Monday, June 28, 2010

Evidence Management Diagram Revisited

Last week I wrote about evidence management and the New FamilySearch Tree. The plan this week was to write about Member Trees. I struggled as I wrote. I decided has a piece of evidence management that isn’t represented in my model. It was time to revisit the evidence management diagram.

Here is how it has looked:

An old Evidence Management Diagram

Since I’m hardly an expert on genealogical methodology, my model draws heavily on experts. Elizabeth Shown Mills “Evidence Analysis: Research Process Map” has these basic components:

 The Evidence Analysis Research Process Map

From sources we draw information. From information we choose evidence. The proof of a conclusion lies in a careful analysis of the source, the information, and the evidence.

Upon this foundation, I drew upon my technical background for my contribution. What interfaces (technically, user interfaces and underlying objects) does a genealogy program need to implement this? Desktop genealogy programs already have interfaces for entering sources and for displaying individuals. I came up with two more: an evidence summary, and a conclusion entry interface. Here are the four interfaces juxtaposed beneath the evidence analysis components:

Evidence Analysis plus Evidence Management

The bottom row became the evidence management diagram shown at the beginning of this article. Some of the boxes are displayed more than once to communicate some technical stuff that I won’t bore you with.

Writing about Member Trees, I realize there is another interface that can play an important role in evidence management. I include it in the new evidence management diagram:

 The Evidence Management Diagram

The circular nature of the new diagram evolved (revolved ;-) from the addition of the new Compare interface.

  1. To evaluate a potential source, we compare all the “facts” we believe about an individual with information in the source. If the comparison is favorable, we have identified a new source.
  2. Through a source interface, we enter a citation and other information about the source.
  3. We take information from the source and create an evidence summary.
  4. To help us make a conclusion, the analysis interface displays relevant evidence.
  5. Our conclusion becomes one of the “facts” displayed about a person.

While names may have changed, the function of the red, green, purple, and blue boxes remains the same as before.

What do you think?

  • Are the changes an improvement?
  • Is it easier to understand?
  • Does it meet the needs of newer users? Experienced users? Genealogy program software engineers?
  • Is the circular format appropriate?
  • Have I correctly applied industry terminology?
  • Do the interface names accurately reflect the function of the interface as explained here and in previous articles?
  • There are technical inaccuracies, to be sure. (For example, information comes from the source, not the source interface.) Are there inaccuracies that can be corrected such that the usefulness of the diagram increases?

I have come to depend upon your feedback during this series of articles. After you’ve had a chance to respond, I hope to have the stamina to go back and revise all the previous articles with the new diagram and the new terminology.

Thanks in advance.

With this new model, I am ready to take on… Next time…

Friday, June 25, 2010

Advisory Notices: Evading Emending?

The National Archives and Records Administration (NARA) invites patrons to inform them of quality problems in NARA collections posted on partner websites. Send information to . Officials point out, however, that some problems are “difficult to resolve in a seamless and timely manner.” In such situations, partners are to post advisory notices.

NARA pointed to the database, “U.S. World War II Draft Registration Cards, 1942,” as an example.

When these draft cards were filmed in the states of Delaware, Maryland, Pennsylvania, and West Virginia, the front of each card was photographed above the back of the previous card. For an example, the draft card for John Henry Sullivan is circled in red on the microfilm snippet, below. (For legibility I added larger frame numbers.) The front of his draft card is at the top of frame 1307 and the back is at the bottom of frame 1308.

The Delaware World War II draft registration cards were microfilmed with the front next to the wrong back

As NARA states, has posted an advisory notice in the database description and one in the record display of a registrant from one of the affected states:

Note regarding the images for the states of PA, MD, WV, and DE. These four states were scanned at the National Archives facility in such a way that the back of one person’s draft card appears on the same image as the front of the next individual. The result is that when you click to view the original image, you will see the correct front side of the draft card, but the back of the previous soldier’s card. Ancestry is aware of this problem, and is working to correct this issue.

