ENG-SHEFFIELD-L Archives

Archiver > ENG-SHEFFIELD > 2003-09 > 1062418068


From: "Linnell, Tim" <>
Subject: RE: [SHEFF] 1837 Online
Date: Mon, 1 Sep 2003 13:07:48 +0100


Regarding the 'Smith' issue (or other common names), there is a reasonable
way of finding the name you want. Given there are 26 letters, assume an even
distribution between the images and divide the number of images by 26. Then
go and have a look at the first first name at the top of the set.

This isn't likely to be the right name (there may be a good many John and
Mary Smiths, and very few Zebedees or Aarons), but will either be before or
after the one you want so gives a basis for further searching. You can then
use the actual first letter of the name you have found to either calculate a
new image to look at based on the letter, or take a stab using trial and
error (depends on how good your maths is). It will take a few goes, but once
done you will have a much better idea of where to look in succeeding years.
You also have to take into account things like the different number of index
entries at different times of year (more deaths in Winter), which requires a
bit more maths to do accurately, but basically what you do is:

if the index number on which you've found the correct name in one quarter is
'I', and the number of index entries in that year is N, and the number of
index entries in the new set is M then:

Good First Guess for Probable Index entry = M*(I/N)

Smiths are always going to be tricky, but genealogy has always been about
using brainpower rather than brute force and searchable indexes!

Cheers

Tim

This thread: