From: Frederico M. <fs...@gm...> - 2010-02-25 19:46:22
|
Hi all, I honestly think that this would be something good to have for a future release, so I'm rehashing the debate. Some context: the issue is to provide a more felxible way to deal with names that do not follow a simple inheritance rule of only one family name. The more well known examples are Spanish and Portuguese names (and former colonies of both, henceforth refered as Iberian for the purpose of this debate, although there are differences between them). The simplest way to define this "compound surnames" thing is to look at a simple example like my name: Frederico Serrano Muñoz. That's given name + mother's surname + father's surname. In Spain I would be Frederico Muñoz y Serrano (or something like that, the connector is optional), but tha's merely a matter of ordering (and that's also the way it was used in Portugal two centuries ago). Some more complex names exist in the form of two given names + two mother's surnames + two father's surnames. This usage doesn't work well with the "Given name" and "Family name" division since Family name is more correctly used for a single surname. Plus, it assumes that the descendants will have the same Family name, which isn't true in the above case. OTOH putting surnames in the Given name is semantically incorrect (and while I've done it, I'm undoing it right now, it looks weird even if it works for purposed of surname inference and ordering). Some previous debates and resources: An open bug: http://www.gramps-project.org/bugs/view.php?id=3161 A wiki-based discussion and attempt at proposal: http://gramps-project.org/wiki/index.php?title=Names_Discussion One of the ML threads: http://old.nabble.com/Proposals-for-additional-name-fields-tt25356341.html I will be the first one to admit being a bit lost. In the ML debate there was a proposal based on connectors (by Benny) that seemed ideal. I added some initial content to the wiki page, and before adding more I would like to get the ball rolling once again: > Am I correct in my understanding? Is this sufficiently flexible. > The only other way would be to not store a surname, but store a > surname list, which would be like a table of surnames, of which one is > the primary surname, eg > > José de Mascarenhas da Silva e Lencastre > > would in table form be stored: > order prefix surname suffix primary? connector to next? > 1 de Mascarenhas / Y N > 2 da Silva / N Y > 3 / Lencastre / N N > > That would be a very different way of input, and would store > everything and be flexible. Obviously only one name can have primary > indicator 'Y' The above is IMO one of the closest to ideal proposals I've seen. I can even see something based on that being used to be able to trace a surname that was inherited "irregularly" (i.e. sometimes by the mother's side, sometimes by the fathers side). This is very common since I've found that in Iberia there was a tradition of daughters inheriting the surname of the mother, and sons inheriting the surname of the father. It makes each surname a "first class" entity in terms of managing and maintains their relationship, which means no information is lost and exports can be done with confidence. Regards, Frederico |