To put it simply, if you want to see the correct back of a card, go to the next image. How could I tell which direction to go? Among the Wilmington cards after frame 1308 there are a few Kent cards. Compare the registrant's residence on the front with the draft board stamp on the back. But I digress…

To Fix or Not to Fix…

NARA says the notice is to be used when the problem can’t be fixed in a “timely manner.” I assume they expect the vendor to eventually fix the problem. published this collection in May 2006. In the four years since then has published billions of records. Isn’t that long enough to address this problem?

Look at Daniel Sullivan’s draft card from New York. The front of the card and the back are on separate images. Would it be so hard to do the same in Delaware?

Thumbs Up and Thumbs Down

A big green-thumbs up to for keeping the frame numbers. This practice conforms with industry standards I mentioned in archive-quality digital record repository and respect des fonds.

A green-thumbs down to FamilySearch for clipping off frame numbers in their “United States, World War II Draft Registration Cards, 1942.”

Two big green-thumbs up for digitizing additional states—IN COLOR! Wow. That’s all I can say. Wow!

Tim Sullivan Draft Registration Card in color
(Sorry, Tim; no disrespect intended to you or your namesake.
This was the first card I came across with a dramatic ink color.)

A tiny green-thumbs down to FamilySearch because their collection is too…  down, that is. As I write this their collection is down on the Pilot RecordSearch site and hasn’t been published yet on the Beta site. But the green thumb is just a tiny bit down; pilots and betas aren’t expected to work all the time.

Finally, a green thumb up and a green thumb down for

The issue is frame 1309 in the microfilm snippet, above. It’s missing. How do I know? Because did the right thing and included frame numbers. That is the point, after all. Give users the wherewithal to independently detect errors. If you view frame frame 1308 and click to the next image, you end up at frame 1310.

Missing a frame could lead a user to associate fronts and backs that don’t belong together.

But wait. This situation is not that simple. I consulted the microfilm and found that frame 1309 is a repeat of 1308. Had left in frame 1309, users might have associated the front of the green card with the back of the red card!

Frame 1309 is a duplicate of frame 1308

Their solution was to silently remove frame 1309 so fronts and backs were associated correctly. But if a user notices a frame is missing, then without an explanation and without access to the original microfilm, a user must assume that one front and one back are missing, leaving two cards compromised.

The correct solution would be to include an advisory notice on frame 1309.

Here’s my advisory notice to and FamilySearch: Can you see why industry best practices are “best practices?”

Wednesday, June 23, 2010

What About Current Problems? added the missing ship Etruria added the missing ship,
Etruria, to Browse of New York pass-
enger arrivals on 13 November 1893. has not totally fixed the misspelling of Throop township only partially fixed
the misspelling of Throop township.

The National Archives and Records Administration (NARA) recently responded to patron concerns regarding NARA record collections posted by partners such as

NARA takes the concerns raised by researchers seriously. We are working with our partners to improve their digital products, including those produced before the partnership agreement, as problems come to our attention. Our partners want to rectify errors and are cooperating in doing so.

Patrons are urged to report specific problems with partner collections by sending e-mail to .

Already has responded to several cases. In one, added the ship Etruria, missing from Browse for New York passenger arrivals on 13 November 1893 (see illustration, top-right).

In another case, searches in the 1920 census, Throop township, Lackawanna County, Pennsylvania failed because Throop was misindexed as Throap. fixed the search index in a timely fashion. However, has yet to fix the spelling used in the browse menus, district descriptions, and headers above images (see illustration, bottom right).

Officials pointed out that some problems are “difficult to resolve in a seamless and timely manner.” In such situations, partners are to post advisory notices.

Next time, I’ll examine one such advisory notice.

Monday, June 21, 2010

The Evidence Architecture of the New FamilySearch Tree

See “Evidence Management” for an overview of this series and for links to other articles.

Let me show you that the New FamilySearch Tree (NFS) has the architecture needed for evidence management. You may not want to try this at home; I’m a trained professional on a closed course.

