APG-L Archives

Archiver > APG > 2004-01 > 1073049487


From: "Jerry Fitzpatrick" <>
Subject: RE: [APG] Re: Name Registry?
Date: Fri, 2 Jan 2004 07:18:07 -0600
In-Reply-To: <89.8a3bed.2d26287b@aol.com>


Donn,

I agree with you in principle, but how you handle the data depends upon how
you define 'persona'. The new data set refers to *some* person, but you're
just not sure if it's one you've already identified.

You can think of a 'persona' as the data set itself, but you can also think
of it as a "presumptive new individual". If you later determine that the new
individual is really the same old individual, you can merge the two
together. You're not really presuming that the data set refers to a new
individual; you have a conservative "working hypothesis" that it refers to a
new individual until you know better.

In any cataloging scheme (computerized or other), you have to have a way to
locate information. The most logical way to do this is to index by person
since individuals are singular whereas names are not. Under these
circumstances, associating a data set only with a source effectively puts
the data in limbo. Alternatively, you could index by source. However, since
one source might identify dozens or hundreds of people, this would be a
rather ineffective way to find data about people (personas).

I may be wrong, but I believe that the GENTECH model (and all the genealogy
programs I know of) effectively index by person (persona). The data set
(assertion) about a persona is still associated with a source, but you keep
track of the information by using the persona, not the source. This seems
simple and effective to me.

Our disagreement is largely semantic, so maybe using different terms would
help. Consider there to be "real people" and "hypothetical people". Give
each one (real or hypothetical) an identifier called "file number" so that
we can look them up in our files. Don't our differences fall by the wayside?

---
Jerry Fitzpatrick
Software Renovation Corporation

http://www.SoftwareRenovation.com


-----Original Message-----
From: [mailto:]
Sent: Thursday, January 01, 2004 7:51 PM
To:
Cc:
Subject: Re: [APG] Re: Name Registry?

Jerry, I would argue that until you're reasonably sure about the person to
whom an assertion should be attached, it's more important to keep the name
and
the data associated with its source. For that purpose, it seems cleaner to
give
the unique identifier to the known data set and source (persona), rather
than
to a presumptive new individual, in a case where attaching the assertion to
an already-known person is possible but still uncertain.

I have a file folder full of assertions about 18th-century married women
named Martha Drake (most without a clue to a birth surname) from southern
New York
and northern New Jersey. At least three different women are necessary to
account for all the persona involved, but whether there are more than three
is not
clear enough to justify assigning facts to just those three. While it could
be done without conflict, it would be an unwarranted presumption.
Meanwhile,
the data sets need to be kept separate, and identified with their sources
rather than an indeterminate number of subject individuals.

Donn Devine

Jerry wrote:

> When the evidence is meager and/or ambiguous, the right thing to do is to
> assume multiple individuals (and therefore separate personas). I'd argue
> that this is the minority case, along the ambiguity no doubt increases as
> you delve further into the past.
---
[This E-mail scanned for viruses by Declude Virus Scanner]



This thread: