You may also find some older one, which I suppose are also good clients for vector scaling (real geometrical maps on 18th!).
etc.
This is another perfect example for why Gramps should support Events on Places.
Just to understand your point better: event/place is already a many-to-one relationship, what would “events on places” add? You mean many-to-many like the ship voyages that have been discussed before?
Probably less ambitious, but I saw this Reddit post on a new development to superimpose two cadastral planes, old and modern, to find the relationship between the old plots and the current ones. If it can help…
It looks like a map inset could be pinned into map tiles fairly easily with their project too.
I regularly acquire a cemetery’s section layout diagram that I’d like overlaid on the map tiles.
You could see the article on this post too
First of all, to be absolutely clear: This is not a criticism of the Domus project at all.
I think Domus, Wikidata, and the OpenHistoricalMap integrations are a great idea and a needed tool in historical research, but not everyone wants to or can work online.
Instead, this is another reference to why Events on Places really needs to be reconsidered as a core feature inside Gramps.
We are seeing a growing demand for this, even if people don’t always see the underlying connection, and the rise of external online platforms for house and local history proves that the current person-centric model is missing a vital dimension.
Let me ask a hypothetical question to clarify the core logic here:
What is your project really about? Is it not the history and events tied to a house or a residence? And is a house not fundamentally a Place?
When we look at history through a local or genealogical lens, so many major events belong inherently to the Place, not to the Person. The people are simply participants. Let’s break it down with a few concrete examples:
- Living somewhere vs. A Place being Inhabited (The Biography of a House): Buying a house or moving to a new address is a person- or family-event. However, the construction of the house is a place-event because it physically alters a specific piece of land. Similarly, residing somewhere is a person-event, but the fact that a building is inhabited is a place-event—a chapter in the building’s own history. If the house is later demolished, that demolition becomes a sub-event of the land’s history. Over the lifetime of a house, dozens of unrelated individuals and families might have lived there. Even if one single person lived there for the building’s entire existence, it remains a part of the building’s biography—making it a place-event with participants.
- Occupation vs. Actual Work (Labor): Having an occupation (e.g., being a blacksmith or a factory worker) is a personal attribute or event. However, the work being performed—the daily operation of the forge, the building of a bridge, or the production at a mill—is a place-event. If your ancestor didn’t do that job, someone else would have been hired to perform the exact same work at that exact location. The labor history belongs to the place.
- Education vs. The Course of Study: Attaining a degree or receiving an education is a person-event; it follows the individual. But the class, the curriculum, or the specific study program run by a university or school is a place-event. The course exists, the lecture hall fills up, and the professors teach regardless of which specific students happen to sit in the benches that semester.
- A Journey (Maritime History): A ship voyage over the ocean or up a river is a dynamic place-based event. It consists of multiple sub-events (departure port, stopovers, arrival port). One or more ancestors might be participants in this journey with specific roles (captain, sailor, immigrant passenger). The journey itself belongs to the history of the vessel and the route (the places), but it also serves as a sub-event to a person’s or a family’s broader migration history.
- A Military Campaign: A military campaign is a place-event, not a person-event. A campaign moves through a landscape, and individual battles or skirmishes take place at specific geographical locations. Hundreds of individuals in your database might have participated, each with different roles (soldier, medic, casualty), but the campaign’s timeline exists independently of any single soldier.
- Censuses and Taxation: Historically, a census or a property tax is tied to the land or the building. A census is carried out at a specific place regardless of who happens to live or stay there on that exact day. The building experiences the census; the people are just the recorded inhabitants at taxation or survey checkpoints.
- Natural Disasters and Epidemics: A forest fire, a flood, or a localized epidemic are strictly place-specific events. A house burns down or a village gets quarantined. The humans are merely participants—either as victims, survivors, or infected individuals. The disaster happens to the locality.
Regarding the discussion about Main-/Sub-events (or Superevents):
No, Place Events and Main/Sub-events are not the same thing, but they are crucial parallel functions required to tell a complete story. A macro-event can contain many sub-events that are a mix of deeply personal experiences and strictly place-specific occurrences.
- For instance, contracting a severe illness or injury is a deeply personal event. However, being treated at a specific hospital, moving to a completely different location for rehabilitation, and eventually having to physically modify one’s residence—these sub-events are anchored to specific places. The rehabilitation facility and the hospital exist and operate independently; the person is merely a participant (a patient) at those locations.
- To draw a parallel to the classic “Gender vs. Sex” discussions on this forum: The internal process of finding oneself and one’s identity is an entirely personal journey. But the medical consultations, the surgeries performed at specific hospitals, and the journeys made to a specific place for convalescence—these are place-specific events. The hospital or the convalescence home would have treated another patient if this specific person hadn’t been there.
Historically, it is a place-event because many individuals can receive the exact same treatment at the exact same location during the same time period.
The person is a participant (a patient), but the event belongs to the history of the place.
If we take this into a broader historical context:
As you can see from these examples, an event is not always a person-focused event. Especially when we want to expand our research into demographic studies, farm/estate histories, military campaigns, migration history, or—as you are doing in this project—the history of houses, we fundamentally need Places to be the primary holders of events, e.g “Events on Places” as short.
