Joshua Porter has undertaken a mammoth task to document the Evolution of Information Grazing. Its great to get this all online so I'll add my tuppence worth here.
I don't agree with Josh's use of the term "Feed Grazing" at all. He defines it as -
"... what most of us do in our feed readers: checking a bunch of feeds to see if there is anything new. Usually the reader notifies us of new content, by either starring the content or sorting it based on newness, or both."
But my definition of "Feed Grazing" refers to our relationship with the feeds and not how much attention we pay to them. What Josh is referring to above is subscription and aggregation. But that's where Grazing is fundamentally different because we are not subscribing to individual feeds. Grazing is a way of accessing the latest items in feeds without subscribing to them, at all.
The Answers.com definition of Grazing I quoted previously is "To touch lightly in passing". Subscription is the act of committing to a feed. Even if we pretty much ignore it afterwards we have still committed to it by giving it a bed in our house. Its a marriage. If we want to be rid of it after we must go through a painful divorce. Grazing, on the other hand, is feed promiscuity. No commitment.
I'd also disagree with Josh's interpretation of Grazing Lists as "a variation of Feed Grazing". As I wrote previously -
It seems that there is a consensus around the idea that a Reading List should be 10 feeds or less so I would define a Reading List as a manually generated low volume, low frequency (of change) list. Whereas a Grazing List is a high volume, high frequency list which is more likely to be generated programmatically, as with Adam's Tech.Memeorandum mashup.
In other words a Grazing List is merely the fodder for a Feed Grazer.
As regards the "Evolution of Grazing" I would say it will depend on Social Bookmarking evolving into Feedmarking, evolving into Nodemarking. In fact I can't see why del.icio.us could not build rudimentary OPML trees on the fly right now based on a tag weighting algorithm. But then again I'm no programmer ;->
Thanks again Josh for the effort to document this and to tie down the definitions.