Most real world computer architectures use some form of
goto (computed and
not computed) in their internal form.
Until the 1960s it was also common for
this to be done in
high level source code.
Then an impractical computer
scientist named Wirth
published a paper called "goto considered harmful",
followed
by an avalanche of educational tools and toy languages to
instill
this attitude into millions of junior programmers
all over the
world.
As a consequence, some "pointy haired" manager types who
only
shallowly understand the tech have issued rules banning
their more intelligent
underlings from using goto and
computed goto even when they cannot be
efficiently replaced
by "structured programming" notations such as while,
repeat,
if and for.
Experienced, intelligent programmers know and
recognize
when goto and goto-like constructs are more efficient
solutions to a
problem than the kludgy tricks needed to
achieve similar results without them.
And anyone who has
actually looked at compiler output (and understood it),
knows that all those fancy high level constructs compile to
goto and computed
goto anyway.
And as the other poster noted, if goto was mentioned in
today's testimony at all (it wasn't, they discussed data
access), it was about
the compiler output, not the source
code.
[ Reply to This | Parent | # ]
| 
