Geneweb Secrets
Handling Names
Jean Vaucher (sept 2010)


Geneweb has many options to deal with variants in people's names like married names, adopted names or alternate spellings, but be aware that it impossible for any software package to handle every possible situation. Furthermore, it is impossible to enter HTML tags in the name fields; so that it is well nigh impossible to display names using your personal style conventions, for example with underline for the usual first name, etc... as in "John Andrew (Lefty) McGovern".

Rather, you should concentrate what you can achieve by using Geneweb's options. There are 3 main objectives: Reference, Display and Search.

By this I mean that the official name, as it appears on the birth certificate, should be recorded somewhere.
This means choosing the name that will be displayed whenever there is a reference (and link) to the individual in his role as a parent, child or sibling, or node in trees. This should be representative but not too long.
How to make the individual easy to find. On the one hand, this means anticipating "First_name/Family_name" pairs that people are likely to enter in the search box on the home page:

For example: obvious search entries like "Bill Clinton", "Babe Ruth","Marilyn Monroe" or "Nicolas Sarkozy" are quite different from those people's actual names.

One should also think about the entries that will be placed in Geneweb's "First Name" and "Last Name" indexes... remembering that most name fields - like "aliases" - are excluded from these indices.

However, you can always use the NOTES to complement the information in the standard fields.

Simple Names

Before we get into the options provided by Geneweb to handle special cases, it is useful to consider the obvious cases. We think of a person's name as being made up of two parts: some first names followed by a family name (or surname). For example "James Earl Jones" is made up of these two parts:
In Geneweb, these two pieces of information would be entered in the "First name" and "Surname" fields of the Modify Person Form (as well as selecting "M" for Sex).
In GEDCOM format, this information corresponds to this line:
		1 NAME James Earl /Jones/
A more complete GEDCOM description would look like this:
		0 @I1@ INDI
		1 NAME James Earl /Jones/
		1 SEX M
		1 BIRT
		2 DATE 17 JAN 1931
		2 PLAC Arkabutla, MS
		1 OCCU Actor
Here is the top of the corresponding Geneweb page (the photo was added via the "send image" link on the Update Form) :

The name shown in large letters at the top of the individual's page has particular importance for both display and search. It is the principal name that will appear whenever the individual is referenced on other pages, either as a child, or a parent or a spouse. It is also the name that should be used in HTML links between user-provided pages and Geneweb pages. Finally, its parts are the only elements entered as keys for searching via the Name lists.

-- Back to the TOP --


To find people, one can enter a name in the search box on the Home page.

Using the first option, you would enter the full principal name "James Earl Jones". Note that spacing, case and accents are disregarded, so that "Jamès EARL    jOnes" would have worked just as well. When you use the other two options, you are directed to lists of individuals who share the given first name or surname, as the case may be. The lists give additional information (dates or relatives) to help find the desired individual. Note that the index of first names is based on the concatenation of all the first names in the principal name. Therefore, neither "James" nor "Earl" alone would work.

On the same topic, notice that the principal name name at the top of an individual's page is actually composed of two LINKS corresponding to the 2 name parts, in our example: "James Earl" and "Jones". Clicking on the first part leads to an index of "James Earls" and clicking on the other to an index of all "Jones's".

Later, we shall see that when a Public Name is specified, it replaces the principal name as the heading of the individual's page with the principal name - in smaller letters - placed below. However, the principal name retains its role as primary reference and this is noticeable because its parts serve as HTML links whereas the Public Name has no such link.

The complete index of surnames (or first names, or place names) can also be browsed from the Home page.

-- Back to the TOP --

Creating HTML links

Knowing the two parts of the principal name, you can create URL requests (or HTML links) for an individual: the first and last names being coded as "p" and "n" parameters [the first names separated by "+"]. To access the page for Albert Einstein on the well-known ROGLO database, you would use:;p=albert;n=einstein
Access to our "James Earl Jones", on a local database would be coded as follows;
A better way to code links to people in the same database is covered in the next chapter.

The examples below access pages on the ROGLO database:

	Albert Einstein;p=albert;n=einstein
	Pierre Elliot Trudeau;p=pierre+elliot;n=trudeau
	Bill Clinton (you need to know his principal name);p=william+jefferson;n=blythe;oc=2

Occurence numbers

If there are several individuals with the exact same (first+last) name, Geneweb uses occurence numbers (oc) to distinguish them, the first individual having oc=0. By default, if "oc" is absent, the system assumes oc=0. Geneweb's reliance on the first+last+oc combination is further explained in the official documentation.

In our example, the request for "James Earl Jones" below is equivalent to the one shown previously:


In the ROGLO example for President Clinton, the "oc=2" indicates that there are other William Jefferson Blythes in the database. Note that, for efficiency, Geneweb normally relies on internal ID numbers -which replace the first+last+oc combination. For example, "999240" is the ID I found for Albert Einstein in ROGLO; but this can change if the database is reorganized and the second link using the principal name is preferable:;i=999240;p=albert;n=einstein

-- Back to the TOP --

The "no_index" option

Often, you can guess the first and last parts of the principal name from the page header but the oc part may not be not obvious. However there is a way to find all parts of the principal name given a numeric ID. To do this, you add ";opt=no_index" to the URL link and display the page again: this gives the 3 parts (first(p)+last(n)+oc) of principal name. For example: presently in ROGLO, Queen Elizabeth II is person #1340919 as indicated in the URL field when we reach her page. Adding ";opt=no_index" shows that, in her case, "oc=2".

-- Back to the TOP --

Standard photos

The first_name/last_name/oc triplet is also used, albeit in a different format to access the standard photos. Geneweb keeps these in a folder with the name of the database (here: a1) in the "bases/images" folder and our James Earl Jones photo is stored as:
The divider here is the underline character and the occurence number is "0". If there was a second James Earl Jones, his photo would be stored as:

-- Back to the TOP --

Handling name variants

Geneweb has many fields which deal with variations in people's names (nicknames, regal titles, double-barrelled names, showbiz aliases, etc...). These are entered in the Modify a Person Form (Click to see) and these are documented HERE.

In databases of prominent people (Kings, Queens and Presidents), TITLES are important. For example, ROGLO relies heavily on them - as well as additional user created indexes - to allow users to locate prominent people; but I have no experience with TITLES; so I won't talk about them. I wil concentrate instead on judicious use of the first 2 blocks of fields.

An important factor to consider before making extensive use of the alternate fields is whether this information can be shared via GEDCOM files. Here is a FORM where we have entered text for all Name and Alias fields. We have even added extra Alias fields (by checking the "Insert ..Alias" buttons and clicking the button... the Modify Form re-opens with these additional fields).

The individual's page header now looks like this:

This starts with the Public_Name+Surname at the top; this will be printed whenever a reference to the person occurs on a page. Then we have - in order:

If we export this to Gedcom, the principal name and public (GIVN) name are listed first. Then, the Surname Aliases as a list on a single SURN line, and finally, the (full) aliases coded as extra NAME lines. But notice that the First_name Aliases are not exported - they are lost.

	0 @I1@ INDI
	1 NAME First_Names /Vaucher/
	2 GIVN Public_Name
	2 SURN SurN_Alias1,SurN_Alias2
	1 NAME Alias_1
	1 NAME Alias_2

This loss as well as the limited use of most Aliases - they are NOT indexed - reinforces our view that you should only enter Aliases when they are widely used alternatives to the principal name parts: i.e. Bill Clinton (the surname of his step-father) or Babe Ruth (who remembers his real first name?). The Notes is where private nicknames (i.e. "Fartface","Babs","BJ", etc...) should be mentioned.

Now we shall consider specific situations: married names, favored first name and long surnames composed of several words.

-- Back to the TOP --

Married Names

In many cultures, women adopt their husband's surname upon marriage. Thus, Cecilia Hart who married our actor James Earl Jones, would be also be known - and searched for - as Cecilia Jones. Similarly, Margaret Roberts is better known under her married name: Margaret Thatcher.

In Geneweb, this is handled automatically. For married women, name search can be done using either her principal name or by combining her first name part and the principal surname of her husband. Note that in the surname index, married women only appear under their maiden names.

In GEDCOM files produced by other software, married names may be recorded in different ways. In some cases, we have seen the SURN tag used for this purpose; in others ad-hoc tags (like _MARNM ) are used. For example:

	1 NAME Cecilia /Hart/
	2 SURN Jones
	1 NAME Cecilia /Hart/
	2 _MARNM Jones
When importing this into Geneweb, the first case would record "Jones" as a surname alias. In the second example, depending on the IMPORT option, the data would be ignored or added to the Notes.

In Switzerland, after marriage both husband and wife may be known by their joint surnames. In my family, John Vaucher married Stéphanie Mouchon and was thereafter sometimes referred to as John Vaucher Mouchon. Apparently, this is rarer nowadays and the wife may even invert the order; but in that case, a hyphen is used to indicate the inversion, as in: Stéphanie Mouchon-Vaucher. To allow search on the composite name you could use a Surname Alias.

-- Back to the TOP --

Usual first name

Quite often, culture dictates that persons be given many first names (often those of the grandparents), but they are generally known by only one (or 2) of them. My own birth certificates gives my full name as "Jean Georges Louis Vaucher" but no one knows me as anyting else but "Jean Vaucher".

In Geneweb, there are two ways to handle this, using either a "First name alias" or a "Public name." The Geneweb documentation recommends the first method (suported by a special IMPORT option). I disagree to some extent. As we shall explain later, the Public Name method is sometimes preferable and never should you use the "first given name preferred" option.

That said, here is what the official documentation says:

We suggest to put the common first name in the first name area. The other first names (complete first names, other possible first names, other spelling of first names, etc) must rather go to the area "first name alias" (below)....

Applying this method to my own case, my web page show up as:

My (shortened) principal name is at the top and the full name below. This is the only place where my full official name will appear. In other places, as a parent or a descendant, I would always appear as "Jean Vaucher".

