Thanks for the tip Benny! I've been putting the patronymic second. The thought occurred to me that I should probably put the patronymic name first, but now I have to fix up hundreds of people in my database.

I haven't opened any defects yet, since I'm still a relative newby on Gramps and I'm still figuring out best practices. I think one problem that should be fixed, though, is with the option to treat a patronymic as surname if no conventional surname exists. Right now, the patronymic is listed twice.

Regarding the sorting of names, I don't know if the current design is wrong or just working as designed. As I mentioned before, I'd like to see enhancements to allow sorting on particular pieces of names, such as unprefixed surname, regardless of how the name is formatted,

Cheers! Hans

On Tue, Apr 30, 2013 at 9:42 AM, Benny Malengier <> wrote:

2013/4/23 Hans Boldt <>
That's a good summary of the situation. For me, name format "Given Patronymic Primary" displays the names acceptably. But then the People view sorts on given name. Any name format with the primary surname first always ends up with other issues.

Pekka, you need to click the + after Origin field to set several surnames, and origin type should be patronymic to indicate it is a patronymic.

As Enno said, the order of the surnames needs to be the one you want them to appear in, so patronymic first sometimes.
If there are problems in display/sort, they are just that, bugs that need fixing. Don't change how data is stored, raise a bug ticket for the display/sort problem. I remember I indicated there was a bug indeed in sorting, but already forgot the details. Was a bug ticket then raised with the details?

When we added patronymic, we decided to group it with surname, not given names. This because normally brothers need to have the same patronymic. We could have made a new intermediate field, but decided against that for compatibility reasons, and to reduce the overhead of working with patronymics.
Note that Swedish users urged for the addition of farm name as origin type. We did not do this, and opted for the more general 'Location' as origin type. Origin is freely editable, so a user can add origin type 'farm name' himself without problems should he prefer that. Personally I would add a note with more details instead of making a lot of origin types.



As we discussed before, in addition to other things we've talked about, I still really want to see an option on the People view to sort names based first on primary surname (without prefixes) and secondly on patronymic. That way, we can be much more flexible in how names are displayed. 

Cheers! Hans

On Mon, Apr 22, 2013 at 6:34 PM, Enno Borgsteede <> wrote:
Hi Hans,

Yes, you can identify which part of a name is a patronymic. However, there is a lot of potential for improvement. For example, in the formatting and sorting of names. That is, I want to see names presented as: (given names) (patronymic) (family name). And if a family name is not given, the patronymic should be treated as a family name. I've fiddled with the Gramps settings, and I can't seem to configure it quite the way I want.
I remember the previous discussion, and I think that I have an idea where things go wrong. It's still sort of an idea, no proof, but it looks like it can explain what happens.

Here are some examples of Dutch names:

'Dirk van de Bunt' - family name is "van de Bunt", no patronymic
'Dirk Mellisen' - no family name, patronymic is "Mellisen", which should be treated as a family name
Dirk Mellisen van de Bunt' - family name and patronymic

One solution is to just not bother using the patronymic attribute in Gramps. However, the problem in research is that in the old records (even sometimes in post-1811 civil records), a person might be referred using patronymic, family name, or both. So it's very useful to be able to identify the patronymics.
In my database, most of the names were imported from programs that don't support patronymics, and in all cases I find the patronymic name behind the given name, unless there is no family name at all, in which case the patronymic name is treated as a surname indeed.

After import, I find surnames like "van de Bunt" in my database, which can then be sorted as Bunt when I let Gramps move "van de" to the prefix field. So far, everything looks OK.

Trouble starts, when I try to convert Melissen to a patronymic in Gramps. When I do that, Melissen suddenly moves to the SURNAME part of the screen, and that's not what I want, because it also means that the name will be sorted as Melissen, even when I designate "van de Bunt" as the primary name. I want the patronymic to be (optionally) shown between the given and the family name, but in most cases it should not be treated as a part of that, unless there is no real family name for a person. And IMO adding an extra name to sort on just to get things right is one step too far.

When I look at the surname types in Gramps, they seem to be treated as exactly that, SURNAME types. And when I think about surname types, they are sort of a scientific explanation of the origin of that name. In todays world, Johnson is a surname with a patronymic origin, but it doesn't mean that a persons father is named John. Likewise, Borgsteede originally refers to a farm, or a village, and can thus be designated as of type location, but it doesn't mean that my personal origin is in that place.

When I see a patronymic name, I like to treat it like an intermediate level, between the given and the family name, just like it is still used in Arab countries today.



Precog is a next-generation analytics platform capable of advanced
analytics on semi-structured data. The platform includes APIs for building
apps and a phenomenal toolset for data science. Developers can use
our toolset for easy data analysis & visualization. Get a free account!
Gramps-users mailing list