Confidence Source vs. Linking

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

Confidence Source vs. Linking

Lasse
Hi all,

I have a problem entering my datas sources, using Gramps 4.2.8.


I have a Child "1", of which I have a marriage certificate stating his
parents ("2" and "3"). I now found a birth certificate for a Person with
the same name as B stating his Parents ("4" and "5").

Both sources are certificates / church records, which I value as "very
high" confidence (I have no possibility to find anything better).

I now enter a Link from "1" to "2"&"3" as well as a link from "2" to
"4"&"5". All data in my tree now has "very high" confidence.

But *I* do not have "very high" confidence in that tree, I am absolutely
not sure, that both the marriage and the birth certificate reference the
same human being.

In short: I am sure about "1"-"2a" and "2b"-"3", but not "2a" == "2b".


I have some custom graphs that use the source confidence to highlight
secured information, sometimes I like to filter out stuff that I cannot
proof 100%.


How would you handle such a situation? I'm missing a "is this source
pointing to that person"-confidence level in addition to the source
confidence.


Kind regards,
Lasse

------------------------------------------------------------------------------
Check out the vibrant tech community on one of the world's most
engaging tech sites, Slashdot.org! http://sdm.link/slashdot
_______________________________________________
Gramps-users mailing list
[hidden email]
https://lists.sourceforge.net/lists/listinfo/gramps-users
https://gramps-project.org
Reply | Threaded
Open this post in threaded view
|

Re: Confidence Source vs. Linking

Dave Scheipers
Hi Lasse,

The Confidence rating is not a rating of the facts of the source but
do these facts apply to your relative.

The baptism is a fact. The confidence rating is how sure are you that
the baptism is for your relative or is it for someone else.

HTH Dave
On Fri, Aug 31, 2018 at 2:50 PM Lasse Schnepel
<[hidden email]> wrote:

>
> Hi all,
>
> I have a problem entering my datas sources, using Gramps 4.2.8.
>
>
> I have a Child "1", of which I have a marriage certificate stating his
> parents ("2" and "3"). I now found a birth certificate for a Person with
> the same name as B stating his Parents ("4" and "5").
>
> Both sources are certificates / church records, which I value as "very
> high" confidence (I have no possibility to find anything better).
>
> I now enter a Link from "1" to "2"&"3" as well as a link from "2" to
> "4"&"5". All data in my tree now has "very high" confidence.
>
> But *I* do not have "very high" confidence in that tree, I am absolutely
> not sure, that both the marriage and the birth certificate reference the
> same human being.
>
> In short: I am sure about "1"-"2a" and "2b"-"3", but not "2a" == "2b".
>
>
> I have some custom graphs that use the source confidence to highlight
> secured information, sometimes I like to filter out stuff that I cannot
> proof 100%.
>
>
> How would you handle such a situation? I'm missing a "is this source
> pointing to that person"-confidence level in addition to the source
> confidence.
>
>
> Kind regards,
> Lasse
>
> ------------------------------------------------------------------------------
> Check out the vibrant tech community on one of the world's most
> engaging tech sites, Slashdot.org! http://sdm.link/slashdot
> _______________________________________________
> Gramps-users mailing list
> [hidden email]
> https://lists.sourceforge.net/lists/listinfo/gramps-users
> https://gramps-project.org

------------------------------------------------------------------------------
Check out the vibrant tech community on one of the world's most
engaging tech sites, Slashdot.org! http://sdm.link/slashdot
_______________________________________________
Gramps-users mailing list
[hidden email]
https://lists.sourceforge.net/lists/listinfo/gramps-users
https://gramps-project.org
Reply | Threaded
Open this post in threaded view
|

Re: Confidence Source vs. Linking

Ron Johnson
In reply to this post by Lasse
On 08/31/2018 01:36 PM, Lasse Schnepel wrote:
Hi all,

I have a problem entering my datas sources, using Gramps 4.2.8.


I have a Child "1", of which I have a marriage certificate stating his parents ("2" and "3"). I now found a birth certificate for a Person with the same name as B stating his Parents ("4" and "5").
Both sources are certificates / church records, which I value as "very high" confidence (I have no possibility to find anything better).

I now enter a Link from "1" to "2"&"3" as well as a link from "2" to "4"&"5". All data in my tree now has "very high" confidence.

But *I* do not have "very high" confidence in that tree, I am absolutely not sure, that both the marriage and the birth certificate reference the same human being.

In short: I am sure about "1"-"2a" and "2b"-"3", but not "2a" == "2b".

But I'm not sure what you mean... :)


I have some custom graphs that use the source confidence to highlight secured information, sometimes I like to filter out stuff that I cannot proof 100%.


How would you handle such a situation? I'm missing a "is this source pointing to that person"-confidence level in addition to
the source confidence.

To me, Confidence indicates how sure you are that the document says what you think it says, like "this is my relative".

Do you know for sure whether either of the two children really belong in your tree?


--
Angular momentum makes the world go 'round.

------------------------------------------------------------------------------
Check out the vibrant tech community on one of the world's most
engaging tech sites, Slashdot.org! http://sdm.link/slashdot
_______________________________________________
Gramps-users mailing list
[hidden email]
https://lists.sourceforge.net/lists/listinfo/gramps-users
https://gramps-project.org
Reply | Threaded
Open this post in threaded view
|

Re: Confidence Source vs. Linking

Philip Weiss
One of the limitations of Gramps is that the confidence level is per-citation, not per application of the citation. To explain:

I have death certificates for for my great great grandmother Maria (Hein) Solle and several of her siblings. Some of them give their father's name as Victor, some as Vincent. I am 100% confident with parts of the each death certificate regarding the date of death. My confidence level regarding the fathers name is nil.

Even if I split apart the citation into multiple citations, one for each part of the document, I still end up with changing confidence levels as I learn more about the fact I'm researching.

Consequently, I don't use the confidence level part of a citation, which appear to be a relic of how genealogists thought about citations back when Gedcom was first devised.

What I do use is an overall assessment of what I know about each fact/event/relationship/etc.. I have a separate source called "Deductions and Proof Arguments". When I have contentious facts, I'll attach a citation for that source to the fact/event, and provide my reasoning and assessment of the overall fact in a text note attached to that citation. And since there's only one of these citations per fact, I can enter the confidence level. (The idea of this source came from someone on the list, I can't remember who. Thank you.)

To your specific example, if I'm unsure if citations and events refer to the same person, I create two and don't merge them until I have high confidence they are the same person. So for a child-parent relationship A->B1 and B2->C, if it looks like B1 and B2 are probably the same person, but I don't feel confident in the citations as a while yet, I've still got them separate. All the citations I know for sure refer to B1 get stuck there. All the ones I know for sure are B2 get put there. And when I've got a few that for sure make them the same person, THEN I merge.

The only drawback I've found to this so far is some reports don't make as much sense. I have a Kate Parker and a Genevieve Parker I think are the same child of James Parker, but I'm not confident in that yet. So for now my family group sheet and family lines charts and whatnot have too many children, and the years of birth are kind of wack to boot.

I hope this helps give you ideas, Lasse. I don't think it's the one true way, but it's worked for me.

Phil.

On Aug 31 2018, at 3:45 pm, Ron Johnson <[hidden email]> wrote:


On 08/31/2018 01:36 PM, Lasse Schnepel wrote:
Hi all,

I have a problem entering my datas sources, using Gramps 4.2.8.


I have a Child "1", of which I have a marriage certificate stating his parents ("2" and "3"). I now found a birth certificate for a Person with the same name as B stating his Parents ("4" and "5").
Both sources are certificates / church records, which I value as "very high" confidence (I have no possibility to find anything better).

I now enter a Link from "1" to "2"&"3" as well as a link from "2" to "4"&"5". All data in my tree now has "very high" confidence.

But *I* do not have "very high" confidence in that tree, I am absolutely not sure, that both the marriage and the birth certificate reference the same human being.

In short: I am sure about "1"-"2a" and "2b"-"3", but not "2a" == "2b".

But I'm not sure what you mean... :)


I have some custom graphs that use the source confidence to highlight secured information, sometimes I like to filter out stuff that I cannot proof 100%.


How would you handle such a situation? I'm missing a "is this source pointing to that person"-confidence level in addition to
the source confidence.

To me, Confidence indicates how sure you are that the document says what you think it says, like "this is my relative".

Do you know for sure whether either of the two children really belong in your tree?



------------------------------------------------------------------------------
Check out the vibrant tech community on one of the world's most
engaging tech sites, Slashdot.org! http://sdm.link/slashdot
_______________________________________________
Gramps-users mailing list
[hidden email]
https://lists.sourceforge.net/lists/listinfo/gramps-users
https://gramps-project.org