At the moment I mostly use Gramps as an archive. All my active research and downloading of sources/documents happens in Zotero, Obsidian, Foam, and the network‑graph and timeline tools I mentioned earlier, completely manually. I’m also working on a project involving OCR and importing OCR/AI‑generated text from sources into Markdown and then into other tools like Gramps.
Be aware that I use Obsidian and Foam far outside what these applications were originally intended for. I often work with Markdown tables containing thousands of lines, set up with wiki‑links to every referenced object — for example census data or a ship’s manifest. A typical manifest table will include wiki‑links for port names, ship name, owner, captain’s name, every sailor, the date, the cargo (if listed), and any other relevant information that might lead me to the next step in my research. I also add anything of general interest in case these datasets can later be shared for other types of social or historical research. So my project goes far beyond traditional genealogy.
If you search for my username here in the forum, you will find comments where I explain some of this in more depth.
Answers to your questions:
-
Yes, I often share notes or attach the same note to multiple entities. In Obsidian I use wiki‑links with aliases in the format [[{Physical document}|Alias]]. I also keep a “Notes” section in my Markdown files where I copy all links. It’s double or triple redundant information, but it gives me the best personal control.
-
I use family notes the same way I use notes for any other object in Gramps or in my Markdown hierarchy — mostly for information that cannot be attached in any other way, or for research documents, links to relevant material, or anything else connected to that Gramps object.
-
My workflow is more historically oriented. I started with genealogy, but at this point it has become a research project on the Norwegian mercantile fleet, with genealogy as a “by‑product”.
I almost always begin by downloading documents and datasets (sources) into Zotero and saving them as linked files to my Zotero items. In Zotero I maintain an extended collection structure that mirrors my folder hierarchy: one collection/folder for each “object”, and a hierarchical structure for sources based on Country → State → Place → Repository → Source → and then the internal structure of the source itself. In Zotero I create virtual copies of each source for every object it is relevant to.
These files are currently stored mostly outside my Markdown vault because I used Zotero long before I discovered the “non‑standard” workflows possible in Zettlr, and later in Obsidian and Foam.
I do all the wiki‑linking manually or semi‑manually. For example, I open Markdown tables in VS Code and add wiki‑links to all columns that are relevant. I only create new “Object Notes” when needed — not for everything in a source, but for what is relevant at the moment. However, I always save and transcribe the complete source, for example all rows in a table.
Then I create a media object in Gramps, or a local file link in any software that supports it. (This is something I’m trying to semi‑automate or fully automate in a Python script project I’m working on, which will eventually handle my entire workflow.)
So yes — Obsidian and Foam are my primary research tools, or what I prefer to call my “hub” that connects all my other software. This allows me to do all my work in one place and later run a sync or export to other formats once my Python scripts are ready (maybe even a Gramplet at some point). But I’m not a developer/programmer, so things take time.
I also use Markdown because it allows me to create hierarchical structures for events. I can link events to places and link people as participants, instead of the lineage‑linked approach that forces you to attach events directly to a person — which is historically inaccurate for many types of events.
A second reason I stick to Markdown is that it can be viewed, read, and edited by any editor. The data remains truly “open data” for the foreseeable future.
-
I think that as a “first version”, the privacy flag only needs to be added so users can decide for themselves. But maybe later it will be possible to add more automated control. If you include the YAML frontmatter key from the start, it will be easier to extend the functionality later if you feel the need. Personally I do not use the flag/tag because I do everything genealogy‑related manually, and I choose very carefully what I share — if I share anything at all.
-
In Obsidian I am looking at the third‑party “export to website” plugin (I never remember the name). I do not use Obsidian Sync, but I might start using some other sync tool at some point. My problem right now is having total control over what leaves my computer (even though I am a Windows user… hahaha). So again, I think the most important thing at the moment is simply to add the privacy flag as a YAML frontmatter key.
All of these things we talk about here, I try to keep as general as possible so they can be implemented both for general use by people who do not use Gramps, as well as for more advanced usage — including GEDCOM‑related workflows or for people without any earlier experience with genealogy. And that is why I think your project has such high potential: just like Gramps, it seems to offer a lot of possibilities and flexibility.
I have not yet started to look at your plugin, only because I need to create a new test vault for it, but I will do so sometime early next year. I clearly see what you are trying to accomplish with it, and I seriously applaud your effort.
And I actually think you can get more great answers in this forum, because several people have asked about my usage of Obsidian earlier.
People might get more engaged after the holidays… 
—
Note: This text was originally written in Norwegian and translated into English with the assistance of Copilot. Some sentences were lightly adjusted for clarity and readability in English, but the meaning and intent remain faithful to the author’s original Norwegian version.