It seems to me, as mentioned, that history will probably become less about evidence and more about the structure of the argument. Less about the manual accumulation of data–the splitting and chopping and stacking of fuel for the stove–and more about the context, the framing, and the discussion.
I often get asked why I study history when the majority of what I do in my “real life” is online and future-oriented. (And it’s worth noting that my like for the subject is often not explanation enough to the kind of people who ask that question.)
I think that blogging and history writing require similar skill sets, namely the ability to aggregate, analyze, and distill disparate sources into a coherent and well-reasoned narrative structure. In that sense, I can see where the author is coming from, but I disagree with his point that history will become less about evidence. A greater number of primary sources doesn’t make the future historian’s task any easier! The challenge just shifts from being about scarcity to being about surplus. Which sources do you choose and why? How does research change when we have too many sources?
That’s why I think that we’ll see future histories deal more overtly with data analysis. Tools like Netbase, Opinion Finder, or the Google- Profile of Mood States [PDF] can and will be leveraged for historical research. Might be a fun project to try something like that at Penn State. Let me know if you’re interested.