APG-L Archives

Archiver > APG > 2004-01 > 1073846491


From: "Jerry Fitzpatrick" <>
Subject: RE: [APG] Re: APG-D Digest V04 #8
Date: Sun, 11 Jan 2004 12:41:31 -0600
In-Reply-To: <5.2.1.1.0.20040111123444.00aa9a20@cpcug.org>


Mary Jane,

The GENTECH data model is basically a logical/conceptual model. It defines
this thing called a 'persona' which lets you collect information about
people without deciding which individuals are which. The GENTECH spec says
that each persona is assigned an identifier. However, the spec does not
require the identifier to be globally unique.

Neither the GENTECH model nor my proposal solve the problem of deciding how
many men are referred to by 100 documents naming John Smith. That's still
something that can only be done by a person.

However, my proposal, by assigning a globally unique identifier to a
specific individual, provides a way to correlate data between physically
separate GEDCOM databases. That is, once the John Smith in GEDCOM database A
is determined (manually) to be the same John Smith named in GEDCOM database
B, they can be given the same globally unique identifier.

Such a decision should not be made in a cavalier fashion. Like any tool, it
can be used wisely or unwisely. Right now, though, there isn't even a tool
to use unless you happen to have a genealogy program that provides it (Tom
Morris had mentioned one).

---
Jerry Fitzpatrick
Software Renovation Corporation

http://www.SoftwareRenovation.com


-----Original Message-----
From: Mary Jane Smith [mailto:]
Sent: Sunday, January 11, 2004 11:41 AM
To:
Subject: [APG] Re: APG-D Digest V04 #8

Thanks, Donn. I see that once again, I am still " underinformed." So much
to learn! I will check out the site you referenced. I suppose my next
question is: if the terrms are already defined, or self-evident, and the
data modeling is already done, how can the model be applied to the
problem under discussion, i.e., Jerry's original idea? Or, does it apply?

Mary Jane

At 09:00 PM 1/10/2004 -0700, you wrote:
>APG-D Digest Volume 04 : Issue 8
>
>Today's Topics:
> #1 [APG] Re: Name Registry? []
> #2 [APG] GenTech [Jan Davenport
> <]
> #3 [APG] Baptismal Date vs. Birth Dat ["Nancy Coleman"
> <.]
> #4 Re: [APG] Baptismal Date vs. Birth [Ray Beere Johnson II
> <]
>
>Administrivia:
>To unsubscribe from APG-D, send a message to
>
>
>
>that contains in the body of the message the command
>
> unsubscribe
>
>and no other text. No subject line is necessary, but if your software
>requires one, just use unsubscribe in the subject, too.
>
>______________________________X-Message: #1
>Date: Fri, 9 Jan 2004 22:57:10 EST
>From:
>To:
>Message-ID: <>
>Subject: [APG] Re: Name Registry?
>Content-Type: text/plain; charset="US-ASCII"
>Content-Transfer-Encoding: 7bit
>MIME-Version: 1.0
>
>Mary Jane:
>
>The GENTECH Lexicon Working Group did the data modeling and defined the
term
>"persona." See http://www.gentech.org/gdm/
>
>I think Jerry was proposing "hypothetical person" as an easily understood
>working equivalent, for purposes of the registry at least, that wouldn't
need
>explanation each time it was used.
>
>Donn Devine
>
>Mary Jane smith wrote:
>
> > Jerry and Donn -- Any way you two can get together and do some data
> > analysis? It sounds to this old data analyst that you need first to
>define
> > your terms; i.e., who or what is a "persona?", a "hypothetical?" and so
> > forth, and then do some data modeling. (No, I am not volunteering,
> but it
> > does sound like a good idea.) As one trying to sort various John
Smiths
>in
> > Virginia, I will be interested to hear your results.
>




This thread: