Re : Is there a particular reason why "Divorce" isn't included as a relationship type?

classic Classic list List threaded Threaded
23 messages Options
Reply | Threaded
Open this post in threaded view

Re: Re : Is there a particular reason why "Divorce" isn't included as a relationship type?

John W. Kitz-3
Enno, Dave,

Hello Dave,

> Maybe you can help by telling us where the Relationship Type affects
> others things we do with Gramps. Where does it come into play?
I have no idea how or where it is used in reports, but I personally use
it for two things:

1. When I find the word "wife" in a birth record, or "widow" or
"widower" in a death record, I conclude that the persons referred too
were married. Saving that conclusion in the relationship is much quicker
than creating a fake marriage event.

2. When I find a birth record where no father is mentioned, or one for
which I know that the father is not the biological one, I create a
relationship between the biological parents with type "not married".
This can only be done when I have reliable hearsay about the real
father, of course.

In situation 1, I know that I may have to look for a marriage record,
and when I can register that, the relationship type is sort of overruled
by the marriage event, which can be followed by a divorce record too,
when available. I leave the relationship type to "married" then still,
because in my and Gramps' logic, that's what the relationship was. A
divorce is not a relationship type.

In situation 2, I use the "not married" relationship type as an
indicator that there is not much need to look for a marriage record,
although there may be one. But if there is none, having the "not
married" relationship type is much easier than a fake "not married"
marriage event, which will lead to very awkward texts in reports.

> Would adding a custom type cause something irregular to happen?
In my mind, yes. Creating a divorced relationship type is inconsistent
with the software design, and creating a custom type also leads to
problems when a language border is crossed. That's why I will always try
to avoid custom types myself.

> Does it show up in reports are cause the relationship to not show up
> based upon what the type says, or if were customized?
I sincerely hope that setting the relationship type to "not married"
will suppress the M word in any report that I may create some time. I
want Gramps to be accurate, and not report marriages because some fool
thought that all people were or should be married.

> Bottom line... Just how important is the Relationship Type.
See above. I need it to record conclusions about the nature or a
relationship, especially when people were not married, and there is no
other way to prove a non event. I also use it for the temporary
conclusion or reminder that a marriage record should be found some time.

Please consider that the relationship type 'unmarried' seems to apply to
just about any relationship type other then 'married', which would
include divorced, separated, widowed and, while pointless from a
perspective of recording a family in Gramps, 'single'.

While there are many ways to accomplish the same, with respect to using
it as a temporary way of recording that one still needs to find some
certificate, you might consider the following:

1. Enter a a couple of generic notes with texts, such as 'Find birth
certificate.', 'Find marriage certificate.' and 'Find death
certificate.', etc. with a type of e.g. 'Research';
2. Attach the appropriate note to the entry of the individuals to whom
it applies;
3. In addition one may, but doesn't need to, create the appropriate
event for an individual with what data one may have found to date, while
taking advantage of the various options available in Gramps. Such as
entering a date of 'unknown' (text), a date range (after, before, range)
or an approximate date (about).

The advantage of doing it this way IMHO is that one only needs to click
one of the generic notes, then references, to find the individuals for
one still has to find the certificate mentioned in the note.

Once that has been done, one clicks on the name of the individual (in
the list of names to which the note refers) for which the certificate
has been found and updates the information with the additional
information found.

Fake events look bad in reports, so Gramps should prevent them where



Regards, Jk.

Check out the vibrant tech community on one of the world's most
engaging tech sites,!
Gramps-users mailing list
[hidden email]
Reply | Threaded
Open this post in threaded view

Re: Re : Is there a particular reason why "Divorce" isn't included as a relationship type?

In reply to this post by enno
For me, if one is "divorced" or "widowhood", then there is a "time" variable.

During investigations, I found logical to have 1 state for Family whatever
change during life time for these adults.
For me, the design sounds good!


Call the relationship calculator

>>> from gramps.gen.relationship import RelationshipCalculator
>>> rc = RelationshipCalculator()

Get the relation between two people

>>> one = self.db.get_person_from_handle(fhandle)
>>> two = self.db.get_person_from_handle(mhandle)
>>> rel_type = rc._get_spouse_type(self.db, one, two)

The default relation type for families

>>> from gramps.gen.config import config
>>> rel_type = config.get('')

Get the relation string between these people

>>> rel_one = rc.get_partner_relationship_string(integer,
where 'integer' will be either:

0: "",

If there is a divorce event or annulement, then the "ex-"
prefix will be returned on the relation string.

As you can see, we do not set any date, only people!

There is a minor complexity under non-english locale, but
we also have localized relationship handlers for any cultural

As we cannot have a divorce without a marriage, first.
"Divorce" seems to be a secondary state.
Same for "Widowhood" without a death event, first.
To handle them as relationship type will make harder any
return for relationship types and partner strings (above samples).

