In a recent PR, an interesting point was raised about improving the labels in the family editor dialog. This post is to summarise the discussion in the PR and to allow the discussion to continue. Firstly, the image below highlights the labels being discussed
This is a good starting point for prior discussion in the PR. I post selected comments from the PR to this thread for convenience.
@Nick-Hall commented in the PR
Generally âspouseâ is going to be preferable to âhusbandâ or âwifeâ and âparentâ preferable to âfatherâ or âmotherâ. However, the family editor is used for families that may or may not have children and the partners may or may not be of the same sex. Do we need these headings in the family editor?
JRalls commented in the PR:
The current headings in the Family Editor are
Father/Partner1andMother/Partner2. Thatâs a bit clumsy. As you say there are families with no children and therefore no Father or Mother at all.How about letting the user choose what the family members are titled?
@emyoulation commented in the PR:
I like when the user can choose.
The Family object editor looks a bit âunfinishedâ with the current âFather/partner1â and âMother/partner2â labels.
The use of âSpouseâ would be a misnomer when the Familyâs Type is âUnknownâ or âUnmarriedâ
I commented in the PR:
whilst adding options sounds appealing, every option comes with a long term maintenance cost (keeping it working, documenting etc.). This option would probably have to be per family as well.
Can we devise a heuristic that does the right thing? Is the label signifying the relationship between the two adults or the relationship between the adult(s) and children?
The table below is incomplete but hopefully gives an idea. gramps would use the first 4 columns to determine which label to display. Also need to consider translation - does this scheme support all other languages?
To be clear. You are proposing that once a Family is formed, with or without children, the labels for the heads of the family will change based upon the configuration of the family and the Type selected.
Do you see any change in the labels before any person is added?
The only issue I see is that semantically (in a few places) the âPartner 1â position is used in the code to search for âfathersâ (male parent), and âPartner 2â for âmothersâ (female parent). It would take some work to remove that connection.
For example, instead of searching for âFathersâ (partner-1), that should be changed to search through partner-1 and partner-2 looking for males. But also connected software (like GEDCOM import/export) is more restrictive.
I was envisaging that the label would be dynamic. Every time the dialog is displayed, or the dialog content is edited, the labels would be refreshed.
So when adding a new family, initially both genders are unknown, no children and unknown relationship type. So, if the table above is used, the labels are âPartner 1â and âPartner 2â
Now if the left hand person is added, the persons gender is Male and the relationship changed to married, the left hand label would update to Husband. If a child is then added, the left hand label would update to Father.
Note that this is only a label in the dialog. Nothing is stored in the database.
The proposal might be overkill and\or might not work in all situations. Hopefully as the discussion proceeds, that will become clearer.
The only issue I see is that semantically (in a few places) the âPartner 1â position is used in the code to search for âfathersâ (male parent), and âPartner 2â for âmothersâ (female parent). It would take some work to remove that connection.
I was only thinking of adjusting the family editor dialog for now. Iâm sure there are many places that could also be updated, and it part depends on if you are searching for the biological father (only 1 person) or who the child calls father (could be 1 or 2 people). I would not be surprised if some languages have distinct words for these.
Do we even want to distinguish between a family with and without children?
I will admit that itâs not really an area I feel knowledgeable enough about to shape the policy. I just wanted to facilitate ongoing discussion outside the PR where it started.
Right, but it should probably be prevented to have âPartner 1â be Female, and âPartner 2â be Male, as that swaps the usage in the code. Oh, I see your chart doesnât even include that option.
My 2 cents⌠leave it as it is. We know how it works or what the fields represent, the change just makes things more complicated with little benefit.
An alternative would be to just remove these labels. Are they really necessary?
Thatâs certainly simpler! The select and add buttons then become a mini-toolbar
The Father and Mother disambiguate for Data Entry.
Adding a âFatherâ (rather than a Partner1) makes it logical to guess Male (and do a Patrilineal surname guess if there is a child providing that context). It suggests conforming to entering data with a father-onâleft, which leads to consistent layouts in charts/diagrams and supports ahnentafel numbering calculation.
99% of the genealogical data entry of Families prior to 2000 is going to be traditional Father/Mother, still the usual binary ordering regardless of whether nuclear or blended.
The Partner1/Partner2 is the exception, not the Rule. Why not default to the traditional and have a preference for polotical correctness support? So it might be âBonding pairsâ with choices of âNuclear familiesâ, âgender-neutral familiesâ, and âgender role familiesâ. And maybe stores overrides as a Family Attribute.
Gramps doesnât support recording true polygamous families either, just multiple bonding pairs. But that hasnât been an issue⌠yet⌠who knows what tomorrow holds.
I agree, and also want to point out that there are also labels in other places that would need to be considered if a change is made.
In the Relationships category, the default view uses the terms âParentsâ, âFatherâ, âMotherâ, âSiblingsâ, âSpouseâ, ad âChildrenâ. (If labels were to become dynamic, then it would be nice if they could show values like âStep-fatherâ etc. when appropriate.)
The âCombined Viewâ of the Relationships category uses the terms âParentsâ, âSiblingsâ, and âChildrenâ but not âFatherâ or âMotherâ (they are unlabeled). So thatâs already an example of doing without the Father/Mother labels.
There are probably many other places (reports etc.).
That would require all the children to have the same relationship. Less ambiguous to have the relationship remain as columns in the Children tab.
There is only one Active Person in the Relationships View. It would be their Step-Father or mother and their siblings.
My mistake. I was talking about object editors where that is not germane. There is no Active Person context while in the Family object editor.
That context only applies in the Relationship views and in Reports.
Sorry to put a downer on this but it strikes as semantic madness crossed with political correctness and within the English language likely to change within a the period of less than a generation anyway.
Firstly let me say that I am quite happy to acknowledge two males or two females can legally marry and by various means IVF, adoption etc have children.
However as far as I am aware the creation of a child requires a meeting of male created sperm and female created egg so there only requires to be labels Male and Female as selected when the individual is entered into GRAMPS and that is far as it needs to go, no mention of father and/or mother, husband or wife, spouse or partner.
Also there is a generation of heterosexual unmarried couples that abhor the term âPartnerâ because of the LGBTQ+ connotations.
This is why I do not use the Family options at all.
Also note I am well aware of individuals who are neither or both male/female and historically always have been and have problems with that.
phil
To finish, my vote is to simply drop all these labels by default.
n 02/01/2025 15:25, Brian McCullough via The Gramps Project (Discourse
Forum & Mailing List) wrote:
Looking at it more closely, I noticed that the labels in question are part of the glade file so it is not a matter of editing a py file to modify the label but creating code to send to the glade file to display the label.
For me, this is a lot of work for a limited return. I see edit windows as behind the scene functionality.
I would much rather see the effort put into making the views and reports that you would show Great-Uncle Henry as you present his family to him be as correct as possible.
