Every tree has a few facts that cannot be true. Most are typos: 1852 entered as 1825, a death date pasted into the birth field. Some are wrong assumptions: a marriage assigned to the wrong John Smith. They are worth catching because other people copy them, and because a chart with a mother born after her child looks careless even when everything else is right.

What GED-VIEW checks

  • Death before birth, or burial before death.
  • A birth more than about nine months after the father's death.
  • A parent younger than about twelve, or older than about sixty (mother) or eighty (father), at a child's birth.
  • A marriage before either party was about twelve, or after either had died.
  • Baptism or christening before birth.
  • A person alive past about 120 with no death recorded.
  • Children of one couple born less than about nine months apart, unless twins.

Each is reported on the Issues page with the people involved and links to their profiles.

The thresholds are deliberately loose. The point is to catch errors, not to argue with the occasional genuine outlier. A woman having a child at forty-eight is unusual and real; a woman having a child at eight is a typo.

What is not an error

  • Approximate dates. "About 1850" and "before 1855" are legitimate and GED-VIEW treats them as ranges.
  • Same-name children. Two children called John in one family usually means the first died young. Worth checking, often right.
  • Old-style dates. Files using dual dating (11 Feb 1731/2) or the Julian calendar can trip a naive check. GED-VIEW reads the date as written and does not convert.

What to fix first

  • Anything where the fix is obvious. A birth of 1925 for someone whose parents were born in 1890 is a transposed 1952.
  • Anything that involves a duplicate. Two entries for one person often each have some right dates and some wrong ones. Merge first, then fix.
  • The ambiguous ones last. If a marriage is dated after a death and it is unclear which is wrong, that is a research question, and the answer is a record, not a guess.

Working the list

Fix the obvious ones in the desktop program and re-export. Pin the ambiguous ones so they stay visible. Hide the ones that turn out to be fine (the same-name children, the genuine outliers). Export the list if it is easier to work from paper or a spreadsheet.

Next

Open Issues and read down the list. Fix the obvious ones today.