While it seems very ideal to have a "system ultimately used for applications original designer never anticipated" for, I can't think of an example readily. Does anyone have anything that comes to mind?
ferozenaina
It means that the design is highly re-useable - a problem from a different domain can be rehashed to be solved with our design.
Only example - OpenGL was designed robustly enough so it could be hacked to perform general purpose computations.
jmackama
I think I'm still a little confused by the use of the would compose. Is there a good example of features or primitives that compose versus those that don't?
Are nodes and edges and graphs examples of primitives that compose because they can have meaningful relationships with each other?
If you were thinking about adding a "pixel" primitive to a graph system, would that be an example of something that might not compose?
While it seems very ideal to have a "system ultimately used for applications original designer never anticipated" for, I can't think of an example readily. Does anyone have anything that comes to mind?
It means that the design is highly re-useable - a problem from a different domain can be rehashed to be solved with our design.
Only example - OpenGL was designed robustly enough so it could be hacked to perform general purpose computations.
I think I'm still a little confused by the use of the would compose. Is there a good example of features or primitives that compose versus those that don't?
Are nodes and edges and graphs examples of primitives that compose because they can have meaningful relationships with each other?
If you were thinking about adding a "pixel" primitive to a graph system, would that be an example of something that might not compose?