Guy Hoelzer  wrote or quoted:
> in article chjf63$1oeq$1{at}darwin.ediacara.org, Tim Tyler at tim{at}tt1lock.org
> > Guy Hoelzer  wrote or quoted:
> >> in article ch8pla$2spd$1{at}darwin.ediacara.org, Tim Tyler at
tim{at}tt1lock.org
> >>> Guy Hoelzer  wrote or quoted:

> >>>> Surely you acknowledge that the term
"information" is used by scientists,
> >>>> including those deeply involved with information
theory, in a variety of
> >>>> different ways, and that perceptive information is
only one of them.  For
> >>>> example, Steven Hawking must mean something
inconsistent with your
> >>>> position when he talks about black holes consuming
information and
> >>>> letting some of it back out.
> >>> 
> >>> A message-eating machine would destroy information related to the
> >>> messages - for all observers except the ones that already know
> >>> everything there is to know about the messages.
> >> 
> >> This seems utterly inconsistent with the perception-based view of
> >> information, which would assert that no information exists for an 
> >> individual if it has not yet been perceived by that individual. [...]
> > 
> > There's no problem with talking about whether an individual would gain
> > information if they witnessed a particular message.
> 
> I disagree.  You cannot know what any observer might take from a particular
> observation, because their interpretation is contingent upon their internal
> states and historical trajectories. [...]

It's certainly true that if you don't have any idea about what the 
observer already knows, then you can't say how much information they
will gain by observing a particular message.

However, you /can/ make statements in the general area.  For 
instance, if you have just witnessed them read an identical
message, it's a fair bet that witnessing another copy of the
message will not teach them very much - and so such a message
would transmit little information to them.

> >>> In the case of a black hole, there isn't likely to be any such
> >>> thing as an observer that knows everything there is to know about
> >>> the messages going into the hole.
> >>> 
> >>> So: I suggest Hawking's use of the term
"information" is quite
> >>> consistent with conventional usage in this particular case.
> >> 
> >> That was my point.  There are several very different
conventional uses of
> >> the term "information".  Hawking assumed a usage
similar or identical to
> >> mine, and inconsistent with the perception-based convention. [...]
> > 
> > We are at cross purposes: when I said: "conventional usage" I
> > was referring to the notion that information is dependent on the
> > observer.  As far as I can tell, this is what almost everyone
> > means when they use the term.  I was asserting that Hawking's use
> > was *not* inconsistent with the perception-based convention.
> > 
> > I fully expect that - if you were to quiz Hawking about what he means
> > by the  term "information" - he would refer to
observer-dependent,
> > Shannon-style information.
> 
> I doubt it, and I think that the perception-based notion of information is
> rarely the meaning assumed by physicists these days.

Some sort of survey would be needed to decide this point.

AFAICS, your view is totally unconventional - and practically everyone
is using the term "information" in a different sense to you.

> >>> Observers may differ regarding how *much* information the black
> >>> hole "destroys" - but they *all* agree that
they can no longer
> >>> gain access the information in the messages.
> >> 
> >> I agree.  This is consistent with my understanding, and IMHO
inconsistent
> >> with the perception-based convention because
"observers" unaware of the
> >> existence of the consumed structures WOULD NOT agree that
anything had been
> >> destroyed under that paradigm. [...]
> > 
> > IMO, they would agree about that issue.  Before the messages were
> > destroyed,  there were messages which could be read - and information
> > could be gained. 
> 
> Not if the observers were unaware of the existence of those
"messages,"
> which is of course the case regarding ourselves as observers and black holes
> as information-destroyers.  It would be non-sensible to even talk about
> information that had never been observed if one assumed the perception-based
> paradigm.  Under that paradigm information has no existence outside of the
> internal perceptions of observers, so the only way that a black hole could
> destroy information would be to destroy observers.  None of this makes sense
> to me.

Nor me :-(

> No such logical inconsistencies accompany the structuralist view of
> information, which has become the dominant view among physicists IMHO.

What is the "structuralist view of information"?

I tried:

http://google.com/search?q=structuralist+information ...and...
http://google.com/search?q=structuralism+information

....but didn't encounter much of relevance.

> > After the destruction of the messages, this is no
> > longer possible.  All observers are affected (though not equally)
> > except for observers who knew the entire contents of the message (and
> > if we are dealing with real material being swallowed, such observers are
> > hypothetical).
> 
> All observers must be hypothetical in a perception-based theory of
> information, because nobody can accurately characterize the perception of
> any other observer, let alone all other observers.  I still see this point
> of view as having very limited utility.

The notion that "All observers must be hypothetical" seems like gibberish
from a philisophy class to me :-(

I don't see the connection with the standard interpretaion of information 
- which has information dependent on the observer - though.

> It is a reasonable model for someone like Shannon focusing on the 
> context of telegraph communications between two people, but I think we 
> have grown far beyond the constraints of that context-specific model.

Shannon's information has proved to be a remarkably useful construct -
AFAICS.
-- 
__________
 |im |yler  http://timtyler.org/  tim{at}tt1lock.org  Remove lock to reply.
---
þ RIMEGate(tm)/RGXPost V1.14 at BBSWORLD * Info{at}bbsworld.com

---
 * RIMEGate(tm)V10.2áÿ* RelayNet(tm) NNTP Gateway * MoonDog BBS
 * RgateImp.MoonDog.BBS at 9/8/04 4:34:54 PM