[Rd] variable scope in update(): bug or feature?

[Michael]

On 22 Dec 2006, Brian Ripley wrote:

> 'update' will update and (by default) re-fit a model. It does this
> by extracting the call stored in the object, updating the call and
> (by default) evaluating that call. Sometimes it is useful to call
> 'update' with only one argument, for example if the data frame has
> been corrected.

Thanks.  I understand now that this is the expected behavior per the
documentation.

It is just that when I call 'update (m1, ~ . - z)' I did not expect x to
change (or everything can be different if mydata has changed).  I only
wanted to evaluate the old model with old data but sans the z variable.

Would it be useful to have an option in update() not to update the data
(i.e., behaves more like drop1())?

Michael