How Floruit is Calculated: Difference between revisions

From London Book Trades
Jump to navigation Jump to search
← Older edit
VisualWikitext
No edit summary
No edit summary
 
(4 intermediate revisions by the same user not shown)
Line 1: Line 1:
<div style="width:90ch;">
<div style="width:90ch;">
A person's floruit — their period of professional activity in their chosen trades — is computed algorithmically from dated evidence across the database.  
A person's floruit — their period of professional activity in the book trades — is computed algorithmically from dated evidence across the database. The algorithm draws on records of apprentice bindings and freedoms, addresses, court appearances, Stationers' Company offices held, and entries in the masters table. Where direct professional evidence is unavailable, it falls back to life dates (birth, death, baptism, burial), age-at-event records, and the date ranges assigned by the original editors of the London Book Trades index — taking account of annotations such as "(active)" and "(died)" to distinguish professional activity from life events. Events relating to relatives are excluded, evidence dated after a person's death is discarded, and statistical outlier detection removes stray dates that fall far outside the main cluster of activity. The result is a start year, an end year, and a confidence grade for each.
The algorithm draws on records of apprentice bindings and freedoms, addresses, court appearances, Stationers' Company offices held, and entries in the masters table. Where direct professional evidence is unavailable, it falls back to life dates (birth, death, baptism, burial) and age-at-event records to estimate a plausible working life, and as a last resort uses the date ranges assigned by the original editors of the London Book Trades index. Events relating to relatives (e.g. "son — Patrimony") are excluded, evidence dated after a person's death is discarded, and statistical outlier detection removes stray dates that fall far outside the main cluster of activity. The result is a start year, an end year, and a confidence grade from A to E.


;A:Both the start and end dates are supported by direct records of professional activity: taking on apprentices, holding a trade address, attending court, holding office in the Stationers' Company, or similar.
The start and end dates are graded independently on a confidence scale from A to E:
;A:The date comes from a direct record of professional activity: taking on an apprentice, holding a trade address, attending court, holding office in the Stationers' Company, or similar.
;B:The date is derived from a life event or editorial attestation: an apprentice binding date plus seven years, a death or burial year, a birth date plus twenty-one years, or a date marked "(active)" in the original editors' index.
;C:No individual dated evidence was found. The date is taken from the editors' date range in the original London Book Trades index, without further annotation.
;D:No evidence was found for this endpoint, or the original date range was implausibly long and had to be narrowed algorithmically. These dates should be treated with particular caution.
;E:A date was computed from the evidence but was then adjusted by a sanity check — for example, an end date that fell after a known death was capped to the death year, or a start date that implied an implausibly young age was moved forward.


;B:At least one of the start or end dates comes from direct professional records; the other is derived from life events such as apprentice binding, freedom, death, or birth dates.
Note that some people will have single year floruits, eg, 1761-1761. This is because the source data only has one item with a date and there is no other basis for establishing a broader floruit.
 
[[Category:Front Matter]]
;C:No direct professional records were found. Both dates are derived from life events — for example, a floruit start estimated from a binding date plus seven years of apprenticeship, or an end date taken from a death or burial record.
 
;D:No individual dated evidence was found in the database. The floruit is based on the date range assigned by the original editors of the London Book Trades index.
 
;E:Evidence was limited or conflicting. Only one endpoint could be determined, or the raw evidence produced an implausibly long span that had to be narrowed. These dates should be treated with caution.

Latest revision as of 12:28, 31 March 2026

A person's floruit — their period of professional activity in the book trades — is computed algorithmically from dated evidence across the database. The algorithm draws on records of apprentice bindings and freedoms, addresses, court appearances, Stationers' Company offices held, and entries in the masters table. Where direct professional evidence is unavailable, it falls back to life dates (birth, death, baptism, burial), age-at-event records, and the date ranges assigned by the original editors of the London Book Trades index — taking account of annotations such as "(active)" and "(died)" to distinguish professional activity from life events. Events relating to relatives are excluded, evidence dated after a person's death is discarded, and statistical outlier detection removes stray dates that fall far outside the main cluster of activity. The result is a start year, an end year, and a confidence grade for each.

The start and end dates are graded independently on a confidence scale from A to E:

A
The date comes from a direct record of professional activity: taking on an apprentice, holding a trade address, attending court, holding office in the Stationers' Company, or similar.
B
The date is derived from a life event or editorial attestation: an apprentice binding date plus seven years, a death or burial year, a birth date plus twenty-one years, or a date marked "(active)" in the original editors' index.
C
No individual dated evidence was found. The date is taken from the editors' date range in the original London Book Trades index, without further annotation.
D
No evidence was found for this endpoint, or the original date range was implausibly long and had to be narrowed algorithmically. These dates should be treated with particular caution.
E
A date was computed from the evidence but was then adjusted by a sanity check — for example, an end date that fell after a known death was capped to the death year, or a start date that implied an implausibly young age was moved forward.

Note that some people will have single year floruits, eg, 1761-1761. This is because the source data only has one item with a date and there is no other basis for establishing a broader floruit.

Retrieved from "https://lbt.bibsoc.org.uk/index.php?title=How_Floruit_is_Calculated&oldid=783710"