When one is writing history, it obviously helps to talk to people who were around when the events happened. Now, WWII is 68 years in the past, and most of the people who fought in it have passed on. So, what can we expect in future historian’s works about it? Of course, you have earlier historian’s works to fall back on-bt can we expect a great deal of “revisionist” history, once the last people who experienced WWII are gone?
Memoirs, news reports, official documents . . . people still find new interpretations and data for the Civil War, and the last veteran died decades ago.
What is your definition of most? There are still a million Americans alone who are WWII veterans.
Germany wins!
Documents, letters, diaries, family history (which wholes embellished can be possessed of a surprisingly degree of accuracy).
History as a discipline is constantly evolving. It’s possible and indeed even likely that future generations might not see it in the same manner as we do. Hell, the Pacific War is already seen as evil versus evil in some parts o Asia.
The meatier part of a historian’s work isn’t to simply establish a set of facts of what transpired. It is to put events in context, by explaining why they happened, what the significance is, and how events fit together. Eyewitnesses aren’t particularly good at most of that stuff. The challenge is finding evidence scattered all over the place, analyzing the credibility of it, and putting into a coherent story. That evidence could be talking to eyewitnesses, reading autobiographies of people involved, finding documents scattered around the globe in different languages, and so on… each of which should be checked against each other to establish credibility of the source, and understanding why different people see the same event in much different ways.
Substantially more than half.
You look at all available sources, decide what should have happened, then tailor the narrative according to your own biases.
The main source of new information now will be government documents as they become declassified and released.
Here is something I found interesting with regard to reconstructing historic events.
The famous Kruschchev shoe-banging event at the UN:
Namely the last paragraph, if the truth of an event that happened within living memory, with hundreds of eye-witnesses and television cameras present can’t be definitively established how can we be sure of events hundreds of years ago. 
Considering that people are still re-interpreting ancient Greek and Roman history, that is a near-certainty.
Every generation of scholars comes up with a new spin on the old stories. They try to correct the mistakes of their predecessors, and they make new mistakes. It is a never-ending process.
Consider George Washington:
One historian will worship him as the Father of the Country.
Another will condemn him as a rich white slave-owner who didn’t want to pay taxes.
Another will praise him as a hemp farmer and whiskey distiller.
The only thing eternal in history IS the revisionism.
Eyewitness accounts recorded about seventy years after an event would be treated by any smart historian with an extra large dose of scepticism. This was a point that oral historians of the First World War had realised as early as the 1960s - veterans’ memories are unreliable and they get much worse over time. Which is why most historians of the Second World War already rely far more on the various forms of surviving contemporary sources and on those eyewitness accounts recorded much nearer the time. There is therefore no obvious reason why the deaths of remaining eyewitnesses should make much difference to their working methods.
There’s some fascinating stuff in that article. One person claimed “the heel of the hand that held the shoe slammed the desk but that shoe never actually touched it.” I’m trying to picture how someone could hold a shoe and strike a desk with the heel of the same hand without the shoe touching the desk. I can’t do it. Then there’s the Life photographer who wrote that he is “certain” that Khrushchev “did not bang his shoe on the desk,” but that “he certainly meant to do so.” How could he know what someone else meant to do?
Eyewitness testimony from folks like this doesn’t seem to be worth much.
I don’t think it’s particularly common for historians to interview people who lived through the event being studies, for a number of reasons. Most of what we consider “history” happened before anyone currently alive was even born. Traveling around to interview elderly survivors would be time consuming and expensive. The elderly person’s memories may be unreliable, and even if they have perfect recall of their own experiences then the average WWII soldier probably did not have any personal experiences that would shed much light on the causes or effects of the war or any lingering mysteries about exactly what happened in a particular situation. There’s certainly a place for first-hand accounts of ordinary people affected by historic events, but it’s not like WWII veterans were a small or poorly documented group of people. We have plenty of military records, personal letters and diaries, photographs, period newspaper stories, etc., covering what they saw, did, and thought during the war.
There are some fields of study where firsthand oral accounts may be the best or only source. When I was a student I worked for my university’s digital library, and one of the collections we were working on making available online was a bunch of recordings from the 1940s where elderly people (who’d been children in the late 19th century) were asked to recite or sing the songs they remembered from their childhood. If no one had bothered to talk to these people while they were still alive then some of these folk songs might have been completely lost to history or at the very least local variations of well-known songs might have been lost.