This is rather an issue for "GEPS_001:_Relationship_type_event_link".

In france too, since 1792, all marriages are "civil", then "religious" as an option.
So, in theory, people who married in France have at least one marriage event, but could
be more. Does "religious" divorce exist for all religions? What will be the relationship type
if there is one "civil" divorce and "religious" marriage? We could find some common
historical "royal" samples in England and France... Does "annulment" event mean "divorced"
or just "no more married"?


 Check out the vibrant tech community on one of the world's
 engaging tech sites,!
 Gramps-users mailing list
 [hidden email]

Check out the vibrant tech community on one of the world's most
engaging tech sites,!
Gramps-users mailing list
[hidden email]
Reply | Threaded
Open this post in threaded view

Re: Is there a particular reason why "Divorce" isn't included as a relationship type?

John W. Kitz-3
In reply to this post by enno

Some final remarks from me in this thread.

On 2017-02-27 21:14, Enno Borgsteede wrote:

> Hello John,
>> On 2017-02-25 16:06, Enno Borgsteede wrote:
>>> Hello John,
>>> The simplest answer is that status and type are different things, and
>>> a
>>> relationship type refers to the sort of relationship that people once
>>> had. Divorce as an event is not a relationship, but rather the end of
>>> it, and divorced is a status, not a type.
>> It looks like we're more or less stating the same (albeit in different
>> words) in that events are cause for some individual civil status[1] or
>> the change thereof, e.g.:
>> o Birth (event) -> Born (status);
>> o Marriage (event) -> Married (status);
>> o Separation (event) -> Separated (status);
>> o Divorce (event) -> Divorced [from so and so] (status);
>> o Death (event) -> Died or Deceased (status) and related Widow or
>> Widower [from so and so] (status).
> No we don't. I wrote that status and type are different things. They
> are different in my mind, and in all the software that I work with, or
> maintain.
>>> When you take the definitions like this, and software is always based
>>> on
>>> formal definitions, a relation type 'married' as opposed to other
>>> types,
>>> does not end with a divorce. A divorce creates a new status called
>>> 'divorced' but that does not not mean that the relationship type
>>> itself
>>> is changed.
>>> I set the relationship between my parents and most other couples in
>>> my
>>> tree to married, because that's what they were. Divorces,
>>> separations,
>>> or deaths, don't change the relationship types that once were.
>> No, but events such as divorce and death do, at least IMHO.
> In yours, yes, but not in mine. For me, a type is like the brand or
> model of a car.

With all due respect, it looks like you need to revisit your
understanding of the meaning of words like status, type, brand and model
as they relate to the topic of this discussion.

> If I were to register my car history, I can write that
> I once had a Volvo 440. This car was recycled quite a while ago, but
> in my car history, it's still a Volvo.

The analogy between certain terminology, such brand and model, that can
be used to describe ones car and comparable terminology, such as status
and type, to describe the past, current or (possible) future status of a
legal contract of sorts that a marriage is, is IMHO questionable at

E.g. you state that for you relationship type is comparable to the brand
or model of a car, but since the brand[1] of a car is something
different than the model[2] of a car the relationship type can only be
compared to one, not to both, or to be specific in the case of your
Volvo 440 the brand is "Volvo" and the model is "440", while one would
typically expect the use of words such as operational, in mint
condition, demolished, inoperable, classic, wrecked, towed, impounded,
or totaled to describe any of the past, present or future statuses of a
model 440 car of the brand Volvo.


> When I think of relationships, I see the type of relationship between
> my parents (marriage) as an eternal fact, which means that they were
> married, just like the first car that I owned was a Volvo.

Referring to the above, the status of a relationship is, as is the
status of a car, by definition subject to change, be it, in the case of
a car, as the result of normal use or an event such as a traffic
accident, or, in the case of a relationship, through marriage,
registered partnership or an event such as death or disappearance (of
one of the spouses), separation, divorce filing, divorce and what not.

> The
> marriage ended when my father died, in 2004, but the type of
> relationship was a marriage for most of the time, and until the end,
> i.e. my father's death.

The present day (not the past) relationship between two spouses ends
when one of them dies and it is my understanding that the civil status
of the remaining spouse changes to widow or widower and as a consequence
the status of the relationship IMHO ought to change from married to

> And on a similar scale, I see my own
> relationship to my father as an eternal one too.

To state the legal fact that someone is the son or daughter of so and
so, IMHO conveys no information whatsoever about the relationship
between the parent and the child in the context of Civil Law, whereas
words like adopted, stepchild, child, biological child, do.

(significant portion removed)

> regards,
> Enno

Regards, Jk.

Check out the vibrant tech community on one of the world's most
engaging tech sites,!
Gramps-users mailing list
[hidden email]