Branches: A welcome addition
Brought to you by:
canajun2eh,
yalnifj
Menu
▾
▴
Thanks for this new feature. I'd like, however, to comment on a few things.
First, the branches listing is built on just a textual comparison of surnames. In this case, I don't think that surname prefixes should be taken into consideration; presence, absence or differences in prefixes should not break the branches. Besides, the internal links to branches produced do not work when several surnames are present: links contain all surnames without the commas. The resulting search does not work.
Second, spelling changes break the branch. It may be interesting to have alternative purely patrilineal and matrilineal branches where surnames are only used to detect key people, so that branches are not broken even in case of complete surname change. Of course, computation would be more expensive, but the results would be very interesting.
Sweet, now - can we print these? Individual branches?
@Julio :
<<First, the branches listing is built on just a textual comparison of surnames.>>
We use textual *and* soundex comparison to get the list. But soundex is based on english and does not give perfect results.
<< presence, absence or differences in prefixes should not break the branches. >>
Should already work : enter DONGEN, and you get all people with surname DONGEN or VAN DONGEN, etc..
<<Besides, the internal links to branches produced do not work when several surnames are present: links contain all surnames without the commas. The resulting search does not work. >>
May you explain which links or give an example ?
<<Second, spelling changes break the branch. It may be interesting to have alternative purely patrilineal and matrilineal branches where surnames are only used to detect key people, so that branches are not broken even in case of complete surname change.>>
There are so many cases (typo variants, migrants, single women and unknown father, etc) that it is difficult to find a perfect solution. Please suggest any rule we may use for better results.
Check:
http://www.enredo.es/gendb/branches.php?surn=la.rreta&ged=sl.ged
i think that, except branch #9, all records are actually the same branch, but get disconnected for one reason or another.
Also, try the link in #4 for Gumersinda (marquesa de Robledo de Chavela) de Mena y Maza (click on her surname), you will ger to:
http://www.enredo.es/gendb/branches.php?surn=de+Mena+y+Maza&ged=sl.ged
That is empty, not even Gumersinda is found.
My site is full of problem branches like those. Feel free to experiment.
As for rules that produced different results, besides improving on what is there, there are alternatives. What this code does is to try to follow surnames. If the surname changes (I have plenty of examples), the branch breaks. Whether this is desirable or not is a matter of preference, even the same person may want to alternate between methods.
For instance, on my site, there are people who had no surnames or could change their surname on each generation or different siblings could get different surnames. Think for a minute about the current practice in Iceland, where most people follow the patronymic system: surnames cannot be followed to build branches. Royal houses exhibit a similar problem because most members did not have actual surnames (many known with surnames are post-fact surnames that they themselves did not use in their time).
That's why I suggested and alternative method to complement the current method. Picture how to answer the question: What is the list of people having the same mythocondrial DNA than a given person? Or the same Y-chromosome DNA? These are pure matrilines or patrilines that are continued even if surnames change. I am not particularly interested on DNA, I just offer the example to explain what I am talking about.
Sorry, the intended link was actually:
http://www.enredo.es/gendb/branches.php?surn=larreta&ged=sl.ged
<< If the surname changes (I have plenty of examples), the branch breaks. >>
Branches is nothing but a list by surname.
For bloodylines up or down, use pedigree/descendancy/ancestry.
I will fix the link problem, and try to find a better way to detect minor surname changes.
@Stephen:
<<can we print these? Individual branches?>>
I know you have a very big tree.
What would you like exactly ?
> Branches is nothing but a list by surname.
> For bloodylines up or down, use pedigree/descendancy/ancestry.
Again, thanks for this, but pedigrees or descendancy lists contain superfluous people, people who do not belong to a patriline. They are not a substitute for what I want.
Imagine I want to create a book with all I know about my relatives. I could start listing descendants of some distant ancestor of mine. Say those of surname "Sánchez-Perdido". Actually that's my own patriline. Unfortunately my grandfather was included in the Civil Register as just Sánchez (and Gómez as a second surname). Never mind, I print this. Then I go to some other ancestor, in this case of surname Gómez-Zurdo. This happens to be the same Gómez surname of my grandfather that was mutilated as well. But even if the surname change did not create a problem, then this list would contain all descendants of my grandfather again. As I add more and more listings, I get more and more duplicates. The way this has been solved by hand in the past is listing each branch just once. A criterion is needed to decide what branch should each person go to. Manual methods leave a lot of freedom and books tend to contain inconsistencies w.r.t. the chosen method. Computers usually need a stricter definition of what a branch is. You have chosen one, but it is not the only possible method. I'd like to see at least patrilines as well. That does not mean I am asking you to do it. It is just something I'd love to see.
Besides patrilines (matrilines are rarely seen), there is another related listing that appears frequently on books, is that of succession (on a property, a land, a country, etc.) that explains how a property passed from hand to hand, inheritance dead-ends, jumps to other family branches, etc. You see, there is no end to the kind of things you can do if you have the data.
Julio,
What IS patriline? Is it a tree of all MALE descendants of a MALE patriarch? Or is it a tree of all male and female descendants, but for each female her descendants are not listed? Are spouses of male descendants included or are they spurious? Are male descendants with no children included or are they spurious? All such trees are strictly genealogical trees, and could be created with a little effort from the data. I feel that depending on the unchanged name is a little silly today. It was not in middle ages, though, so perhaps such a tree may be of historical interest. What are the criteria for inclusion?
_Marek
In the most common definition, a patriline contain descendants of either sex, but only males propagate further down. Their spouses are not part of the patriline but may be mentioned together with other facts. Same for matrilines, they contain descendants of either sex, only females propagate. Other words used for patrilines and matrilines are resp. agnatic or uterine lines.
I don't know about other countries, but in Spain surname stability is backwards: regular after 1870, very variable before that, and mostly chaotic before the Council of Trent (1563). During half of the middle ages (before ca. 1200), people only had patronymics, so surnames would change every generation, just as in Iceland today.
So following surnames is useful in modern times, but breaks easily in earlier times.
Julio,
Most European countries have names reaching at most 14-15 century, see an excellent Wikipedia article http://en.wikipedia.org/wiki/Last_name. Which n.b. adds a problem in PhpGedView: how to enter the kings: In Henry III is III his "last name", or is the last name unknown and III is just the suffix?
With your definition, it should be trivial (and I don't mean trivial amount of work, only "trivial" mathematicians use it - all the information is there) to write the code that would prune all children of female descendants. (And the male for the matrilinear line - after all only mothers carry the some unchanging digital code, while fathers carry only a 1/2-1/2 mixture of subroutines). I feel this would use the data in the Gedcom much better than relying on the spelling of the name. In my tree, wherever I reach far enough into the past, name eventually changes.
-Marek
Marek
Regarding "Henry III is III his "last name""
British 'tradition' is that such Kings (and Queens (before Eliz II) were named after their "House" or lineage. So the normal surname for Henry III (and Henry II, Richard I and John) is PLANTAGENET.
"III" is definitely a suffix. Still confusing genealogically, but there is at least an accepted practice.