Hi Matthias,

Let me explain a bit about what that final parameter is being used
for.  But first let me state that if you use circles to draw Venn
diagrams, then for three or more circles there are many consistent set
relationships that cannot be drawn accurately.  An example of this is
three equal size sets, A, B, C, where each set overlaps each other set
by say 50%, but the intersection of A, B, C is zero.  Using circles to
draw this, the overlap of A^B^C won't show up in your diagram as
zero.  This is an example of a very common kind of relationship, and
one that can't be drawn perfectly accurately using circles.  So what
the algorithm does for cases like this is try to come up with a layout
that is "pretty good."  Where "pretty good" means "not too
misleading."

In a nutshell, the algorithm works like this:
1. Figure out the radius for A, B and C.
2. Figure out how far apart the center of A has to be from the center
of B to satisfy the pair-wise constraint, and do the same for B & C,
and A & C.
3. If distance relationships satisfy the triangle inequality, then it
lays them out in a triangle, and then try to further improve the
layout taking into account A^B^C by using an optimization technique to
try to find a layout that is the least misleading.
4. If the distance relationships do not satisfy the triangle
inequality, then the center of C is placed at the intersection of the
edges of A and B and the stochastic hill climber is started.  In these
cases an "ideal" layout is much less likely to occur.

Does this help explain what the final parameter is used for?

Justin

On Dec 10, 12:00 pm, Matthias Kluwe <[email protected]> wrote:
> Hi!
>
> IMHO the seventh parameter of the data v set in cht=v seems to be
> redundant. The sizes of the three circles are clearly fixed by the
> first three parameters and their relative position by the parameters
> four to six.
>
> It would be helpful for the user if the documentation mentioned
> this... The last parameter for the should be optional, a mathematician
> would say it should not exist.
>
> Regards,
> Matthias
--~--~---------~--~----~------------~-------~--~----~
You received this message because you are subscribed to the Google Groups 
"Google Chart API" group.
To post to this group, send email to [email protected]
To unsubscribe from this group, send email to 
[email protected]
For more options, visit this group at 
http://groups.google.com/group/google-chart-api?hl=en
-~----------~----~----~----~------~----~------~--~---

Reply via email to