Ulrich

Your use of the phrase "say 100 years after the event", led me to believe that you were arbitrarily pulling the number out of a hat. Had you provided this information earlier I doubt that I would have taken you to task, sorry for the derail.

Personally, I would say that both sets of material should be viewed as highly suspect. Of course the longer you go between an event and its recording, the more chance you have of getting things entirely wrong, but I think 40 years is sufficient time for even an eyewitness to screw up nearly every detail.

I sometimes have problems remembering exactly what my boss told me 5 minutes ago, especially when it concerns working as opposed to visiting IIDB.

John A. Broussard

By now this should be an eternal verity. Eyewitnesses are notoriously unreliable. Just a cursory check of Psychology journals will give you plenty of evidence. Best one I've heard about recently was where someone in a gorilla suit walked across a basketball court in the middle of the play. Later interrogation of witnesses found that a majority of them never even noticed what should have been very obvious.