Meeting notes

August 5, 2008

John, Julia, and Syd met to discuss the initial personography specification:

Parts of personal names, ordering of names, etc.

For authors, we're assuming that we will be dealing with a surname, one or more first names, and one or more middle names. For some of our authors we will also have a title or honorific of some sort, and possibly additional name(s) that include generational or epithet information. Female authors will often (though not always) have a birthname that we will record as well. We want to have an algorithm that will order these various pieces correctly.

The typical case will involve a surname, a forename, and a middle name. Less typical will be cases that involve more than one fore/middle name and various kinds of titles or honorifics. To handle these cases, we can specify a sort order for the appropriate name elements.

Difficult case: how to deal with the difference between inherited and acquired titles (e.g. Lady Mary Wroth vs. Mary, Lady Chudleigh). We said we could add @type to rolename specifying whether the title in question is inherited, acquired, religious, or some other type we haven't yet identified.

Taxonomy of roles for people

Julia proposed that rather than using a strict taxonomy, we could employ a "faceted" approach that permits us to apply multiple descriptors from multiple domains to any given person. Using the existing FileMaker taxonomy as a starting point, we decided that it makes sense to divide role types into three general domains: "cultural," "geographic," and "functional"

Cultural = the general cultural field from which the person comes or within which the person has (or had) some particular significance; possible values for a controlled vocabulary include: history, myth, fiction, classical, biblical, and proverbial\(?)
Geographic = the broad geographic region from which the person originates or in which she/he resides; values would be based on continent with a more precise set of terms for subdividing Anglophone regions: North America, South America, Africa, Asia, Europe; British colonies, British Isles
Functional = the domain of textual production that includes aspects of authorship, publication, circulation, and intertextual reference; possible values include: author, imprint, reference

Other information to include in personography entries for authors

We discussed a number of possible categories of information pertaining to textbase authors (these are in addition to the basic set that includes names, date of birth, date of death, dates of primary activity, sex, and key):

Marital status
Religious affiliation
Parenthood
Languages spoken/read
Place of birth
Place of death
Socioeconomic status
-Education-
-Occupation-
Nationality or country of habitual residence

Marital status

Decided to capture information about authors' marital status as an event that occurred one or more times before death, rather than as a state. Discussed using from= and to= or notBefore= and notAfter= before deciding this was cumbersome and unnecessary, since all we're interested in is indicating that a person had experienced a particular state (or states) at some time during her life.

We seemed happy with the idea of using <event> with a child <desc> element and controlled vocabulary to indicate the specific type of marital event:

marriage
divorce
annulment
widowing

Single will be the default state; in cases where there is no <event>, it is assumed that she/he remained single throughout her/his life.

Religious affiliation

Decided to use the TEI <faith> with a controlled vocabulary to specify an author's known religious affiliation(s). Like marital status, this will simply be a way of indicating the faith (or faiths, in the case of individuals who undergo one or more conversions) an individual claimed at any point during her life.

Our proposed vocabulary would be:

Orthodox
Catholic
Church of England
Other Protestant

We did not discuss non-Christian faiths, though we may want to do so in the future.

Parenthood

Decided to treat this as a binary state using the <state> element with an @type that can take either "parent" or "childless" as its value. Use of an optional <desc> element would permit us to provide some descriptive detail in cases where we think it will be useful (e.g., "Two children, both of whom died in infancy" or "Three miscarriages"). For our purposes, we decided to treat all pregnancies, even if they resulted in miscarriage or death, as a form of parenthood.

Languages

After initially deciding not to deal with languages known, we revisited the decision in light of the fact that a number of references sources include this information when it is available. We decided to use a <langKnown> element for this information, with an @level for indicating the level of language proficiency.

August 24, 2007

Jacque, Brian, Syd, and Julia met to discuss the scope and strategy for this proposal.

NamesGrantMeetings

Contents