experts disagree on "call-by-reference"

[Gordon McMillan]

Greg Weeks wrote: 

>Lisp, CLU, Java, and Python implement objects, assign objects, and pass
>objects to functions similarly.  [Although Jave differs regarding its
>"primitive" types.]  So I was curious about how various Lisp, CLU, and
>Java texts dealt with the issue of "references" and "call-by-whatever". 
...

Back in '96 or so, I had a knock-down, drag-out fight on c.l.j.? (they call 
it a "discussion" there) with a well known Java book author and Sun 
employee. He would tell posters "Java is pass by value". I got him to 
reform that to: "Java is pass by value: for primitives, it's the value of 
the primitive, for references, it's the value of the reference."

I wasn't particularly happy with that either (he should've at least added 
"...*not* the referenced object"), but it's technically true, and it should 
ring warning bells in the heads of people who have been brainwashed by Herr 
Wirth into thinking that "by value" and "by reference" are magical things 
the language does for you, without ever stopping to consider how it works.

My experience with these folks is that most of them can grasp what a 
pointer is, but less than half ever get "pointer to a pointer".

Pascal-causes-brain-damage-ly y'rs

- Gordon