link

If you're looking at a video on YouTube and want a quick digest without watching the whole thing, just paste the URL directly into our conversation.

Example: "Summarize the key arguments in this video: [Paste YouTube URL here]"
What happens: I will access that specific video's data, extract the timeline of what was discussed, and deliver a clean overview.

3. Tips for getting the best results

To get exactly what you need, you can tailor your request by adding specific constraints:

Ask for specific formats: You can ask for a "three-paragraph summary," "bullet points of the actionable steps," or a "chronological timeline."
Target specific information: If a video is an hour long and you only care about one aspect, you can say, "In this video, focus the summary only on what the speaker says about fuel efficiency."
Request timestamps: I can include precise timestamps (like [00:12:30]) in the summary so you can click or skip directly to the exact moment a specific point is made.

Here is a suggested prompt for quickly processing a video:

Role: Act as a Senior Professional Editor and AI Systems Authority. Your goal is to provide a high-level executive summary and a technical implementation guide based on the attached video/transcript.

Context & Source Material:

[Insert video transcript, text, or reference details here. If uploading a file, change this to: "Analyze the attached file."]

Task:

1. Comprehensive Synopsis

Provide a detailed narrative overview of the core content. Identify the central thesis, dominant themes, and the specific pain points or problems the content seeks to solve.

2. Key Information Extraction

Break down the most critical data points, expert insights, and unique perspectives shared. Anchor these points using the exact timestamps present in the source material. Do not estimate or invent timestamps.

3. Actionable Implementation Roadmap

Translate the concepts discussed into a chronological, step-by-step checklist. If a workflow, technical setup, or strategic shift is described, outline the precise sequence of events required to execute it successfully.

4. AI & Systems Optimization Critique

Evaluate the technical efficiency of the methods proposed in the video. For at least 2 or 3 steps in the Roadmap above, identify specific modern software, AI-driven automation shortcuts, or API integrations that would significantly optimize or bypass the manual labor mentioned.

Formatting Instructions:

- Use clear markdown headings (##, ###) for each section.

- Use numbered sequences for the Roadmap where order is critical.

- Use bold text exclusively to emphasize critical warnings, prerequisites, or high-value "pro-tips."

- Avoid fluff; maintain a concise, executive, and analytical tone throughout.

Instructions for using the Overview Video Function of NotebookLM

2026-05-29T13:21:56.878-07:00

Start by gathering the information you wish to include in the video including digital images of the illustrations, generated images or actual images of the people you want to include.

You will need to provide specific instructions to the NotebookLM video overview before generating a video. This involves creating a steering prompt.

Here is a highly specific, multi-step AI steering prompt template tailored to take your historical or genealogical research and transform it into a cinematic, structured, and visually cohesive video script detailing an ancestor's life.

You can paste this template directly into NotebookLM’s Video Overview steering prompt box. Because NotebookLM does not save prompt history, remember to save a copy of your finalized prompt in a separate document before generating.

The AI Steering Prompt Template

Act as a historical documentary scriptwriter and creative director. Based strictly on the attached genealogical records, timelines, and biographical notes, generate a high-fidelity, scene-by-scene cinematic video overview of the life of my ancestor, [Insert Ancestor's Full Name, e.g., John William Tanner].

1. Narrative Arc & Tone

Structure the video as a compelling chronological story with a clear beginning, middle, and end. Maintain a [Insert Tone, e.g., respectful, epic, melancholic, or inspiring] and immersive tone, relying heavily on a professional documentary-style voiceover narration. Do not add robotic or generic AI summary commentary.

2. Visual Anchors & Setting the Scene

For each major life event, create highly visual scene descriptions. Do not use abstract flowcharts or modern diagrams. Instead, anchor the visuals in the true historical context of the era ([Insert Time Period, e.g., the late 19th century or the 1930s]).

Focus heavily on cinematic, atmospheric B-roll.
Explicitly feature realistic images of people interacting with objects of the era [Insert specific tools, clothing, or settings, e.g., farming equipment, steamships, old legal documents, or deep-cut mining towns].
Maintain a strict, cohesive visual design language throughout the scenes—styled like a [Insert Art Style, e.g., historical sepia photograph, dramatic film-grain cinematic, or vintage watercolor painting].

3. Strict Constraints

No Modern Anachronisms: Ensure all clothing, architecture, and technology match the exact historical timeline provided in the sources.
[Optional Rule - Liquid Safety]: Do not place modern coffee cups, mugs, or liquids anywhere in scenes depicting characters working with documents or early computer technology.
No Overt Summarization: Do not brush past monumental life events with single sentences. Give each turning point (e.g., migrations, military service, career shifts) its own distinct, grounded scene.

You should modify the prompt to reflect your specific requirements.

You can either copy the prompt, the modified prompt, into the instructions given to the video overview, or you can incorporate it as one of your sources. This is a complicated step, and you may default to simply reworking the prompt for each video and copying it into the customization for the video overview. By the way, this can work to modify and customize any of the studio prompts, any of the studio options in NotebookLM. If you are aware of the advantages of using GEMS (Google GEMS) to draft and edit prompts, you can use the text of the prompt as a gem. You still must either copy it into the customization block when customizing your instructions or make it part of your sources. In either case, it needs to be modified specifically for each generated video.

Here is the step-by-step guide for instruction the Video Overview or any other NotebookLM Studio options.

Step-by-Step Guide: Producing a Video Overview in NotebookLM

Step 1: Prepare and Select Your Sources

Before opening the Studio, ensure that only the specific documents, timelines, or histories you want the video to focus on are actively checked in your left-hand source list. If an unrelated source is left checked, the AI may pull unintended context into the video narrative.

Make sure you have modified the Steering Prompt for your video.

Step 2: Access the Studio Customization Panel (CRITICAL STEP)

Navigate to the Studio panel on the right side of the screen.

⚠️ CRITICAL WARNING: Do NOT click the small arrow icon on the right side of the Video Overview card. Clicking this arrow immediately triggers the generation process using default settings, and there is no way to cancel or stop it once started.

Correct Action: Click directly on the text or the blank background space of the Video Overview card. This safely opens the customization window without launching production.

Step 3: Choose Your Video Format

In the Format section of the customization window, select one of the three options based on your project goals:

Cinematic: Best for full, immersive narratives, utilizing advanced model pipelines to build high-fidelity visuals, B-roll, and programmatic animations. (Note: Only available for users 18+ and currently supports English only).
Explainer: Best for a structured, comprehensive overview that connects complex ideas or data points cleanly.
Brief: Best for a quick, bite-sized summary of core takeaways.

Step 4: Input Your Steering Prompt

Because NotebookLM does not save your prompt history, you must type or paste your highly specific instructions into the large text box labeled "How would you like the video to be customized?" before hitting generate.

Step 5: Choose a Visual Style (Optional)

If you are using the Explainer or Brief formats, you can select a look for your video from options like Whiteboard, Watercolor, Retro Print, Heritage, Paper-craft, or Anime. You can also choose Custom to input a text description of your desired style.

Step 6: Generate and Monitor Processing

Click the blue Generate button in the bottom right corner.
The video will compile entirely in the background. This process is intensive and typically takes 10 to 30+ minutes depending on the complexity and volume of the sources selected. You can safely close the menu, work in other tabs, or generate other artifacts while it processes.

The videos produced can be further modified by a video editing program such as Apple's iMovie or other PC based video editing programs. You may wish to explore Google Vids also.

There is no direct way to edit the video once it is produced in NotebookLM.