[Fis] Reply to António (A Priori Modeling)

Marcus Abundis 55mrcs at gmail.com
Sat Jun 18 12:24:02 CEST 2016


Hi António – thank you for your questions. I paraphrase your questions
below and respond in the order as given:

*• “How does a map constitute metadata?”*
You immediately strike the key issue.
Metadata is often known as “data ABOUT data.”
We might simplify this to “ABOUT-ness.”
Or, we might expand this to:
“*descriptors* ABOUT a *described*,” or “a MAP *about* a TERRAIN” to covey
targeted meaning.
This answers the "map as metadata" question.

But this example might be expanded further (as you suggest) to:
“a LEGEND, *about* a MAP, *about* a TERRAIN.”
But a map’s legend has “certain order” (why we know it is a legend). So we
now have:
“an ORDER, *about* a LEGEND, *about* a MAP, *about* a TERRAIN.”
You then mention the ALPHABET (ordered letters) used to create that legend
. . .

I hope you see the infinite regress I develop here, where all terms cover
ABOUT-ness – each is a type of metadata – except for the TERRAIN itself.
To answer your question, we must explore TYPES of DATA, and their
relationships (type theory). But then, we have still not begun to detail
the most important thing, the TERRAIN itself (why a map exists, das Ding an
sich). And, we have not named the *many* types of maps (tectonic,
geological, biological, hydrological, topological, etc.) possible for this
*ONE TRUE* TERRAIN.

In this infinite regress, we may attempt “truthful” exchanges ABOUT the
TERRAIN, but with no real grasp of the other person’s meaning . . . *unless*
we are clear on the types of data being discussed. For this reason, I use a
third analytic track (beyond simplify or expand) to *deconstruct*
(universalize) metadata in a priori terms, hoping to minimize confusion.

I am careful in my answer here as it covers a basic concept (oft taken for
granted, or forgotten?), that can lead quickly to complex confusion. One
must be bloody-minded in viewing this matter of "type," or confusion
quickly erupts. This is why I mention Deacon’s notion of “keeping our
levels straight” in my opening comments.


*• “How does delta z describe Darwinian fitness?” *
I am not sure if I understand your question. Delta Z, by itself, does NOT
describe fitness or a cost function. Delta Z merely describes “the ground”
prior to the emergence of Darwinian agents/agency. Thus, at *this level of
analysis* fitness is not relevant or even evident. You then ask . . .
*• “Should we interpret ‘meaning’ as quantifiable?”*
Yes, if we see survival as meaningful and quantifiable. For “higher-order”
functioning, parametric quantification is a complex matter. Here, going
further leads to a discussion of “types of meaning” and matters explored in
the paper (resource #2) “Is the Internet, etc Conscious” – an advanced
topic (a bit early to go into this?). You add . . .
*• “Why a Darwinian process and not another?”*
Because Darwinism is one or our most successful scientific models, and
offers a good starting point. I am aware of the problems with Darwinism
(again in the paper just noted), but this simple “scientific base” (even
with its problems) supports a “scientific effort” in modeling information.
I am interested to hear alternative foundations you find more interesting
or useful.


*• “What do you mean by ‘role’ and ‘model’ . . . one relating to function?
. . . isn’t 'model of the world' more useful?”*
This is another “map or terrain” question. Yes, the ROLE/terrain relates to
functioning (a “thing” that is DIRECTLY meaningful – survival/demise), and
MODEL/map refers to “other things” (diverse informational facets, many map
types). Both are useful, but in very different ways. I believe DIRECT
effective functioning is *most* useful (survival role), and INDIRECT models
of that utility are “very interesting” – if you take my meaning. If I ever
seem to confuse “role” and “model” (map or terrain) in my material, this is
an error – please tell me so that I can make the correction.


*• “How do you justify continuous “direct versus indirect” roles in the
informational reduction of entropy?”*
This is a big question, and actually invokes the entire body my work.
Still, I hope the infinite regress I name above (continuum), and the notes
just above (direct versus indirect, terrain versus map) offer insight on
how this question is already answered. If not, please let me know and I
will make a better effort.

Also, your note on “entropy gap” is well framed. In fact, THIS is THE KEY
MATTER! Shannon signal entropy must somehow tie to “material entropy”
(tendency to symmetric material dispersal). I label this “Shannon’s Gap” or
“a meaningful void.” I cannot say more here as I am unsure of what exactly
*you* mean by entropy gap. Still, one might see how an “informational
continuum” might conceptually close/fill that gap.


*• “Examples of the various entropy types you discuss would be helpful.” *
I am unsure how to respond as most of the video (to my mind) details just
such examples. The call for more examples here echoes your opening comments
– I am sympathetic.

Yes, the video is challenging, but then so too is the topic. If this were
an easy topic, central questions would have been answered long ago. More
examples push the video past an already-long 23 minutes. I purposefully
offer a video as this makes difficult material available in a concise way,
to be re-watched as needed, for clarity. If one picture is worth a thousand
words, a moving picture is worth thousands of pictures. I have done the
best I can (for now) and hope to make improvements, following the end of
this session.


*• “Why is Bateson not discussed simply as "relations between data”?*
The points you raise in closing your post – I think we largely agree. I
tend to focus on Bateson’s “Differences themselves must be differentiated”
rather than “a difference that makes a difference.” DIFFERENTIATED
differences, I think, speaks to the relational essence you (I believe)
point to. Also, I already point to the relational nature of things in my
first answer above. I am unsure I entirely grasp all that are you saying
here – but *my interpretation* of this relational essence is perhaps best
typified by delta S (in the video).

So, in closing, what you call for (I think) is what I offer.


Again, thank you for your interest and your questions.

Marcus
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://listas.unizar.es/pipermail/fis/attachments/20160618/888f13ed/attachment.html>


More information about the Fis mailing list