Before we dive into New FamilySearch, it will help if you know some NFS technical terminology.1

  • An evidence summary (a green box in the evidence management diagram) is called a persona.
  • An individual (blue box) is called a person.
  • A conclusion (purple box) is called an assertion.

If I specify both my term and the NFS technical term, I will separate the two with a slash, like this:

conclusion (assertion)


To illustrate evidence management in NFS, I again use an example from David Rencher. I created an evidence summary (persona) in NFS for Angeline Clements for each of the five sources, below.

Source Name Persona ID Evidence Summary  
Image 1850 Census Peyton C Clements LWXX-TPN Evidence Summary: 1850 Census Peyton C Clements Link
Image Peyton Clements probate finalized LWXX-5Y3 Evidence Summary: Peyton Clements probate finalized Link
Image 1880 Census W H Goldsmith LWXX-BGT Evidence Summary: 1880 Census W H Goldsmith Link
Image Death certificate A J Goldsmith LWXF-M6T Evidence Summary: Death certificate A J Goldsmith Link
Image Marriage W H Goldsmith LWXF-9ZV Evidence Summary: Marriage W H Goldsmith Link

Click the image link to see a digital copy of the source.

Name (Persona ID)

I included evidence summary names in the table even though NFS doesn’t support them. Instead, NFS uses persona IDs. Remember, a persona in NFS takes the place of an evidence summary.

Evidence Summary and Link

Click the thumbnail to see an image of the evidence summary. Compare a source and summary to see that the two match.

I included the images of the summaries (personas) because once connected—combined as NFS calls it—to an individual (person), NFS provides no way to see just the summary. I included links to the summaries, but once connected (combined), the link brings up the individual rather than the summary.

Creating Evidence Summaries in NFS

With such a perfect architecture under the covers, it is unfortunate that FamilySearch chose to dumb-down the interface to the lowest-common denominator. I think it is possible to adequately serve patrons of all genealogical maturity levels. Because NFS doesn’t give direct access to personas, and because I know that creating an individual (person) also creates an evidence summary (persona), I clicked on Add Information and then Add an Individual That Is Not Connected to My Family Tree.

Then I added the evidence from the source:

Entering evidence for: Peyton Clements probate finalized 
Evidence from the “Peyton Clements probate finalized” source.
Click on this and subsequent images to enlarge.

Since the summary was to be combined with Angeline, I entered only her information. Because it was not obvious how I arrived at a birth date of “Before 25 November 1852,” I clicked on the note icon and entered: “Probate file states Angelina is over 21 years of age.”

For each summary, I clicked on Source details and added a citation. A true evidence summary has but one source for the entire summary. Because NFS doesn’t have true summaries, I clicked the checkbox, Use this source for everything… As mentioned in the past, NFS source templates are inadequate. Consequently, I entered the entire citation in the Comment field, as illustrated below.

Source of evidence summary: Marriage W H Goldsmith
Source for the “Marriage W H Goldsmith”
evidence summary

Once I completed an evidence summary, I clicked Continue. For summaries without death information, NFS gave me the opportunity to enter it.

NFS message: Death info must be added

I appreciate the reminder to enter information that might have been inadvertently missed. But It is unfortunate that NFS insists on adding information—a death flag—even when the source did not contain death information. It seems superfluous, given that NFS is already completely convinced that the person is dead.

NFS then let me review the information.

Review evidence summary: Angeline death certificate 
Reviewing the evidence summary, “Angeline’s death certificate”

Notice that all the information is repeated twice. This is one of the nice features of NFS. I can enter information exactly as it appears in the source document. Below the original text, NFS displays its standardized interpretation so I can see if it understood. Most programs either throw away the original text, or silently—and perhaps incorrectly—interprets it.

If NFS truly supported evidence management I would click Review Possible Duplicates to see if anyone else had already entered an evidence summary for this source and this person. Then I would click Done.

After I clicked Done I saw the completed evidence summary (persona).

