TMG-L Archives

Archiver > TMG > 2004-05 > 1083458075


From: "Kevin Sholder" <>
Subject: RE: [TMG] Tag vs Individual Sentences
Date: Sat, 1 May 2004 20:34:35 -0400
In-Reply-To: <6.0.0.22.2.20040429112314.038bc9f8@pop3.norton.antivirus>


Thank you for the insight AND advice.

Kevin

-----Original Message-----
From: Lee Hoffman [mailto:]
Sent: Thursday, April 29, 2004 12:36 PM
To:
Subject: [TMG] Tag vs Individual Sentences

I have been re-reading a few messages that were posted on here in the past
(some back in 2002) and mulling over in my mind some of their points.
Primarily, I am refering to the discussion about whether it is best to make
changes to the Tag and Witness Sentence via the Master Tag Type List (MTTL)
or make the change to the individual Tag and leave the default Sentences
alone.

There are a number of pros and cons either way. Against making the change
in the MTTL is that you are only intending that change to apply to the
single instance in which it is used. The problem with this is that later
you find that you have made individual changes to a number of individual
Sentence and mostly the changes are the same. So, on reflection, you decide
to make the change to the MTTL and keep the Sentence consistent. This means
that you will have to go through the data set and find those persons using
this Tag Type and whose Sentence has been changed. On the other hand, if
you had spent some time early on and studied things, you _might_ have
foreseen this situation and made the change early. Then if you wanted to
alter the change default Sentence, it would be easy.

On the other hand, there are a couple of reasons to just make changes to the
individual Sentences. One is that each is likely to be a little different
and thus your narrative reports would appear to be less "canned." This is
of more importance to me although some may prefer the consistency of
standardizing the Sentences.

But a reason against making the change in the MTTL is that if you then merge
that data set (A) with (and appending to) another data set (B), the MTTL
changes in data set B will disappear. The Sentence in the MTTL for data set
A will remain and thus override those of data set B. Of course, you could
insure that the data set B Sentence remain by "reversing" the merge such
that data set A is appended to data set B. But if you had made changes to
the MTTL Sentences in data set A, they would be lost in the
merge. However, during either merge above, no individual Tag Sentence
changes would be lost and all would be retained in the merged data set.

The problem with the merge situation is that while you might be able to
retain the changed Sentences in the MTTL depending on the direction in which
you do the merge, that option may or may not be suitable for other reasons.
For example, your main data set might be A and have persons assigned ID#s
which you do not want changed while data set B has only a few people and you
are willing for their ID#s to be changed. Then there is the situation in
which you and another researcher have exchanged data (both using TMG to
simplify things). If the other researcher merges your data set (B) with
his, he may lose data depending on what you changed in the MTTL Sentences.
Conversely, if you merge his data set (B) into yours, you may lose data
depending on what he changed in the MTTL Sentences.

Of course, if the "real" data is in the various field of the individual
Tags, and only the Sentence Structures are changed as to what Variable
prints where, then no data is lost although some of it may not print in
reports since the Sentence may not refer to that Variable.

Now a Custom Role added to a standard Tag in one data set will be copied to
the same Tag in another data set and any Sentence for that Role will be
included in the merged data set -- IF the "original" data set does not
already have that Role in it with a different Sentence.

Thus there are pros and cons to how Tag and Witness Sentence should be
changed. The main thing is that depending on what you do regarding standard
Tags (and to some extent Custom Tags) and their Sentence and Roles, you need
to take care when merging data sets so that you don't lose work that you
have done. This is especially so if you have spent hours developing Roles
and Sentences in various Tags.


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)


==== TMG Mailing List ====
To un-subscribe from TMG-L (in MAIL mode), send a message to
<> [to <> in Digest
mode] with just the word "unsubscribe" (no quotes)in the text and turn off
your signature.

This thread: