G Boyle's Genealogy

Descendants and relatives of Thomas and Henry Boyle,; and Sir Richard Boyle, First Earl of Cork

ABOUT AIDs, D'ABOVILLE NUMBERS, AND FOUNDING FATHERS

On this page you can read about how I use Alphabetic IDentifiers (or AIDs), which are unique alphabetic codes which I assign to each person in my BOYLE genealogical data base.

To the left is a graphic called a Wordle, created using  the site www.Wordle.net/create.  If you feed a text file in, you get an interesting graphic out in which the size of each word is scaled to its frequency in the file.  After you have read the description of AIDs below, come back and examine this wordle.  I have entered the AID prefixes for all founding fathers, once for each descendant.  I have also, for interest's sake, filtered out spouses and inlaw relations. You will note that spouses and inlaws are a significant percentage of the people in my database, but BOYLEs, RILEYs, FORESTs and descendants of many other "founding fathers" can be found here.

A searchable list of all descendants and relatives of Thomas and Henry BOYLE, the brothers who immigrated to Canada in 1823 as Peter Robinson settlers, can be downloaded in DOC format if you click here (493 Kb).  Let me be clear.  This is not a readable document.  It is just a sorted list of names which, I am hoping, will be scanned by Google and will bring people to this site.  The names of all 11,000 relatives found at this site are listed either in this file, or in the next file.

A searchable list of all descendants and relatives of Sir Richard Boyle, the First Earl of Cork, can be downloaded in DOC format if you click here (104 Kb).

About Alphabetic IDentifiers (AIDs) 

and “Founding Fathers”

The following text is available in DOC format if you click here.

When you are working with a family tree it is often the case, especially when working with the generations of people born before 1950, that many cousins will have the same names.  The practice was to name children after Royal personages, people in the Bible, or grandparents, parents, aunts and uncles.  So, in a large family tree, it was common to have several people with very similar or identical names.

This is a serious problem when you are writing family stories, or corresponding with people.  It becomes necessary to identify clearly who you mean by “Aunt Mary” when there are many Marys and almost every Mary is an aunt to someone.  For example, in the Davis family tree there are seven people with the name “John Davis”, six with the name “Joseph Davis”, five with the name “Sarah Davis”, and four with the name “Margaret Davis”.  If we are to write a story about any of these people, and if it is to be easily decipherable by those who are untutored in the intricate relations between them, we must find a unique identifier for each.

Modern genealogical software solves this problem by assigning a unique code  or number to each person.  The code is usually of this form:

 - The founding father (or mother) is assigned a root code;

 - The code of any child of the founding father contains the code of their father on the left, but has a serialized code appended to the right of it.  The serial code is determined by birth order, or by alphabetic order if birth dates are missing, or gender, or some other means.

 - The code of any grandchild of the founding father, and all of their descendants, has the code of the parent on the left, but has a serial code appended to the right.

For example, using Henry numbers (explained below) the codes might be assigned as follows:


PERSON                 CODE              EXPLANATION

Founding Father        1                     1st generation, 1 is default

Second child             12                   2nd generation, 2nd child

His/her third child       123                3rd generation, 3rd child

His/her second child   1232              4th generation, 2nd child


For the researcher who is compiling a family tree that involves British Nobility, establishing the correct order within a family is a little different.  Publications such as Burke’s Peerage list the males first, in birth order if known, and then the females, in birth order if known.  So, children may be listed in birth order, in alphabetic order, in gender order, or some combination of these.

Genealogical software will generate such unique identifiers for your family tree each and every time you print a report.   You must identify a “founding father” (or mother) and generate a printout of all descendants of that person.  The relation of each descendant to that founding father then determines the unique identifier of each person.  

The software that I use, called “Legacy”, has three types of numbers that can be so generated:

Henry Numbers (H#s) – Illustrated above, the founding father is assigned an H# of 1.  The next generation (his children) receive an H# of 1x, where the x is their rank in birth order (or order in this tree).  So, the first child gets a 11, the second child a 12, and the third child a 13 etc.

D’Aboville Numbers (A#s) – The founding father is assigned an A# of 1.  The next generation (his children) receive an A# of 1.x, where the x is their rank in birth order (or order in this tree).  So, the first child gets a 1.1, the second child a 1.2, and the third child a 1.3 etc.  This is seen as an improvement on H#s when a family has more than 9 children.  A#s such as 1.12.2.14 are possible, and H#s would use a kluge such as 1.c.2.e in which alphabetic letters are substituted for digits above 9.  (Like hexadecimal notation.)

de Villiers Numbers (V#s) – The founding father is assigned a V# of a1.  ‘a’ is an indicator of generation.  ‘1’ is an indication of rank within the generation (e.g. birth order).  The next generation (his children) receive a V# of a1 bx, where the x is their rank in birth order (or order in this tree).  So, the first child gets an a1 b1, the second child an a1 b2, and the third child an a1 b3 etc.  This is seen as an improvement on H#s and V#s when a family in the tree has more than 9 children, and there are a lot of generations, but less than 26.  A#s such as 1.12.2.14 would be translated to a1 b12 c2 d14.

However, apart from the minor problems in using one type of number over another, there are three larger problems with the computer generated numbers.  

Firstly, spouses and inlaws are never assigned identification numbers.  This means that almost half of the people in your family tree are NOT assigned a unique identification number in a listing.  This can be a problem.  E.g., if you have over a thousand families, there will be a lot of Marys marrying into the family, and, without unique identifiers it is difficult to keep them straight.

And then, secondly, if you add or delete a person, or insert a missing birth date, or otherwise change the order of listing of children in a family, then many or all of the unique identifiers change as soon as you print another listing.  Such “unique” computer-generated numbers are very ephemeral, and often only valid in the report in which they appear.  So, if you write a family story using the unique identifiers from a current printout, and, in response, you are informed of a person missing from your tree, the insertion of that person makes many of the identifiers in your story obsolete (i.e. mostly wrong and unreproducable).  E.g. if I learn about a child that did not live to adulthood, and insert that child into my data base, and then print a new family tree for someone, many or most of the computer-generated identification numbers will change.

And thirdly, if you choose a different “founding father”, then those people who are descended from both founding fathers will have two identifiers, and they are no longer unique.  For example, if I identify my 16 Gt-Gt-Gt grandparents, I will have eight sets of founding parents, and I will appear in eight different trees with eight different computer-generated identification numbers.  Not only do I have eight different IDs, for any one of these IDs it is possible that eight others share it.

I, personally, prefer the Henry numbers or the d’Aboville numbers over the de Villiers numbers when I must use computer-generated numbers, depending on the circumstances, but, for my own files, for the reasons described above, I use a version derived from the techniques my grandmother (Mary (Taylor) Boyle) used, which I call AIDs (Alphabetic IDentifiers).  

AIDs work as long as no person in the tree has more than 26 children.  Similar to the other types of numbers, the founding father is assigned an AID which is prefixed to the AID of all descendants.  [Note: When writing an AID in a story, I include the ‘AID=’ symbols, or I enclose the AID in curly brackets, so the reader knows it is one of those unique identification numbers.]  The first difference is, AIDs use only capital letters.  This sets them apart from the other three computer-generated numbers.  

So, if the AID of the founding father is A, it may be written as {A} or AID=A. [I will modify the way AIDs for founding fathers are assigned below, but for now, set it at {A}.] The children of the founding father will be given an AID of AID=AX, where X is a capital alphabetic.  So the first child in the founding father’s family will get AID=AA, the second AID=AB, and the third AID=AC, etc.  These AIDs are inserted into the Legacy software in a field called ‘User ID’.  I don’t know if other packages offer such a field.  In listings produced by Legacy, you can ask for the User ID to be appended to the name, and it is printed in curly brackets like this: Garvin Boyle {AB}.  It is stable, user determined, and works well.

Next, lets tackle the problem of multiple founding fathers.  Most people in the genealogy hobby are tracking the families of several founding ancestors (i.e. earliest men and/or women in a line), and the  descendants would have a different AID in relation to each such ancestor.  This is NOT a trivial problem, and the solution is a bit messy.  The solution I am about to describe works for me, because I have been working on my family tree(s) for over forty years, and they are mature, in the sense that I am rarely able to discover the parents of a designated “founding father”.  If I make a list of all of the ‘founding fathers’ or ‘founding mothers’ of all of my branches of the tree, that list changes rarely now.  So, having established a stable list of such earliest discoverable ancestors, I assign each a unique prefix, which is applied to all of the AIDs for their descendants.

