How do you manage your research contacts?

I like to add distant relatives to my research tree, filling in the barest outline of the divergent branch to connect us. (The Data Entry Gramplet is great for quickly sketching in the missing bits. FamilySearch, WikiTree & WeRelate all have a … ummm… QuickView … delineating a direct genetic connection. I just have to find the raw end in my tree and outline.)

It doesn’t matter if the connection is unconfirmed (or even absolutely wrong)… my Gramps data is a RESEARCH Tree for managing overwhelming number of leads… not my Proven Tree. I collect any & all information and winnow out the wheat from the chaff later. (I wish the Confidence feature extended symmetrically into the negatives and for all objects. I currently record disproven Events but have to use Disproven custom Roles for this purpose. It saves having repeatedly retrace disproving logic when the same bad ‘fact’ pops up again.)

I add a qualified Occupation (like ‘Genealogist, hobby’ or ‘Genealogist, net published’) to these distant cousins who share the obsession. And have a Custom Filter to run after making a brickwall breakthrough. It reminds me of living researchers to contact who are interested in that brickwall.

When eMailing those contacts, I add a Note with the message text & apply it to the list of those who have been carbon copied. (Or BCC’d for the few who have requested protecting their privacy.)

What other ways do you use Gramps to remind you to do structured sharing of leads?

1 Like

@emyoulation, thank you for starting this thread. It gets at several things I’ve been thinking about lately.

Do you maintain a separate proven tree? If so, how do you manage the transfer of data from your research tree into your proven tree? Or do you maintain only the research tree, and use on the Confidence level of citations to somehow create a filtered “proven” subset? I, too, collect many source records that may or may not turn out to be relevant, but I tend not to add people and their records into my tree until I’m fairly confident they belong there.

I’ve wondered how to handle situations where I reasonably expected to find a record for a person in a given source, but did not. To avoid repeated searches in the future, maybe I should create a “negative” citation in which I enter “NOT FOUND” in the Volume/Page information?

Do you have a similar method for tracking disproven parent/child relationships? For example, I have a great-great-grandmother whose parents are unknown to me. I often see her in other trees with incorrect parents, and I have posted the disproving evidence on Family Search. I have not entered those incorrect parents in my tree, but maybe I would if I had a good way of indicating the disproven relationship.

How to do you mark something as a brick wall, as distinct from the many other things that are simply waiting to be explored further? And how do you track which other researchers are interested in a particular brick wall?

2 Likes

If I ever learn to program python, the very first thing I would like to add would be a real research panel, where you could make research logs and add contact information, communication and other things thats relevant to the research, but maybe not so to the tree itself, other than as a history…

And I also really miss a way to register Disproven Research
At the moment I register all that in Zotero, Adding a Disproven TAG to document or source that do not give me a result…I have created a “template” system for my self in Zotero
Where I add and move any entries to the different collections as my work progress goes on…

I also use a few other tools for research, because no software are complete…

1 Like

There are lots of ways to mark your brick walls.

You could use Tags… but I like to reserve those for extremely Temporary use because a long Tag list is overwhelming. (And SLOWS down Tag use.)

You could use a shared brickwall note. But while that’s great for keeping a harmonized master notebook, it is not very visible. Sharing an image is something that would also show in a lot of the Charts & visualization tools. You could use an animated GIF like:
https://i.gifer.com/2zzW.gif or https://media1.giphy.com/media/h8DLTFQf4c2Xr8PhWy/200.gif. But I prefer an icon sized GIF instead. Looking at the References tab for that Media object is your list of brick walls.

ToDo Notes are the general mechanism for manual task management. I also reserved the highest ranking 3 ToDo notes for higher priority purposes.

The ToDo Gramplet lets you scroll through your list of ToDos but how it decided which are at top of the list is opaque & baffling. So I created 3 additional ToDo notes and transcribed the arbitrary top 3 then shared to their References. That allowed me to blank the top 3 for re-use in other purposes.

There’s an experimental ToDo Gramplet but I haven’t found a viable workflow for it. It’s too quirky.

As for tracking people with special interest, that’s just a People filter… looking for living descendants of a person who have a ‘Genealogist’ occupation. Because the descendants can be a long processing filter, I tend to use a temp Tag. It allows me to expand or strip the list. (I might want to start with the parents instead to include 1st cousin descendants. Or might expand the descendant to include genealogists within a couple degrees of separation from the genetic descendants.) After finding the Living Genealogists among the Tagged people, I’ll send those people a notification & flush the temp Tag.

For historically significant ancestors (who have unrelated researchers or where a professional genealogist was hired) I’ve added a Note to the ancestor… with that Note shared to the unrelated Researcher’s profile.

Some of the Genealogist have a DoNotContact custom attribute and so I filter them out. (They’re just people who don’t like to accept help, are needy, are disagreeable, or are leeches. Not because I have any ‘restraining order’ custom attributes… yet.)

2 Likes

You can still add the false parents. In the child’s reference editor, you could make the record private and add all the relevant documentation and notes that disproves the relationship. The private would prevent the relationship from being printed or disseminated thus preventing furtherance of the falsehood. Although not marking it as private will allow your disproving research to be displayed.

The problem would what cues do you add to the “parents” that would display in any unfiltered views. I do not enter records with capitalized surnames. Maybe entering these “parents” capitalized may be enough of a cue. You would have to use the surname override to get the person grouped in with the non-capitalized “relatives”. After all, in future research you may find that these “parents” are actually an aunt and uncle.

Just an idea.

I often will find a wife with just a given name. Other researchers claim she is of one surname or another. When there is evidence that she is NOT the women with the surname, I will create an alternate name with the Type “Person is NOT” with all of the citations and notes. These I do not mark private while the surname is still in question.

2 Likes

The thing about a Proven Tree is that it is a snapshot of a fleeting moment in time.

I use a custom attribute to flag records that are ready to Share or Publish. When I reach a cut-off point, I will export only records with that attribute. Then move the export to a fresh machine with a new download of Gramps. A new Person is created as a container for production notes and storing any critical .ini settings BEFORE importing. Then I import & generate my publication from that Dated data file.

After the publication is complete, the exported file is archived along with the Tree & rest of the job material. The ‘Proven Tree’ file & Gramps is stripped from the computer. (If the job is particularly rigorous, an archive restoration is tested on another box before flushing.) And I go back to the overflowing Research Tree on my original system.

Yes, it is painfully meticulous to validate & set an Attribute on every item before it can be allowed in publication or shared for distribution. And, yes, it is WAY more than needed for amateur genealogy. But it follows a workflow I adopted 20 years ago for releasing shrink-wrap COTS products. Quality Control was added to my responsibilities after several sequential releases had recalls. There had always been delays due to production errors. This process tripled the initial QA burden but unexpected delays & recalls went away. There were only 3 times we needed to reproduce a discontinued product from archive and all 3 went smoothly.

I’ve only done a small distribution with a Gramps Tree once so far. The process still needs work.

3 Likes

I should have added that I also have the name Type “Could Be” when a name is not clear.or records do not agree.

1 Like

Earlier in this thread, a question is raised of how to record Postulated Theoretical families or lineages.

There is an absence of a functional system that supports Negative Confidences to “fully disproven”. The current GEDCOM-based (null,0-to-3 indexing levels) “Quay” system is merely decorative as implemented. (Because EVERYONE acknowledged the Quay “standard” is useless, it is dismissed as a foundation for development.)

As described above, I abuse the Roles feature in Events to record “Disproven” events. (Only some features check for Primary role when doing analysis.

Is there a way to twist the Family containers in the same way? Where Gramps could create Ghost Family objects to outline (but normally ignore) complete lineage postulations? Moreover, to groups the Ghost Families (as a named conjecture) that temporarily Reorders Families so that a normally hidden set of Ghost Families take precedence as the Preferred lineage analysis/printing/diagrams? (Which necessitates temporarily demoting the normal Primary Family to a hidden ghost state.)

In this way, I might have an “Alfred E. Neuman conjecture from Incestry.com” conjecture which can be contrasted with my own postulation AND possibly against the primary conjecture of Families validated with high confidence.

Further, since sometimes only a person or 2 differs in a conjecture (which of 3 possible Martha IDs is the mother) some sort of CSS would be ideal: based on Family 33, replace mother Mary1 with Mary2 & null out child John16.

I’m pretty confident that Gramps cannot be forced to do such contortions now. Ghost Families would undoubtedly create loops in the Tree. And they would give the Relationship Calculator conniptions