TMG-L Archives

Archiver > TMG > 2001-04 > 0986237420


From: Linda Lawhon <>
Subject: Re: [TMG] TMG for Dummies + Merge
Date: Mon, 02 Apr 2001 13:50:20 -0500
References: <4.3.2.7.2.20010401232733.00aaeaa0@pop3.airmail.net>
In-Reply-To: <200104020513.f325DB723876@mail.redshift.com>


said:
>Why not, before merging the two files, first import the one to be
>merged as a separate TMG file. In that file create a special flag,
>then when they are merged the imported records will be identified by
>the special flag.

Thanks for all the input from all. I think Bob's idea is the best at
present. Too bad he didn't tell me that before I thought of merging. :-)

This was/is a very good lesson for the newcomers to TMG. The
concensus of opinion ( I think) is it is better to have one big
database. That is not to say you cannot have or create some subsets
for specific purposes. I am not a newcomer to TMG and I have used
the one main database. I have also said I would never merge another
dataset with mine. Pass the ketchup, please. I did start a little
sub dataset and I never expected I would want it in my main
set. Then I was sent a dataset from another researcher which I added
to my little subset. It was rather easy to clean up the duplicates
in this set. Well, the day came -- I decided I certainly did need
all these people in my main dataset. Altho I am not new to TMG,
because of my previous pontification about merging someone else's
material with mine, I have not messed with the find duplicates nor
the merge. The only time I have messed with it is when I found two
people in the picklist that I knew were duplicates -- so I merged
them right there on the spot.

If I had had the experience of merging datasets, I might have done
things a bit differently. Like -- go back to the archives and read
all those posts written by those who do it all the time. Sigh. Or
-- I might have considered asking the list before I jumped in and did
what I did. Unfortunately -- I didn't know what I was in for until
the deed had been done. So far I have made two dumb moves -- more to
come, I am sure. First I opted to delete duplicate tags ***and***
names in the find duplicate merge screen. What I did -- because I
went at alphabetically -- was unlink all my grandchildren from their
mother. I can only thank the stars that I noticed this
immediately! Second -- I merged about 25 duplicates without doing
any editing in the find duplicate merge screen. This second move was
the reason for my original post. I remember some of them -- but not
all, so I am now going through the pick list one by one. If I recall
the presentation by Bob at GENTECH -- this kind of thing will be
*much* easier in v5.

Linda

This thread: