[Fis] Five Clouds over Fundamental Information Science

[Marcus Abundis]

Dear Xueshan,


Thanks for your interesting post.


I wonder if a term ‘inforware’ is truly needed, where
‘information/informatic domain’ mostly covers the same, including material
substrates — to ‘discuss’ material substrates is to ALREADY make ‘matter’,
abstract. Only truly-direct material experience conveys anything different.
Otherwise, I think inventing (unnecessary?) terms can confuse things,
rather than clarify.


For example, I have found it more useful to differentiate between
*informatic treatment* of ‘matter-in-itself’ (carbon and oxygen atoms,
etc.) in contrast to ‘matter specifically used to encode information’ (HDD
bits, DNA, etc.) I think both can be treated in ‘informatic terms’, but as
‘differed levels of abstraction’. I have the same reaction to notions
of communication,
computation, semiotics, cybernetics, adaption, etc. — I believe, simply
differed levels of abstraction that must be closely mapped in informatic
terms. This notion of ‘levels of abstraction’ dates back to Alfred
Korzybski, predating Shannon’s model. In fact, I wonder if your named
‘clouds’ actually attempt to ‘map levels of abstraction’ (without calling
it that). Even your cloud 5 seems to suggest such a unified informatic
mapping approach.


Your ‘*’ an ‘**’ notes seem to suggest ‘some things’ CAN be treated
informatically, and ‘other things’ CANNOT be treated informatically? What
EXACTLY is the difference between the two? Is that what you really think,
or did you have something else in mind?


Lastly, of course you likely already know of my own Theory of Meaning [ToM]
(presented at IS4SI 2021, and in the upcoming Mark Brugin memorial book)
that covers many of the issues you raise here, but which many FIS members
oddly seem to hope to ignore. I am never sure how to respond to this –
other than to refer people to the latest version of papers I work on . . .
but now targeting AI communities. The current draft is being revised for
the upcoming International Conference on Learning Representations, so the
last 2 pages are presently in a bit of a jumble, awaiting completion (end
December): A Simplified A Priori Theory of Meaning
<https://urldefense.com/v3/__https://drive.google.com/file/d/11r5LHrzAV2wAPXAPbEMxO58GuTR48hdP/view?usp=share_link__;!!D9dNQwwGXtA!QIGgryqf5nNEFVwrfy4r6VvwIrWH9wDQ6pNZtfPlGgan824tiyVQMLWwAXpdyEfnYpXOV0jLgXUxR6OC$ >.
But the prior 6 pages I think plainly address what you raise herein, while
later pages address 'further AI issues'.


I look forward to hearing more of your thoughts.


Marcus
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://listas.unizar.es/pipermail/fis/attachments/20241202/a4fe0878/attachment-0001.html>