Re: dynamics: comment about MBdyn...

On Wed, 16 Jul 2003, Pierangelo Masarati wrote:

[...]
> Actually, what you're suggesting is in contrast with  one
> of the peculiarities of Modelica -- one that allows A LOT
> of flexibility and code reuse. 
Indeed! And that flexibility is the reason I don't feel comfortable
with it :-)

[...]
> In this sense, how can you leave "dangling" connections,
> according to the factory pattern, and force users to
> resolve them later, provided multiple connections are
> legal and multiple combinations of connections can result
> in a legal model?  

It's not a problem to reconnect connections at any time. It's also no
problem to have partial models, as long as they are consistent. And
no, one single object without any connection is _not_ a valid partial
model. Even a free floating single rigid body needs to be connected to
the "world" before it can be of any use. In other world, a simulation
based on a model is only valid if the model is "grounded" to a know
reference. So, if you send me a complex Modelica file and I cannot be
sure that it represents a valid model, ready for simulation, than you
are doing me a big disservice...

I am all for flexibility, but it should be structured :-)

Herman

-- 
  K.U.Leuven, Mechanical Engineering, Robotics Research Group
<http://people.mech.kuleuven.ac.be/~bruyninc> Tel: +32 16 322480