Abstract vs. Concrete Labour
""Concrete work/labor" and "abstract labor" aren't two different _kinds_ of labor, they are just two different _aspects_ of labor. _All_ labor/work is concrete in so far as it produces some concrete use value; when we ignore the specific use value produced and just consider labor as "expenditure of human labor power", what's left is "abstract human labor" (cf. Wikipedia: http://en.wikipedia.org/wiki/Abstract_labour_and_concrete_labour ).
In capitalism this abstraction from specific use value is very important since it forms the basis of _exchange value_ (or just _value_): the value of a commodity depends on the amount of abstract human labor that's necessary to produce it. The abstraction therefore is a "real abstraction": it takes place in the real world, not just in our minds.
While it is clear that _all_ work, in any society, is always "concrete useful labor (or work)", it's a matter of debate whether it makes sense to speak of "abstract human labor" in regard to non-capitalist modes of production, where production is not based on (exchange) value and where the "real abstraction" doesn't take place. I'm inclined (but with no strong feelings one way or the other) to think that we can always speak of "abstract human labor", but that we should carefully distinguish whether this abstraction takes place in the real world ("real abstraction", as in capitalism) or whether it takes place in our minds only (as when we regard the work of somebody writing free software or a Wikipedia article as abstract "expenditure of human labor power", disregarding the concrete use value they're producing -- since they aren't producing any exchange value, this abstraction takes place in our mind only, it doesn't have an effect on the real world).
Now, the real abstraction in capitalism is an automatic process, it happens "behind the back of the producers" as a result of price negotiation between market participants.
On the other hand, abstraction in the peer economy, as I describe it, is the result of a conscious decision of people. A decision that they make in order to deal with a problem that they'll doubtlessly have to face, namely: how to distribute tasks that don't distribute themselves automatically" (Oekonux mailing list, September 2008)