On Fri, Mar 17, 2017 at 6:50 PM, Lyndon Nerenberg <lyndon@orthanc.ca> wrote:

> On Mar 17, 2017, at 2:08 PM, Dan Cross <crossd@gmail.com> wrote:
> While they tend not to use the bitmapped graphical icons of other systems, I argue that limiting the definition of desktops to being characterized by icons representing objects such as files and applications while being present on the screen seems like an implementation detail and unnecessarily limiting.

But I didn't say a desktop requires iconic representations of objects. I don't think the early Oberon implementations had them (but there are >20 years of memory loss between then and now).

Sorry; I thought that's what you were saying but I was wrong. But I confess confusion. For instance, you mention Oberon here as not having graphical icons but then in the next sentence two sentences it didn't meet your definition of what a desktop is. So that sort of seems like a non sequitur. What, then, is you definition? (And I'm not asking that to be combative; I'm truly interested.)

Was Oberon a desktop? Not to my mind. It was a bitmapped interface vs a text-cell-based interface to a cooperating group of programs. Conceptually I don't see any difference between Oberon and screen(1) in that regard. Would you consider screen a 'desktop'? And likewise, Oberon? I'm not asking this rhetorically. These concepts have fuzzy definitions for a lot of people, and I'm curious to see how they map out.

I would definitely call Oberon's graphical interface a desktop (btw, the graphical sorting demo was *cool*).

But I'm clearly using a different definition than you are.

- Dan C.