Exported to GEDCOM, my data looks like this - a single NAME line with the preferred first name in quotes:

	0 @3@ INDI
	1 NAME "Jean" Georges Louis /Vaucher/
	1 SEX M
This coding scheme works both for export and subsequent import. It also works when the preferred appellation involves two or more consecutive words - even if they are not at the beginning of the list. For example:
	0 @3@ INDI
	1 NAME Jean "Georges Louis" /Vaucher/
	1 SEX M
But the is a PROBLEM If the "First name" is not a subsequence of the "First name alias." In that case, the official list - coded as "First name alias" - is not exported to GEDCOM and will be lost. Example:

when exported gives (losing Georges):
	0 @3@ INDI
	1 NAME Jean Louis /Vaucher/
	1 SEX M
Generally, this is not a problem as the usual first name(s) is part of the full list... or we can make sure it is. But it's a reminder that trying to use "First name aliases" to record nicknames like "Shorty" or "B.J." is not a great idea. It is better to mention them in the Notes area.

The First name OPTION:

When importing a GEDCOM file, there is an option to treat the first of the "First names" as the usual first name for all individuals. That is to treat

	0 @3@ INDI
	1 NAME Jean Georges Louis /Vaucher/
	1 SEX M
the same as
	0 @3@ INDI
	1 NAME "Jean" Georges Louis /Vaucher/
	1 SEX M
At first sight, this seems a useful option... leading to more compact lists, but I recommend against it. First with a small database, it is easy to put quote the usual first names of the people you know.

When you have a large database or import data from various sources, this option means that the people you know are mixed up with a multitude of strangers with the same first name. It makes close relatives hard to find. It also makes it harder to identify duplicates and clean up a database.

My suggestion: don't use the import option and use the recommended preferred first name technique only for close relatives and people you really care about.

-- Back to the TOP --

Public Name

The other option for a preferred first name is to leave the full list in the "First name" area and to enter the preferred one - when it is known - as the "Public name". In this case, the Public name (with the surname) will appear in all references to the individual... BUT it will not be entered in the search indexes, nor will it be used to locate the standard photos or in external HTML references. Those roles are reserved for the principal name which is still based of the first+last+oc combination.

The public name has its own GEDCOM tag: GIVN which works both for import and export of data. Importing GEDCOM files which use this convention automatically sets the "preferred name."

This public name option (basically doing nothing) is the one I recommend for the bulk of individuals in large databases: mainly distant ancestors and people you have imported from other databases. This option keeps the official names in indexes and displays facilitating duplicate identification. This is why I strongly discourage the "first name" import option.

CAVEAT: In many GEDCOM files I have received, the GIVN line systematically duplicates the first names in the NAME line and does not indicate "preferred" status. In such cases, I use a text editor to eliminate all such lines before importing.

-- Back to the TOP --

Surname Aliases

Surname aliases can serve in cases of adoption (i.e. Bill Clinton); but they are generally used to provide a shorter version of the family name.

Take, for example, the current president of France: born Nicolas Paul Stéphane Sarközy de Nagy-Bocsa, he is better known as Nicolas Sarkozy. Skipping over the first name problem, (ROGLO used technique 1), we need a surname alias.... BUT, now the choice is which version to use for the Surname and which for the Alias. Remember, surname aliases are not put in the indexes nor are they used for display; they serve only for search. It's your database... the choice is up to you!

Take a different example. In the Iberian tradition, surnames are a composition of the parents' (and even grandparents') names. For example:

	Felipe de Jesús Calderón Hinojosa (Spanish convention)
	José Eduardo Santos Tavares Melo Silva (Portuguese convention)
The usual surnames (underlined) derive from the father (patriarchal bias). I everyday life, the individuals would be known as Felipe Calderón and José Silva. Since the surname combination changes with every generation, there might be little advantage in entering the full official surname in the indexes and we might decide to use the usual short version as the "Surname" and the official version as the "Alias." This means that family descendants would be together in the indexes.

-- Back to the TOP --

Dit Names

In places, like Québec, soldier-settlers sometimes used their "nom de guerre" as a "dit" addition to their patronym, "dit" meaning called in French [Google references]. Here is a typical example:

En 1665, les frères Jean et Pierre Delguel dit La Brèche sont arrivés au Canada comme membres du Régiment de Carignan.

The "dit" names are inherited, much like some British double-barrelled names (i.e. Helena Bonham Carter). I suggest using the whole thing (Delguel dit La Brèche) as Surname and adding the patronym (Delguel) as an Alias. In this particular case, recent descendants are known as Déziel so I would also add that as an alias.

Surname addenda

In small towns, where a large number of the inhabitants had the same surname, it happens that qualifiers were added to distinguish the various family branches. In my case, browsing by surname shows two main branches, Vaucher de la Croix and Vaucher Des Moulins - in addition to some spelling variations.

The listing shows that I chose to enter the full surnames as principal name. Though it's not apparent, I also chose to enter the underlying patronym (Vaucher) as alias. The listing also shows the usefulness of browsing by surname instead of searching by surname because alternative spellings tend to be grouped together.

-- Back to the TOP --

Updated: dec 2011

Creative Commons License
GeneWeb Secrets by Jean Vaucher is licensed under a Creative Commons Attribution 4.0 International License.
Based on a work at