Mark Wilden said:
I've used MI a buncha times. It's really not a big deal. I had problems with
it when I first used it. I also had problems understanding pointers.
Unfortunately I'm finding it hard to
Oh, I completely grant that any statement could be confusing to anyone. But
"could be" is different from "is."
Good job I originally talked about it being "potentially confusing" and
it being "likely to lead to confusion" then
I'd have to see an example.
"When a reference type value is passed by value, it is the reference
which is passed, not the object; it is still passed by value (by
default) rather than by reference, but you need to be clear that the
value in question is a reference."
I often find myself making statements *like* that (hopefully a bit
clearer) because people associate "reference type" and "pass by
reference" when they shouldn't - and I believe that's largely *because*
the word "reference" appears in both.
I know it seems like a bad idea to you. I just don't know if anyone actually
has trouble understanding it. You understood it. Dustin understood it. I
understood it. No one that we know of doesn't understand it. So there's
actually no evidence that the definiton is hard to understand.
No, but I think it's
That said, it doesn't hurt to be absolutely clear. I was really just talking
to my "hypothetical idiot" point (which is admittedly a pet peeve of mine).
I think the difference is that I'm considering a hypothetical newbie,
not a hypothetical idiot. Were you an idiot when you didn't understand
pointers? No - just new to the idea.
In this case, we don't even need to look for a hypothetical newbie -
the post was a response to a *real* newbie (garyusenet) who has never
hidden his "newbieness". What we *don't* know is whether he found it
confusing, or whether he would have found it confusing if there hadn't
been all the other posts talking about types, or whether it may have
laid the foundations for confusion when he learns about interfaces.
I do argue that that's a valid statement. It's completely, 100% true.
The example isn't pertinent here, though, since this newsgroup is about C#
(and explanations thereof), not music, and one's knowledge of C# is quite in
evidence here.
And in this case, the OP's "newbieness" is in evidence too.
I do have evidence for my belief about the definition:
1) Based on my knowledge of and experience with programmers, I don't think
the definition would be confusing to most or even some of them.
Programmers with how much experience though? Having had to try various
ways of explaining certain concepts in the past, I believe a C# newbie
could very easily be confused by the different uses of "interface" in
the statement, either when they first hear it or later when they then
start to use C# interfaces.
2) No one has expressed confusion.
Hence, 3) It seems that the definition is not confusing.
I disagree with the logic there, because it's so easy for it to have
confused someone a) without them knowing or b) without them wanting to
display that ignorance.
Do I -know- that invisible pink hippos don't orbit Saturn? It just seems
pretty likely that they do not, based on reasoning enumerated above.
Given that I disagree with you on point 1, it's unsurprising that I
disagree with the conclusion.
Pace Donald Rumsfeld, the absence of evidence -is- evidence (though not proof) of
absence.
It's evidence, but not particularly strong evidence - whereas I believe
that the evidence of common sense is that overloading terms
unnecessarily (within a context where there is a specified meaning of
those terms) leads to an unnecessarily high potential for confusion. In
this case, I think the potential is great, and you disagree, which is
fine.
I've seen enough people getting confused by relatively clear
descriptions of the type system to want to err on the side of caution.
