I guess that's how abstractions work in general. You start reasoning in them explicitly when the need (or benefit) to do so is sustained for some time. Then you discard them when they're not needed anymore.
There's a difference between "abstraction" as used in computer science and software development, and "abstraction" as an intellectual act or process.
The former itself seems to have a few different but related uses, but generally it is often a matter of modeling.
The latter is generally that act of the intellect that draws out something from a whole. So it's kind of an opposite process, in a very loose sense. The concept "greenness" is abstracted from images (or phantasms, in classical parlance) of particulars we encounter through the senses. We never encounter "greenness" in the world as a thing, only things which are green.
Where risk is concerned, I would expect the intellect to seize on the concept through abstraction (2), but any modeling of risk would involve the use of (1).
I fail to see the difference between these two abstractions. Is modeling, as you put it, not drawing something out from a whole? I think if abstraction meant different things in different contexts, the definition would be to blame.
The former itself seems to have a few different but related uses, but generally it is often a matter of modeling.
The latter is generally that act of the intellect that draws out something from a whole. So it's kind of an opposite process, in a very loose sense. The concept "greenness" is abstracted from images (or phantasms, in classical parlance) of particulars we encounter through the senses. We never encounter "greenness" in the world as a thing, only things which are green.
Where risk is concerned, I would expect the intellect to seize on the concept through abstraction (2), but any modeling of risk would involve the use of (1).