FLHIG

Family and Local History
Indexing Group

Selected articles from the Newsletter

This article first appeared in SIGGNL 13   (August 1996)

A decimal numbering of a one-name study card-index
by David Doidge

Introduction

[This article is reproduced, slightly edited and by kind permission of David Doidge's widow, from a letter to Tony Rydings, who has for some years been running a one- name study for Ryden, Ryding, Rydings and variants.  David died suddenly some five years ago, and his papers, including those on his Doidge-Dodge one-name study, have been passed on to Exeter University.  The item is not on indexing as such, but since I am embarking myself on a one-name study for Nanskivel and Nancekivel, with variants and likely corruptions such as Nankivel, Chivel and Kivel, I felt any helpful tips would be worth publishing. - Ed.]

Description

You asked me about my decimal numbering system: the origin of the system was simple.  Soon after I started collecting Doidge-Dodge data, I put the data onto 5" x 3" file cards in a certain format.  When I had a few hundred cards, I realised the need to uniquely identify each card - if only to clarify links such as husband and wife, parents and children, etc.  The simplest way was a number system of 1 to 1200.  At that time I could not envisage the scale being inadequate.  Abel was 1, and William about 1130.  Obviously, as the files increased, I found that I was running out of numbers for some names.  Also new names began to appear, which had to be accommodated in the number system at the correct alphabetic position.  The only way to continue was to introduce decimals of one, two or three orders.  As a system it is arbitrary but very effective, and theoretically can accommodate new names, repetitions, etc.  My file cards are each numbered in order, and are alphabetical.  At present, the "worst" number has six digits, eg 181.236.  There does not seem to be any foreseeable need to introduce seven digits.

My computer programme allows up to 8 digits, and when I perform a search, the ultimate act is to identify the card numbers I need, and thence pick them out of the file drawers.  The computer is a search tool - the file cards are the real records. Each card is virtually held in the computer memory with key words (ie names of people and places) indexed for fast searching.

Card no 1 is number 1.0 (not 1.00000): I do put a decimal zero after whole numbers, eg 1.0, 250.0, 987.0.  I do not otherwise pad out numbers; thus a typical series of three numbers might be 592.1, 592.11, 592.2 etc.  When introducing new data necessitating a new card, I frequently position the new card following a card to which it is related (if there is the same forename), and the card number follows logically.  My card file index is, of course, alphabetical by forename.  People with other surnames do not have individual cards, though their names are always indexed by the computer.

If you propose to introduce a number system, I suggest you use a base from 1 to 99,999.  You then have to approximately divide your base into alphabetical areas [work is being done by the FFHS on frequency of forenames - Ed].  Now, using my computer, out of 7000 "cards" there are 300 x Ann or Anne, 560 x Elizabeth, 136 x George, 130 x Henry, 215 x Jane, 230 x Joan, 760 x John, 680 x Mary, 320 x Richard, 107 x Roger, 140 x Sarah, 27 x Susan, 280 x Thomas, 480 x William. This gives you some idea of the frequency of names I have encountered.  Some families have particular favourite names.

Once you have decided where the major names are to be numbered, less used names can be allocated numbers.  Problems will arise at some future date when a new name arises that has to be inserted between the end of one name series and the beginning of another, the current next name.  You may then be forced into decimals - but so what!!  Preserving both alphabetic and numeric sequence is extremely advantageous.

Within a given name series, there is no particular order.  For general interest, I enclose a typical file card, plus below a screen print-out from the computer [which I have been unable to include; contact Tony or me for a photocopy, including sae and a few stamps - Ed].  Deciphering the data is not difficult.  Finally, my computer program is Cardbox, which allows you to format your card as you wish.  I obviously chose to format the data in the same way as my file cards.  You will see I do not put data sources on the computer; it is seldom required and is readily available from the file card.

Ç TOP