Re: Another Perspective on the Cause of the Keyhole Problem

"White Wolf" <[email protected]> Sat, 8 Nov 2003 17:54:48 +0200
Newsgroups gmane.comp.programming.keyholes
Message-ID <000701c3a610$a3543c00$1e90ba50@Evo>
Scott Meyers wrote:
> How can somebody say "Who does he think he is?" if the argument is
> compelling?

Just like that. :-)

> Sigh.
>
> That was a rhetorical question.  Sometimes compilers are so much more
> reasonable than people.

Compilers have no soul (or blues ;-) ), but people do have.  Eh, what
really matters here is pride and rationalization.  Have you ever seen
those firedoors in sci-fi films?  They are shut within a second as the
alarm goes off.  This is how people stop listening if you push the wrong
button.  That very fireproof glass door shuts down and you never get
through again.  I am not saying you do push that button, I have no idea,
I am not (yet) a web designer.  Others - however - warned you might.  It
is up to you now. ;-)

We (C++ people) are used to your frank style - and try not to think
anything more to it than what you actually say.

IMO there is a big difference between what you do now and the C++ books.
The C++ programming language has a standard.  There are implementations
out there.  So the topic is (more or less) a rather objective one.
However with keyholes, you talk about perception.  So with C++ there
were objective rules/facts to refer to and one viewpoint - the
developers who wants to understand C++ and make good C++ code (no
crash/fast etc.).  There is no real need for a motivation - other than
technical proof.

In the case of keyholes there are two viewpoints and it is more about
subjective perception rather than objective rules.  I mean there might
be rules out there (there are, you are discovering them) but they are
not part of any standard or programming language definition.  So in the
case of keyholes you need to do more/something different.  You do not
only need to show (already motivated) people how to do it right, but you
need to convince people that they are doing it wrong - without pushing
their buttons...

Your C++ readers bought the book because they thought that they do not
know enough and do not do their tasks effective enough.  With the
keyholes topic you need to "sell the idea" to people that they are *not*
doing their job right, while still keeping them reading the book.  I
think this is the most important difference...

White Wolf