The short of it:
You have a dude named Charlie. You know his grandpa was Albert, and you know Albert had three sons Bud, Bob, and Bill. But you don't know which of those guys was Charlie's dad. How would you yourself, practically speaking, represent this in your PGV database?
The long of it:
The first Hawk man in Georgia was Andrew. He had four sons Jacob, Peter, Andrew Jr., and William. These four men sired generation3, or the 3g's, as I'll call them. But the 3g's were born and raised during the 1820 - 1840 census years, when only the head of household was listed by name. Due to the Civil War, burned courthouses, rampant illiteracy in the family, and so on, records are very scarce during that time. From the 1850 census onward, I have good records of the 3g's on down to the present day. The problem is connecting the 3rd generation to the 2nd. My first strategy was to leave them unconnected, but then I lost the tree. I then connected them by guess and by gosh, with explanatory notes, but of course in gedcom this looks so formal and irrefutable. The visual affirmation of parent/child structure is stronger than a thousand words of notes. My current strategy was to create a new fictitious individual "AJP" Hawk, which stands for "Andrew, Jacob, or Peter," and assigned all 3g's of unknown parentage to him. He is connected to Andrew, so the generational structure is now correct, but it looks so weird. Given that gedcom structure doesn't support a probabilistic assignment of parents and so forth, I'm very interested to know of any workarounds or such that you've done in your own lines. I'm hoping someone has thought of a better solution than mine.
-Victor
If you would like to refer to this comment somewhere else in this project, copy and paste the following link:
Anonymous
-
2009-01-22
Victor - your solution sounds fine - but so does you original one (guess the father, and cover with notes).
I can only offer one other idea - if you don't have enough FACTS to complete an entry - don't enter it.
Personally, I like you last idea the best - with plenty of notes all over the place.
If you would like to refer to this comment somewhere else in this project, copy and paste the following link:
> Given that gedcom structure doesn't support a probabilistic assignment of parents and so forth
I have a similar situation. I have a Simon in my tree born 1861 or 1867 (can't tell, but a separate record has it as 1861) in Switzerland with no further information. I have a freestanding branch of Largiaders in the US who originated from a Simon who was born in the early 1860s (based on immigration and offspring dates) in the right area of Switzerland (based on a census). At the moment I don't know for certain that it's the same guy but the 'main trunk' is pretty well filled out and there aren't any other loose Simons matching the description. And I don't know how much more info there is to have, so it would be nice to have a shaded line, dotted line or something representing a less than 100% solid link.
But OTOH this field is not about representing guesswork, so I can see that side of it also.
If you would like to refer to this comment somewhere else in this project, copy and paste the following link:
<<this field is not about representing guesswork>>
I prefer to call it "making the best interpretation of the available facts", rather than "guesswork", but the distinction between the two is little more than the experience of the guesser.
To recap, you have some grandchildren, but don't know which of several siblings is their parent.
In these cases, I tend to create an extra "dummy" sibling, and attach the children to them. Call this person "Dummy /Surname/", to make sure they stand out, and add suitable notes.
You can, of course, attach them to your "best guess" sibling, but you run the risk of someone not seeing your notes, and taking your assertions as fact.
If you would like to refer to this comment somewhere else in this project, copy and paste the following link:
The short of it:
You have a dude named Charlie. You know his grandpa was Albert, and you know Albert had three sons Bud, Bob, and Bill. But you don't know which of those guys was Charlie's dad. How would you yourself, practically speaking, represent this in your PGV database?
The long of it:
The first Hawk man in Georgia was Andrew. He had four sons Jacob, Peter, Andrew Jr., and William. These four men sired generation3, or the 3g's, as I'll call them. But the 3g's were born and raised during the 1820 - 1840 census years, when only the head of household was listed by name. Due to the Civil War, burned courthouses, rampant illiteracy in the family, and so on, records are very scarce during that time. From the 1850 census onward, I have good records of the 3g's on down to the present day. The problem is connecting the 3rd generation to the 2nd. My first strategy was to leave them unconnected, but then I lost the tree. I then connected them by guess and by gosh, with explanatory notes, but of course in gedcom this looks so formal and irrefutable. The visual affirmation of parent/child structure is stronger than a thousand words of notes. My current strategy was to create a new fictitious individual "AJP" Hawk, which stands for "Andrew, Jacob, or Peter," and assigned all 3g's of unknown parentage to him. He is connected to Andrew, so the generational structure is now correct, but it looks so weird. Given that gedcom structure doesn't support a probabilistic assignment of parents and so forth, I'm very interested to know of any workarounds or such that you've done in your own lines. I'm hoping someone has thought of a better solution than mine.
-Victor
Victor - your solution sounds fine - but so does you original one (guess the father, and cover with notes).
I can only offer one other idea - if you don't have enough FACTS to complete an entry - don't enter it.
Personally, I like you last idea the best - with plenty of notes all over the place.
> Given that gedcom structure doesn't support a probabilistic assignment of parents and so forth
I have a similar situation. I have a Simon in my tree born 1861 or 1867 (can't tell, but a separate record has it as 1861) in Switzerland with no further information. I have a freestanding branch of Largiaders in the US who originated from a Simon who was born in the early 1860s (based on immigration and offspring dates) in the right area of Switzerland (based on a census). At the moment I don't know for certain that it's the same guy but the 'main trunk' is pretty well filled out and there aren't any other loose Simons matching the description. And I don't know how much more info there is to have, so it would be nice to have a shaded line, dotted line or something representing a less than 100% solid link.
But OTOH this field is not about representing guesswork, so I can see that side of it also.
<<this field is not about representing guesswork>>
I prefer to call it "making the best interpretation of the available facts", rather than "guesswork", but the distinction between the two is little more than the experience of the guesser.
To recap, you have some grandchildren, but don't know which of several siblings is their parent.
In these cases, I tend to create an extra "dummy" sibling, and attach the children to them. Call this person "Dummy /Surname/", to make sure they stand out, and add suitable notes.
You can, of course, attach them to your "best guess" sibling, but you run the risk of someone not seeing your notes, and taking your assertions as fact.
<<it would be nice to have a shaded line, dotted line or something representing a less than 100% solid link>>
There is always this gedcom structure
1 FAMC @F123@
2 STAT challenged
2 NOTE blah blah blah
However, I'm not sure whether many applications support it. PGV doesn't, and will ignore it.
I tend towards the dummy mode when I'm missing a generation I'll create one called
"Missingname /surname/"
Then add a note stating the problem and fill in the information later either by a merge or re-link to the correct person.
I would put in all three names as
0 @I###@ INDI
1 NAME Andrew? /Hawk/
1 NAME Jacob? /Hawk/
1 NAME Peter? /Hawk/
and use ASSO (if there's a way) to connect him to all three of the "real" guys.
That way, the "fake" appears in searches and lists under all three names.