Evidence Summary: Peyton Clements probate finalized
The “Peyton Clements probate finalized” evidence summary

Reaching Conclusions

Conclusion entry in NFS for name of Angeline ClementsAs I’ve mentioned, the big payoff for evidence management comes next, when making a conclusion. NFS requires that evidence summaries (personas) be connected (combined) to individuals (persons) prior to using the evidence to make conclusions. The process is messy and reflects the duplication problem that FamilySearch painted itself into. Indulge me if I gloss over it and go right to entering conclusions. Suffice it to say, I connected all the summaries to the individual.

I then clicked on the Summary tab. In the context of evidence management, it could be called the Conclusion tab. The Conclusion tab shows the basic vitals: name, gender, birth/christening, and death/burial. Next to each is a down arrow (pointed out by the mouse pointer in the illustration to the right). I clicked the down arrow and NFS displayed all the values entered into the evidence summaries. I pointed to the one representing my conclusion (pointed out by the hand in the illustration) and clicked it.

Below, compare my suggested conclusion entry (left) with that of NFS (right) for the birth date of Angeline Clements.

Suggested conclusion entry for Angeline Clements birth date  NFS conclusion entry for Angeline Clements birth date

Comparing the two you will notice several shortcomings in NFS conclusion entry:

  • Without a summary name, it is difficult to remember where each piece of evidence came from.
  • The notes I entered in the summary are not displayed.
  • Obviously, since NFS has no provision for recording the attributes of the evidence, none can be displayed.
  • There is no place to enter analysis about each piece of evidence, so other users have no way to know if contrary evidence has been handled.
  • There is no place to enter the overall reasoning used to reach the conclusion.

No Benefits

The evidence management compliant architecture of NFS has given FamilySearch nothing but problems, so I expect they will discard it. One reason it has proven problematic is FamilySearch’s decision to preload millions of junky, sourceless evidence summaries (IGI patron submissions, Ancestral File, and Pedigree Resource File). But I digress.

One reason that NFS users have derived no benefits from the NFS evidence management architecture is that NFS designers failed to give users a way to see whole evidence summaries. Clicking on the  details tab of Angeline Clements doesn’t allow users to see the evidence summaries or even a list of the summaries. Instead, it shows all the evidence interlaced and out of context:

Details of Individual LWFZ-RDZ

Use the Combined records option to see the connected (combined) evidence summaries:

Use the Combined Records option to see NFS evidence summaries

This option gives users the ability to disconnect (separate) summaries (personas) mistakenly connected to the wrong person. When you do this, all the assertions from the source come out in a group. And when the summary is reassigned to the correct person, all the assertions come in as a group.


Because I understand the NFS architecture, I can tap into the strength of its evidence management. But doing so is painful. FamilySearch users have not benefitted from their architecture for two reasons. First, since FamilySearch preloaded garbage into NFS, garbage is what they’re getting out (GIGO). Unfortunately, user interface and architectural decisions are now being driven by the resulting ripples. Second, FamilySearch chose not to provide a different user experience to immature users and mature users. FamilySearch then had to simplify the user experience by hiding the existence of evidence summaries (personas).

To conclude this presentation of the New FamilySearch Tree architecture, let me say that it is extremely impressive and ideally designed for evidence management. Hopefully, FamilySearch will one day leverage their technical superiority by opening up persona management—or, as I call it—evidence management.


     1.  Rob Lyons, “Family Tree Combine/Separate,” FamilySearch Developers Conference, 2008; online archive, “Recorded Presentations,” FamilySearch Developer Network: for Software Programmers ( : accessed 18 June 2010). Also “Glossary,” FamilySearch Developer Network.

Friday, June 18, 2010

NARA Responds to Issues

Microfilm Documents Missing

The National Archives and Records Administration (NARA) recently responded to accusations that posts NARA record collections with numerous quality problems, including missing documents.