In other words, I do not assign a default AID of {A}, to a founding father.  Instead I assign an AID as a prefix that is unique to each founding father.

For example, the earliest known BOYLE ancestor is the (un-named) father of Thomas and Henry Boyle, and I assign him an AID=Bo- or {Bo-}.  All such founding father prefixes are two or three letters, and end in a hyphen.  The AID of Thomas is then AID=Bo-A or {Bo-A}.  Henry, the (presumed) fourth child is AID=Bo-D.

Here is a list of the surnames of my earliest known male ancestors’ surnames on my father’s side and the prefixes that I have assigned to them:


   SURNAME          AID PREFIX

 - BOYLE               {Bo-}

 - RILEY                {Ri-}

 - HIGGISON         {Hi-}

 - DAVIS                {Da-}

 - TAYLOR            {Ta-}

 - FOREST            {Fo-}

 - STEWART         {Ste-}

 - STRINGER         {St-}

 - McINTYRE         {McI-}

 - McALLISTER     {McA-}

Note 1:   My mother’s side of the family has produced a similar list, not reproduced here.  I hope to organize that information for presentation in the next few months.

Note 2:  There are other surnames belonging to female earliest known ancestors such as BOYDE, GILCHRIST, ALEXANDER, CARTWRIGHT, COWAN AND JOHNSTON.  However, since I do not know names of parents or siblings, and just their surnames, they are the only people in the tree with these surnames, I have not yet assigned a prefix for them.  If I ever discover who their parents or siblings were, and begin an independent tree of descendants for that bloodline, then I will assign a new prefix for each such blood line.

Note 3:  Based on this list, I personally would have ten different AIDs, one for each of my earliest identifiable male ancestors.  On my mother’s side I have several more.  Hardly unique!

The surnames of Annie’s family, with the assigned AID prefixes, in priority order, are:


   SURNAME         AID PREFIX

 - LAMPINEN        {La-}    My inlaws, ancestors of my wife

 - EKSTROM         {Ek-}    My inlaws, ancestors of my wife

 - TELENIUS        {Te-}    My inlaws, ancestors of my wife

 - ANSAMAA         {An-}    Relatives of my wife


In addition, in my work as a self-designated family historian focused on my own ancestors and cousins, I have from time to time researched the ancestors of my relatives on their behalf.  I have assigned some AID prefixes for their earliest ancestors, though they are not my ancestors.  Their data is found in my Legacy data base and it is connected indirectly to me:


   SURNAME          AID PREFIX

 - SMITH                {Sm-}    Married into the RILEY family

 - PARCHER          {Pa-}    Married into BOYLE, DAVIS and FOREST families

 - PORATH           {Po-}    Married into the BOYLE family

 - BRISCOE          {Br-}    Married into the BOYLE and DAVIS families

 - ROGERS           {Ro-}    Married into the Taylor family


You may come across any of these prefixes when reading the listings in this website.

Next, we need a protocol to solve the multiplicity of my own personal AIDs and reduce my ten+ AIDs to one unique AID.  

The prefixes in the above list are shown in priority order.  If a person has several valid AIDs, they use the AID with the prefix of highest priority.

Priority is determined by marital connections.  A RILEY married a BOYLE in an older generation.  A DAVIS married a BOYLE in a later generation.  These marriages determine priority.

Essentially, if you look at an ‘ancestor fan’ (see below) you will see priority is assigned left-to-right.  E.g  all people who are in a direct line from my earliest BOYLE ancestor use the {Bo-} prefix and associated AID.  BOYLE descent is highest priority.  All people who are in a direct line from my earliest RILEY ancestor use the {Ri-} prefix and associated AID, unless they are also the offspring of someone with a higher priority prefix (i.e. a BOYLE prefix).  In those cases they use the BOYLE prefix and AID.  RILEY descent is, currently, second in priority.

