[LON-CAPA-dev] RFC: XML for <imageresponse> rewrite

[Guy Albertelli II]

Hi Guy,

> >> <numericalresponse unit="N/m" answer="15">
> >> <responseparam name="tolerance" type="tolerance" default="3%" />
> >> </numericalresponse>
> >>
> >> seems to follow a different philosophy.
> >
> > Because it right now only (in most usage) accepts a single number.
> 
> I think what Gerd is trying to say is that although a vector is 
> characterized by two numbers, the input accepts a single click on an 
> image. Gerd's tolerance refers to the click coordinates, while Guy's 
> tolerance refers to the interpreted result.
> I'm with Gerd on this one - we should be tolerant to the user's mode 
> of response.

Either mode would be tolerant to users response and equally
transparnet to the user about it. The difference is in
how the instructor thinks about it.

When he wants a vector, (which now is position independent and thus
not tied to any particualr locale on the screen). Does he think 

"I want a vector that looks like this one, but the head could be off
by this much."

Or

"I want a vector of this kinda angle and this kinda length."

Additionally. the instructor might care about the angle alot, but
length very little, and thus want to allow the user to draw a vector
of artibtraryish length but with a fairly accurate angle. (or vice
versa).

> The student does not calculate the angle and length of 
> the vector separately and reports it as two numbers, but graphically 
> appreciates where the vector's head should be in relation to the 
> diagram, and should be graded on this.

This would be an <arrow> not a <vector> (arrows care about absolute
position in the diagram vectors are positionless and thus arrows have
untweakable tolerances specified in the objects that underly their
<from> and <to>)


-- 
guy@albertelli.com   0-7-2-1-27,137