A spokesperson for NARA wanted to make it clear that has unilaterally digitized and published more than 300 NARA microfilm collections with over 70 million images prior to entering into a contractual relationship with them. digitized, indexed, and placed these images online using NARA microfilm publications that are available to anyone by purchase from NARA.  This was strictly the work of Ancestry[.com], with no involvement, oversight, or quality assurance work by NARA.

NARA has posted a list of all their collections online at partner websites, whether the collection was produced before or after their agreements. The list is at and will be updated on a regular basis. But I digress…

At a digitization facility in Silver Spring the two have started digitizing records per the contractual arrangement. NARA preps records for scanning, does the scanning, and NARA conducts quality control checks. According to NARA,

Both staffs ensure that every page has been imaged. NARA does a page-by-page quality control check on 5% of the boxes scanned. If a problem arises, mistakes are rectified immediately and the percentage of review on that camera operator’s work is increased. An operator must image two consecutive boxes perfectly before the audit returns to the 5% level.

For a recent project, 9 boxes out of 130 were checked. The highest error rate for any one box was 4 pages missing for every 1,000 pages. Missing pages were immediately digitized before processing continued. The overall error rate for all the boxes reviewed was 7 missing pages out of every 10,000. Most missing documents hadn’t actually been skipped, but scanning failed to pick up a light stamp or mark on an otherwise blank page.

“NARA considers that digitizing thousands of documents and having them available online with unprecedented indexing is worth the small percentage of error.” As one attempts to drive the error rate to 0, the cost explodes exponentially. In other words, dropping the error rate from 0.07% to 0.007% might cost 100 times as much, and dropping to 0.0007% might increase costs 10,000 times.

While NARA seems to feel that the current quality/cost ratio is acceptable, a spokesperson made it clear, “NARA does not want errors.” Next week I’ll tell you what NARA officials recommend you do when you come across problems.

Wednesday, June 16, 2010 Missing Documents

Microfilm Documents Missing Staff at the National Archives and Records Administration (NARA) recently responded to accusations that posts NARA record collections that are missing documents.

Earlier this year at the Annual Bloggers Day Todd Jensen briefed us on their new NARA scanning facility in Washington D.C. (I alluded to the presentation in one of my articles. I was waiting on for photographs before I wrote my article. Now I probably can’t remember enough to do the presentation justice. But I digress…)

At that time I asked if still dropped images from NARA collections when they published them. Andrew Wait assured us that their policy has always been to publish every image. Another employee in the room (I don’t remember who) leaned over and whispered some circumstance in which they had dropped images.

My hearing isn’t all that sharp, so I didn’t hear the circumstances mentioned, but it is well known by most subscribers that has always done so. seems to feel they are doing everyone a favor by chopping and dicing up census microfilms:

  • Dropping images with no legible names:
    • microfilm headers
    • NARA publication booklets
    • covers
    • census totals
    • blank forms
    • pages that can’t be read because they were imaged too dark, too light, or too blurry
  • Rearranging census districts according to alphabetical jurisdictions
  • Preventing going past either end of a group of images

These changes are perfectly reasonable to decision makers that dabble in genealogy just enough to be dangerous.

And, in fact, these changes might indeed be an improvement if also allowed unimproved access. The former without the latter has serious repercussions:

  • Removing context removes information
  • Tampering with evidence decreases its evidentiary value
  • The changes rob users of any way to detect documents that were inadvertently dropped
  • Removing illegible images gives NARA staff members no way to know that access to the originals is warranted

For these reasons, members of the Association of Professional Genealogists (APG) have criticized’s practices. Last year Peggy Reeves pointed out that all but one of the first 25 soldiers from roll 402 are missing from’s publication of T-288, General Index to Pension Files, 1861-1934. If allowed unimproved access to this NARA publication, Reeves would have discovered one of two things. Either the images were illegible, or had inadvertently left out all the index cards from “Charles Roe” to “Allen Rogers.”

That’s a lot of missing documents.

Digital publishers might want to take a lesson from microfilming practice. FamilySearch (as the Genealogical Society of Utah) always filmed every document, but when an original document was illegible, included a label indicating “Illegible Original.”

