There are some clueless "researchers" and "genealogists" who use "first date of reference to the person" as "date of birth", when it's usually 20+ years too late. I've run into this from time to time.
This one http://freepages.genealogy.rootsweb.ancestry.com/~hwbradley/aqwg697... thinks c. 1070 is more likely, and they may be right.