Re: The wisdom of the object mentors

Dmitry A. Kazakov wrote:
> On 27 Jun 2006 07:43:22 -0700, Marshall wrote:
>
> > Dmitry A. Kazakov wrote:
> >> On 26 Jun 2006 19:06:43 -0700, Marshall wrote:
> >>
> >>> Bart Wakker wrote:
> >>>> frebe73_at_gmail.com writes:
> >>>>>
> >>>>> A algorithm could must obviously know about the data structure.
> >>>>
> >>>> Not at all! I'm currently writing many algorithms that get their data
> >>>> passed in as java objects. The algorithm does not need to know where
> >>>> the data came from and how it is stored in the database.
> >>>
> >>> "Where the data came from" is not the data structure.
> >>
> >> Replace "where the data came from" with "the data structure at the place
> >> they came from."
> >
> > The claim was that it is not the case that an algorithm must know
> > about the data structure it operates on. This claim is bogus on
> > the face of it. Your introduction of the idea of different schemas
> > for different modules does not change how bogus it is.
>
> That depends on how we'd define both, which has a danger to move us into an
> exercise in amateur philosophy. Especially whether we are talking about
> inputs and outputs or internal states (both in some imperative mindset), or
> about ADTs (mixed), or wider about abstract mathematical structures (purely
> declarative). Consider an algorithm of enumeration of any [infinite] set in
> ZF. I don't know what's the data structure there. So, as a behaviorist, I
> don't believe in existence data.

What on earth is going on with you that you cannot admit to a simple truth, that an algorithm must know the data structure that it operates on, and must instead try to obscure the issue by bringing in lots of irrelevant abstractions. Why are you mentioning ZF?

I really dislike Bob's snake-oil salesman hypothesis, but I have to admit I haven't a single other plausible hypothesis that would explain the above.

Marshall Received on Tue Jun 27 2006 - 17:59:33 CEST