Next time I’ll share what NARA had to say about all this.

Monday, June 14, 2010

Genealogist or Gossip

If they don't depend on true evidence, scientists [including genealogists] are no better than gossips.

Penelope Fitzgerald

In the example presented in “Evidence Management Explained” we saw that the big payoff of using evidence management occurs when a conclusion is drawn. Drawing from the example of Angeline née Clements Goldsmith again, suppose a week or more passed between discovering and entering each piece of evidence. As each piece of evidence about her birth is entered, we might well revisit our conclusion about her birth date. Each time we click to edit the birth date, the evidence manager displays a conclusion entry window.

For convenience I have reproduced the conclusion entry window from the example. I have incorporated most of your feedback. If I didn’t get yours, remember that this example is only conceptual.

Summary Name Asser-tion Evidence Notes Created Date Link Analysis
Automatically Selected Evidence
1850 Census Peyton C Clements Age 2 Image copy of federal copy 1850 Source The earliest record; at just two years of age, it is highly likely that the 1850 census correctly implies 1848.
1860 Census P C Clements Age 12 Image copy of federal copy 1860 Source Next earliest records agrees with 1848
1870 Census P C Clements Jr. Age 18 Image copy of federal copy 1870 Source New orphans with all birth dates wrong suggests a 3rd party supplied the data
Peyton Clements probate finalized Age Over 21 years Image of original. Primary information 25 Nov 1873 Source 1848 and 1850 are consistent with father’s probate record
1880 Census W H Goldsmith Age 25 Image copy of federal copy 1880 Source Census ages ending with 0 or 5 are suspect
Death certificate A J Goldsmith Birth date 5 Feb 1850 Image copy of original. Secondary information 1939 Source There is no reason to doubt 5 February even though the 1850 is not possible according to the 1850 census
Gravestone A J Goldsmith Birth date 1850 Secondary 1939 Source Likely same informant as death certificate
Manually Selected Evidence
Marriage W H Goldsmith Marr-iage Date 15 Jan 1873 Explicit 1873 Source Birth from 1843-1858 is likely.
1850 Census Peyton C Clements Sibling Eleanor Age 1 1850 Source To have a 1 yr old younger sibling in 1850, Angeline must have been born in 1848.
Conclusion for Birth date: 5 February 1848
Reasoning: It is clear that the earliest records have the correct birth year. While there is no collaborating evidence for the day and month, there is currently no reason to doubt it.


How do the vendors do with conclusion entry? Here is what I found:

Conclusion entry facilitates intelligent conclusions by design. No No No
Pertinent assertions from all relevant evidence summaries are gathered together and displayed in one place. Yes-ish Yes Yes (but not grouped together) if a person page is considered a conclusionary person, but no if the person page is an evidentiary person
To encourage critical thinking, notes can be entered for each piece of evidence and for the conclusion. Yes-ish Yes-ish Yes-ish
Evidence is automatically selected for analysis based upon the conclusion type. For a birth date conclusion, evidence about age and birth date are automatically displayed. No Yes Yes (but not grouped together) if a person page is considered a conclusionary person, but no if the person page is an evidentiary person
The user can manually select other evidence. No No All evidence is displayed in all cases.
Attributes are displayed for each piece of evidence and its source. For the source, these might include: original or derivative, derivative type, recording date, and recorder. For the evidence, these might include: informant, primary or secondary, direct or indirect, and supportive or contradictory. No No No


I realize these tables are barely better than nothing. I need to display some screen shots from each of these products to illustrating what I’m talking about. I had planned to start those today, but articles for Wednesday and Friday caused a digression and my weekend is over. Stay tuned…

And just so you (the vendors) know, I’m always open to answer questions you may have regarding evidence management… no extra charge.

Saturday, June 12, 2010

Rrrrr -- SPAM Comments -- Rrrrrr

I apologize for the SPAM comments that are being posted to my website. I know some of you subscribe to comments and have to wade through them.