There is a special case when someone marries into a lineage.  If the person marrying into the lineage is not descended from any of the identified ‘founding fathers’, they do not have an AID of their own prior to the marriage, and they simply take on the AID of their new spouse, with a suffix (S).  However, if they are directly descended from one of the ‘founding fathers’ and have an AID of their own, I have assigned them two AIDs.  Their native AID is listed first, and the AID of their spouse is listed second, with the (S) suffix added.  The offspring derive their AID from the spouse with the higher priority AID. 

Note that the assignment of prefixes and priorities is highly dependant upon:

(a)   The person whose array of ancestors has been identified.  It would be the same for me and my brothers and sisters, but for nobody else.  The array of possible earliest identifiable ancestors is determined by my unique ancestry.  And, 

(b)   The current state of knowledge of my ancestors.  It would change (new prefixes added) as my knowledge of my ancestors advanced.

Here is the current ancestor fan of my paternal grandmother.

By examining the fans, you will see how priority order has been established.

There is one last problem to resolve, before we have truly unique AIDs for each person in the family tree.  In the interest of simplicity, the solution I describe here is only a partial solution.  There is the possibility that many people of interest who are not in direct descent from my ancestors will nevertheless be in my family tree, i.e. spouses and their relatives, otherwise known as inlaws.  I addressed this above, but I would like to go into a little more detail here.

First and foremost among those are, of course, the spouses of the direct descendants of ‘founding fathers’, as discussed above.  The spouses of my direct ancestors are, of course, also my direct ancestors, and each has the possibility of opening up a new prefix, a new tree of relatives, a new project of genealogical research into my distant cousins.  This is, in fact, how the list of priority prefixes was established.  But, every other person in the data base who is not ancestral to me may also have a spouse, so there is a need to assign them a unique AID as well.  This gets complicated, and much of this you will not see in the listings I have posted on the internet, but, for completeness sake, I provide my entire AID protocol here.

List of all AID suffixes:
Template        Description                                       Examples
XXY                Child of AID=XX                                 {BD} is the fourth child of {B}, standard suffix for a child
XXX(S)           Spouse of AID=XXX                           {BD(S)} means spouse of {BD} when there is just one spouse, standard suffix for a spouse
XXX(Sn)          nth spouse of AID=XXX                      {BD(S1)} means first spouse of {BD} when there is more than one, 
                                                                                {BD(S2)} means second spouse of {BD},
                                                                                etc.
XXX(SS)       Spouse of a spouse                            {BD(SS)} means an non-related spouse of a 
                                                                                twice-married person {BD(S)} who also married {BD}.  
                                                                                There are some like {BD(SSS)} meaning
                                                                                spouse of a spouse of a spouse.
             OR
                     Sibling of a spouse                            {BD(SS)} means brother-in-law or sister-in-law of {BD}, 
                                                                                but not otherwise related to me.
                                                                                Distinguishable only by context.
XXX(SP)      Parent of a spouse                              {BD(SP)} means the father-in-law or mother-in-law of {BD} 
                                                                                who are otherwise not related to me, 
                                                                                i.e. the parents of {BD(S)}
XXX(SPP)     Grandparent of a spouse                   {BD(SPP)} means grandparents of {BD(S)}
                                                                                {BD(SPPP)} is great-grandparents

Note that AIDs for inlaws are not always unique, and are ambiguous in meaning without context.  Since my online listings only include descendants of ‘founding fathers’, and spouses of such descendants, most of these ambiguous variations will not be in the online listings.  They exist only in my Legacy database, or in some GEDCOM extracts that I may decide to make. 

I am aware that this AID protocol is not for the faint-of-heart, and will probably not be used by others.  It works for me.  The AIDs assigned in my BOYLE BULLetin in the early 1980s are now largely obsolete, but those I assigned in the early 1990s using this protocol are, mostly, still valid.  This has given me the ability to manage a database of over 14,000 names, and always know where a particular person fits in.

My approach is this.  I manually assign stable AIDs, which are not computer generated and are unchanging.  Since I try to keep the AIDs stable over decades of research, sometimes the people are not listed in alphabetic order according to their AID.  

I know.  It’s complicated.  But it’s the best I can do.  It's what I do.

Nakemiin!

Last updated: 22 October 2012