Once a Place holds the event, individuals, or even other place-related objects like ships, trains, or historical caravan wagon trains, can be dynamically added to these events as participants with specific, contextual roles.
When we try to record these things in Gramps today, we are forced to create some kind of workarounds, like adding a census or a land-tax regime as a personal attribute or a simulated event on a person.
Ultimately, Places as primary holders of events and Main-/Sub-events are two distinct, yet equally important concepts.
They must work hand in hand to successfully document complex, multi-layered histories and chronological sequences of historical events.
If Gramps natively allowed Events to have Places as their primary object, we wouldn’t need to reinvent the wheel externally. We could map out the entire historical, architectural, and social context of a location natively, and simply link our ancestors to these events whenever their lives intersected with them.
Note: This text was drafted in Norwegian to articulate the core concepts and examples, and translated/edited into English with the assistance of an AI collaborator for structure and clarity.
Sorry, I still don’t quite understand.
Event has .place, and db.find_backlink_handles(place_handle, ["Event"]) returns all events linked to the place, whether they are linked to a family or not. I don’t see how that relation is different from Wikidata P276 (location).
What we don’t have is metadata on the link, like an event_ref_list on Place that supports citations or notes. Is that what you’re missing?
No, it is not just about metadata or citations on the link, even though having typed and cited relations would always be a very welcome bonus and upgrade for any relationship in Gramps. It is about a fundamental shift in who owns the event—and to be absolutely explicit: this is not about changing or replacing any of the existing structures in Gramps, it is strictly about adding a missing dimension. It is about moving away from looking at a Place solely as a passive attribute to a person’s life.
When Gramps runs db.find_backlink_handles(place_handle, ["Event"]), it treats the Place as a filter or a text tag attached to a person-event. In historical and demographic research, a Place is not just a passive backdrop—it is an active historical entity with its own independent timeline.
Let’s look at why the current database model breaks down in actual research:
- Floating events vs. Place ownership: Technically, you can create a floating event in Gramps and assign a place to it without linking it to any person. But this event remains a detached, orphaned object in the database. It has no primary ownership. For house history, the event shouldn’t just float in a vacuum—the Place itself should be the primary owner and holder of that event, forming the core timeline of that location’s biography.
- The event belongs to the Place, not the Person: When a hospital treats 10,000 patients during an epidemic, the epidemic treatment operation is an event historically owned by the hospital (the Place). The patients are just participants. If you use the current person-centric model and link all those thousands of separate individuals to that event, it creates a significant logical error: the system will make it appear as though all these thousands of people have a direct personal or historical relationship with each other. In reality, they don’t. They only have a shared, independent relationship with the place-centric event. Forcing a person-centric model here completely skews the relational data.
- Wikidata P276 (location): In Wikidata, an event is its own independent entity (a separate item), and P276 simply points to where it happened. But in Gramps, an Event is structurally treated as a sub-object designed to fill the timeline of a Person or a Family.
So to answer your question: What is missing is not just notes or citations on a link. What is missing is the ability for a Place to be the primary object/owner of an Event, where people (or mobile objects like ships and trains) can be added as participants with contextual roles—exactly like you add participants to an event today.
The broader genealogical research world consists of event-oriented research, document-based research, and place-oriented research, alongside traditional lineage-linked tracking. When recording the biography of a house or a locality, having the Event natively owned by the Place is the only way to avoid complex database workarounds.
Note: This text was drafted in Norwegian to articulate the core concepts and examples, and translated/edited into English with the assistance of an AI collaborator for structure and clarity.
I have been following this discussion with interest not that I have any curiosity in following the ownership of buildings or plots of land more in that it highlights a fundamental dichotomy in GRAMPS development.
On the front page of the GRAMPS website we have GRAMPS Genealogical Research Software
Definition of Genealogical Research
Genealogical research involves examining historical records to extract information about individuals and families, aiming to trace ancestry and kinship ties. This process is essential for understanding family histories and relationships.
No mention places, buildings, ships
The addition of another set of targets makes an already complex application with a steep learning curve even more complex with a steeper learning curve. The outcome of which will be to put even more new users off and overload the developers.
Perhaps there should be GRAMPS MkII (Replacing People with Places in terms of function) Building Research Software
phil
Yes, the Gramps data schema is already more than sufficient in its foundation. Only small adjustments are needed, and in its simplest form, it could just be a pointer with a relation type. However, that alone only solves the linking or references—it doesn’t fully address the core issue of event ownership.
So we are largely in agreement that this is a very simple addition, in much the same way as the upcoming changes regarding main-/sub-events seem to be. It is not about rewriting the schema, but about unlocking that final piece of native flexibility.
By the way, the interface of your app looks really great!
Note: This text was drafted in Norwegian to articulate the core concepts and examples, and translated/edited into English with the assistance of an AI collaborator for structure and clarity.
@DavidMStraub are we straying off-topic here?
The initial posting is about Domus. And specifically says it is not being integrated into Gramps Web (yet). Although it does mention house pins.
The latest postings seem to be about trying to redesign Gramps (whether the redesign is backend or frontend) to roll the Domus functionality/philosophy into the core of Gramps for desktops.
As OP, do you want that content digression split into another topic?