This has forced me to enable a feature on my website that requires me to review each comment before it is published. Unfortunately, I do much of my Insider work on weekends so that it doesn't infringe on my employment. That means it will be difficult to review your comments in a timely manner.

I am very grateful for your comments. You are an amazing group and happily I learn from what you have to say. Hopefully this will be a temporary change so that you can continue to respond to other commenters in addition to responding to me.

This change is effective immediately.

-- The Ancestry Insider

Wednesday, June 9, 2010

No, You’re Not Traveling Through Time

While it’s true that the Bloggers Day was held back in January 2010, I never got around to putting together a table of contents to all the articles I wrote about it. I’ve rectified that with the other article you’ve seen published today: “ Bloggers Day 2010.”

If you caught all the articles as I published them, you can ignore this. Otherwise, I hope you find it useful. Bloggers Day 2010

Click each of the following links to read about each presentation made at’s Bloggers Day 2010.

Monday, June 7, 2010

Close, But No Cigar

FamilySearch Places Records Into Folders 
In the New FamilySearch tree, records
are combined into folders from which
summary values are selected. Image
courtesy: FamilySearch International.

See “Evidence Management” for an overview of this series and for links to other articles.

Vendor Support of Evidence Summaries

Each vendor has some features that tantalizingly approach evidence summaries.

The new FamilySearch Tree is close. Evidence summaries are placed into a folder for an individual from which users choose a conclusion (they call it a summary value). FamilySearch is the current technology leader. But then they preloaded evidence summaries with oodles of worthless, source-less, secondary informational, derivative sources: Ancestral File, Pedigree Resource File, and user-submissions to the International Genealogical Index. This, in turn, generated IOUSs, which in turn crashed their servers.  Further, they surfaced none of an evidence summary’s usefulness to users. Consequently, they see no advantage to their technology leadership.

FamilySearch has announced their intent to separate artifacts from individuals. This moves them in the right direction. But because their technological advance has given them nothing but problems, they will be tempted to abandon it entirely. Keep your fingers crossed.

Footnote is perhaps closest. They have created evidence summaries for individuals mentioned in collections such as the Social Security Death Index and the 1930 Census. They have created person pages that allow users to store conclusions. However, they have positioned the two kinds of pages as equivalent. Consequently, users are frustrated because there is no ability to combine the two.

Footnote should slap an “Evidence Summary” moniker on their evidence summary person pages. And they need the ability to attach an evidence summary to a user contributed person page in such a way that the person page inherits evidence from the evidence summary. Inheritance also allows users to inherit from other users’ person pages. This allows users to share their contributions without worry that other users will modify their pages.

Features Ancestry
Evidence summary stores abstracted evidence separately from conclusions No
Yes. The NFS tree documentation uses the terms record, folder, and summary to explain the evidence summary (green box), individual (blue box), and conclusion (purple box), respectively. All a person's records are placed (combined) into their folder and users choose a summary value when conflicts exist.
No for user contributions. No distinction is made between records and folders.
Yes-ish. Duplicate person pages come close, but there is no concept of an evidence summary person page versus a conclusion person page.
Evidence summary stores abstracted evidence separately from sources Yes-ish for provided sources. Yes for preloaded. Yes
Each piece of evidence in a summary is categorized by the assertion type (e.g. name, gender, age, birth date, birth place, marriage date, place, death date, place, burial date, place, and so forth) No Yes for preloaded. Yes
User can view and work with an entire evidence summary No Yes when combining or separating records. No
User can give descriptive name to an evidence summary No No No
Can generate a list of evidence summaries No No No
Can sort and filter lists by name, subject, source, evidence creation date, informant, evidentiary weight, etc. No No No
Evidence summary is linked to source (entry) No Yes for preloaded. Yes
User can characterize evidence as primary information or secondary information, supporting or conflicting, direct or indirect. No No
User can enter notes about a piece of evidence No No Yes-ish