TMG-L Archives

Archiver > TMG > 2004-03 > 1079967350


From: Lee Hoffman <>
Subject: Re: [TMG] Projects and Data Sets
Date: Mon, 22 Mar 2004 09:55:50 -0500
References: <007601c40f47$32bd06c0$6400a8c0@E2004><200432191851.505056@TERRY2><6.0.0.22.1.20040321093839.023e0820@pop.ca.inter.net>
In-Reply-To: <6.0.0.22.1.20040321093839.023e0820@pop.ca.inter.net>


David Reed wrote:
> I am new to using TMG and I'm not sure I understand this concept
> of data sets and projects and their relationship. I had a look at the
> archives and from what I read, it appears that the data can be split into
> subsets, but it is still treated as if it were a single large data set.
> Is this correct?
>
> I split my data into subsets many years ago when the program I
> was using (Roots 3) limited the size of the data set to how much memory
> was available. The downside of this was that my grandparents were in two
> of the data sets and I and my parents were in all three. It would appear
> to me that I can now put them in a single project while retaining them as
> three separate data sets and, by fixing the links, could eliminate the
> duplicate individuals and still retain the correct links.

I used Roots3 prior to converting to TMG some ten years ago when it was
first released. At the time, my database consisted of some 9600
persons. Then I also started researching my wife's families as well as my
own and the number of persons quickly increased in size until today it is
over 39,000 persons in a single data set in a single project. Many times
over the years I have debated about splitting the data set and with the
advent of TMG v5x, I again considered splitting the data into multiple data
sets in a single project.

Many users have split their single data set into multiple data set and do
fine. In many cases, this requires that a few persons are duplicated in
two or more data sets with possibly some duplicates being only basic data
and the "original" data set (for that person) having full data. A lot
depends on the user and what they want as well as the persons in the
data. For example, in my situation, I have found many family connections
between my wife's family and my own. Their is, so far, only "by marriage"
connections and no real kinship, but if I had not had the persons in a
single data set I might not have discovered these connections. Mostly,
these connections are more of casual interest than anything else but I have
also found other connections as well that have helped me to break down some
of my brick walls via the "back door".

Thus I would suggest that you, at least, import your multiple Roots3
databases into a single project. Then you can (after making a backup of
that project) merge the multiple data sets into a single data set. Then
you can see how that works for you. If it does not work, you can always
revert to your backup of the multiple database import project.

There are a number of TMG users with well over 100,000 persons in a single
project/data set. So that should not be a problem for you. On the other
hand, you noted that you had split your Roots3 database due to computer
resources. That can still be a problem for you with TMG v5x depending on
the size of the project and your resources. If you have a reasonably
current system with plenty of memory, you should have no problem.

Hope this helps -

Lee Hoffman/KY
TMG Tips: <http://www.tmgtips.com>;
My website: <http://www.tmgtips.com/lhoffman>;
A user of the best genealogy program, The Master Genealogist (TMG)

This thread: