some events have unreasonably big dates

[VladimirAlexiev]

Histogram of number of events per date:

SELECT ?year (count(*) as ?c) {
  ?x a sem:Event; sem:hasBeginTimeStamp ?date
  bind(xsd:integer(replace(str(?date),"-.*","")) as ?year)
} group by ?year order by ?year

There's plenty of events in the near future, but I guess that's possible/reasonable:

However, there are also events in the very far future, which doesn't seem reasonable. Also, this prevalence of digits 9 in the dates is suspect: