GENBOX-L ArchivesArchiver > GENBOX > 2003-05 > 1054045778
From: "Paul J. Harris" <>
Subject: RE: [GENBOX] Places - Almost there!
Date: Tue, 27 May 2003 10:29:38 -0400
From: Don Zochert [mailto:]
Sent: Tuesday, May 27, 2003 8:15 AM
Subject: Re: [GENBOX] Places - Almost there!
>Paul Harris wrote:
>> I am...particularly attracted to the handling of places through
>> linking, creating a single unique single record for each place
>> division. This is a feature I have long desired. Although the
>> mechanics of making the links can be frustrating at times, I'm
>> catching on and have created such sophisticated variations as
>> a town that has had three names, which is located in a township
>> that has belonged to two counties in one state!! ;-)
>Paul, what are the advantages of this feature over, say, creating a
>record for each variation of a town name with a note of how and when the
>name changed? Does the complexity add value?
Well, to my mind, that would be analogous to creating a separate person
record for each name variation that you find on a person, of course, noting
the fact that this is the same person as another person. Imagine how
unwieldy that would become! I really think if you start to look at places
the way we look at people in a genealogy database, this system begins to
Let me use an example of Chesterhill, Ohio, formerly known as Chester Hill
and Chesterfield. When linking an event to this town it would be appropriate
to use the name variation that existed at the time of the event, or that
which is in your source for the event. At any rate, over time, you will have
events in Chesterfield, Chester Hill, and Chesterhill. With a flat file
place database you would have three separate place records, and if you want
to add notes about the name changes, as you suggest, you would have to
REPEAT the notes in all three records. To find all of the events that took
place in this town you would have to search for all three name variations
(and remember there are three).
With a relational place file, such as exists in Genbox, I can create a
SINGLE record for the town and give it three different name variations with
a date range for their application. Any notes about the town go in this
SINGLE record. As I link events to this town, using the appropriate NAME
VARIATION, I am linking events to this SINGLE LOCATION, even though the name
varies. Now, when I look at events linked to this location, I get everything
that happened in the town regardless of whether it was recorded as
Chesterfield, Chester Hill, or Chesterhill.
To me, that is a great advantage. To others, it may not be. However, I
believe that such a system allows ME to manage my place records more easily
>To extend the question, what value do you see in a complicated,
>multi-part source template system over directly recording the format
>of each source as you want it to appear in a full footnote, second
>reference, or bibliographic note?
This question does not appear to be directly related to my post on places,
but I will express my opinion. The main advantage of a "complex, multi-part"
(hierarchal) source template is consistent output and flexibility to control
and change that output. Directly recording the format in the source template
will give consistency for THAT source, but will you be consistent in the
other 74 "similar" sources that you enter?? I know I wouldn't.
On the other issue of flexibility of CHANGING output, consider that people
decide to change citation forms for a variety of reasons, not to mention
that a new "standard" emerges, or we evolve our own. Say you have the above
74 similar sources, each with the format directly recorded in text of Item
A, Item B, Item C, and Item D. One day, you are compelled to rearrange the
output to Item D, Item A, Item C, and Item B. You now have 74 source records
to retype. If you had used a "complex, multi-part" (hierarchal) source
template, you could simply change the order in the template and all 74
sources would come out the way you want. Each of us has to decide for
ourselves which system is more complex.
There is an ongoing battle over whether one should use GENERAL or SPECIFIC
sources. Those who argue for general cite the "problem" of too big a source
list. Why, it becomes unmanageable! So they create a GENERAL source and put
all the data items in the Citation Detail field, which is free form text.
They wind up with one general source, but perhaps hundreds or thousands of
free form citation details!!! If they ever want to change the format of
their citations in any way--------! No thanks! I don't care how big my
source list gets. That's why I use a computer to manage it. I want as many
items of data in my source citation out of the CD and into template fields
in the source. This is equivalent to "tagging" each piece of data so I can
control its output. We're probably all going to be using XML based programs
someday, and the more data that is broken down into fields, the easier the
transfer will be. <g>
|RE: [GENBOX] Places - Almost there! by "Paul J. Harris" <>|