TMG-L Archives

Archiver > TMG > 2005-07 > 1122134115


From: Lee Hoffman <>
Subject: RE: [TMG] List of Folders for Validate File Integrity
Date: Sat, 23 Jul 2005 11:56:13 -0400
References: <00cf01c58f93$d7832db0$6501a8c0@charliexvi><000201c58f94$42c118d0$6902a8c0@Moms17>
In-Reply-To: <000201c58f94$42c118d0$6902a8c0@Moms17>


Teresa Elliott wrote:
>Thanks Jim,
> If we could point to a directory and have it search all folders, wouldn't
>this option become obsolete, since you could have it search C:/ or D:/ etc?

Well, the checking in this case (Exhibits) does not take long -- especially
if the folders are specified and only the specific folder is checked. I
would suggest that doing a tree search just starting from the root
directory of each disk may be impractical. I have two hard drives on my
main computer and both are about half full. When you consider that is
about 60 gigabytes of data with tens of thousands of files, then you can
see that the time could grow almost exponentially. On the other hand,
doing a tree search from say C:/Pics down to say two or even three levels
then the search would have a small limit and would go more quickly.

>This is a fairly interesting topic, since most of us have no idea what VFI
>really does.

Well, first you need to know that database management is based on entering
something only once. For example, places in TMG are only entered
once. When the exact same place is referenced in a different record then
that original entry is "pointed to".

Now TMG is rather unlike many other database programs in that there not
just a single database containing your data. Rather, TMG uses more than
two dozen different databases to hold the information for a single
project. This design allows greater flexibility and especially provides
the user with increased capability to enter much more data than could be
efficiently handled in a single database.

Because larger project can cause operational speed to decrease slightly,
the designers have provided more indexes to decrease the time to access
certain data. But the speed of data access can be greatly increased in
other ways as well. One of these is that some data can be stored in two
(or even more) different databases. One will always be the master data for
that specific information but the rest can be used to more quickly access
some data.

Thus a lot of VFI is mainly just making sure that data is correct for the
place it is stored (for example, you can have sex of M, F, or ? but a sex
of G would not be valid). So a lot of the checking is like that but then
there is also a lot of checking that redundant data agrees with the master
data for that data. Also there are other ways of ensuring that data
entered and stored is correct. All this checking is done using standard
data storage techniques.

For example, one method of data checking is with what is called a "check
sum". Most people see this all the time with credit cards and account
numbers. Often this seen with a single digit appended to an account
number. For example, when the number is entered into the credit card
machine, the computer adds up the number and compares the value of the
"check number" with the sum of the rest of the account number. If they
match then the account number is valid.

In any case, this is a very general kind of description of what VFI
does. Any more would be too technical and more boring that what I have given.

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: