Re: Ditch OWL:Thing = god? SemWeb 2.0

1 view
Skip to first unread message

Owen Ambur

unread,
Sep 9, 2022, 12:21:20 PMSep 9
to timothy...@gmail.com, paul....@gmail.com, AboutThe...@googlegroups.com, ch...@chriscfox.com
Timothy, responsibilities can be documented as roles in StratML Part 2, Performance Plans and Reports.  Logically speaking, they should be linked to goal and objectives.

As the saying goes, a goal is a dream with a deadline.  The difference between a strategic plan (set of goals and objectives) versus a performance plan is that the latter specifies stakeholder roles and performance indicators by which accountability can be ascribed, progress can be measured, and success can be recognized.

I've removed those whom I previously copied because I don't want to overload them with TMI that may not be useful to them.  The reason I copied them is that I have interacted with them about StratML services.  I still hope to engage some of them in helping to develop StratML-enabled query services and thought the information in our exchange might be of some use to them.

I'm looking forward to the day that the benefits of RDF/OWL services can be applied to documents in StratML format and particularly the >5K files currently available at https://stratml.us/drybridge/index.htm, all of whose URLs are listed in sitemap format at https://stratml.us/sitemap.xml.

That's why I responded to your message and remain interested to learn whether you and/or Paul might be interested and capable of making progress along the lines outlined in my plan at https://aboutthem.info/SQS.xml.

Of course, too, I'm always interested in learning when others have plans that have been sufficiently well-thought-out to warrant rendition in StratML format.  Please let me know if you have the makings of such a plan(s).

BTW, I share your interest in making dialog like this publicly available.  So I'm copying my Google discussion group site at https://aboutthem.info/  However, I'm even more interested in reducing the need for "dialog" by making it easier for those who have common and complementary objectives to "cut to the chase".  See https://www.linkedin.com/pulse/cut-chase-skip-nonsense-proceed-directly-doing-good-better-owen-ambur as well as the informal plan that Chris Fox and I drafted to move to a higher level of information management maturity than E-mail.



-----Original Message-----
From: Timothy Holborn <timothy...@gmail.com>
To: Owen Ambur <owen....@verizon.net>
Cc: paolad...@googlemail.com <paolad...@googlemail.com>; pradee...@ictect.com <pradee...@ictect.com>; kra...@thoughtflow.io <kra...@thoughtflow.io>; carlma...@gmail.com <carlma...@gmail.com>; acu...@01communications.com <acu...@01communications.com>; Paul Werbos <paul....@gmail.com>
Sent: Thu, Sep 8, 2022 9:42 pm
Subject: Re: Ditch OWL:Thing = god? SemWeb 2.0

Hi Owen & others,

I've cc'd paul - who i'm collaborating with a great deal at the moment.  I would prefer these sorts of conversations to be on the lists / public, either via the peace project list or w3c - or both!  but as that's not the circulation of this email, i didn't add it to them.  

Feel welcome to do so, if the 'non-list' consideration was in error...  or, per below, if its not considered w3c cg relevant, nb: https://groups.google.com/g/peace-infrastructure-project - which is WIP. (work in progress).  There's also a discord group, to help with more in-depth conversations about particular topics, etc.  but the online infrastructure is presently in its infancy (notwithstanding the historical works to draw upon, etc.).

I think firstly - i hope my considerations were not interpreted as to suggest that i want to do away with OWL entirely!  this would be in-correct, there's an enormous amount of ontological infrastructure built from this predicate concept of 'thing', that's often extensively useful - yet, there's more to life than things.  thereafter - is owl the suitable solution to support 'scientia et vita omnia'...  whilst constructing solutions (as has been the mission, for a while now) that's 'human centric', where wisdom is out-of-band (maintained in the minds of natural observers, etc.); empowered by knowledge, curated via technology of the time (we're a bit beyond libraries of papyrus, or indeed in other regions - songlines - yet - we struggle with the issue of 'fake books', as may be the appropriate translation...?);  as such, the considerations are not about restructuring the way we represent considerations relating to 'things', rather, i question the universal appropriation of the term & whether there's something i've missed - or whether there's a real-problem, that might be something worth addressing (regardless of how well its supported).  
  

On Thu, 8 Sept 2022 at 06:46, Owen Ambur <owen....@verizon.net> wrote:
Timothy, I don't feel qualified to comment on this matter in any detail but here are a couple thoughts about your personhood/human rights ontology,

I don't feel qualified, yet also - when i speak to the experts, it doesn't appear to me that i'm alone with this feeling of not being qualified :) 

important issue to address, in-time, given the role ontologies play - with how our world is organised, etc.


 
for whatever they may be worth:
  1. Some folks may object to your use of the word "brotherhood" in the description of the ontology.
I asked a local indigenous man, what is the indigenous language word (in this area) for 'peace be with you', or 'shalom' or 'As-salamu alaykum',  he responded - and said the terms translated as - learn to be brothers - he then me more insights - and he said - the additional term translated as - learn to be honest...  

So, whilst this likely not quite  (given the context, we were both out at night when the opportunity to learn, occurred) - its a description of a consideration, where -  it seems to be useful to investigate & consider - various forms of social ontology, and figure out how to support the many; rather than shoe-horning people into a few - which is of-course, distinct to how others may in-turn form inferences between personal ontologies / group ontologies, that may in-turn change overtime...  


  1. Human rights cannot be usefully separated from responsibilities.
Entirely agreed..  Noting - i suspect there's a bit of vocab out there for responsibilities - i'll have to check where i'm upto with that generally otherwise - but there wasn't ontology for human rights;  i started that work a while back https://github.com/WebCivics/ontologies - circumstances at the time, complex, i won't bore you with it all - but since, there's been some work done - whilst an absence of work in other, contributory areas of importance, imo.


 
This ngram hightlights the problem:


I personally am not so interested in how people describe (identify) themselves or to what they believe they are "entitled" simply because they exist as a human "thing" as I care about: 

a) what they plan to do with their lives, 
b) what they are actually able to accomplish, and 
c) how it impacts others.  

Human Entity - or indeed, quantum entity - consciousness with its interactions being notated - to have both internalities & externalities, often thereby also - in our modern times - some level of interference via software agents (ie: autocorrect); 

 

That pretty well sums up why I care more about StratML than RDF/OWL.  On the other hand, of course, I am always open to learning more about how RDF/OWL might be use to support my interests.  For example, I'd love to see someone demonstrate the benefits of its usage as applied to the >5K files in the StratML collection.

The vision of the StratML standard is:  A worldwide web of intentions, stakeholders, and results.

Another way of characterizing it is as the *Strategic* Semantic Web -- linkages not merely for their own theoretical sake but in support of the practical realization of human objectives.

I think there was a discussion about StratML - perhaps about 8 years ago?  i'll have another look at it.

there's other areas (other than the topic of this convo where - from the best of my recollection - StratML had a bunch of invaluable work, yet to be improved upon?  (conversion across to RDF (from XML) was one of the problems?  maybe i'm confusing StratML with something else, idk. 

generally speaking - the broader international / earth - works - i'm curating via https://groups.google.com/g/peace-infrastructure-project/c/JrYcxP9Bazc 

which is a bit different to the historical 'web civics' concept - although involving works, that most certainly relate.  

There's some broad-ranging work required to address 'digital slavery' 'endentured survitude', other real-world problems; both, more broadly, and also in how we ensure people - can contribute towards the development of solutions to address these sorts of problems without being expected to be sacrificed, as independent human beings - able to stand-up for the needs of our human family (as is sometimes distinct to the interests of patent-pool related W3C members via their representatives (who participate as an agent / representitive of those group entities) - not for the purpose of deploying rents (usury), but to ensure support for acknowledgement & stuff that has otherwise been the subject of widespread concerns, over a long period of time.


some of the latter diagrams noted: https://medium.com/webcivics/inforgs-the-collective-info-sphere-67a660516cfd might help.  but as noted, i think forming a modern ecosystem might be better done via a group - that reflects much of what W3C CGs do - but tries to address the social challenges, peace - infrastructure - in effect - as we've got no right to engage in violence for gainful purpose, nor be agents in support of acts of war, whether via methods defined as war today - or methods associated to areas in which its called a 'not a war', notwithstanding the lived experience of persons / discouraged senses of 'common sense' - that its certainly not, broadly - a peaceful time...  nor is lawful remedy available, to the many, all too often...  

Hope That helps... 
 

Timothy Holborn



-----Original Message-----
From: Timothy Holborn <timothy...@gmail.com>
To: Paola Di Maio <paolad...@googlemail.com>
Cc: W3C AIKR CG <publi...@w3.org>; public-cogai <public...@w3.org>
Sent: Wed, Sep 7, 2022 11:16 am
Subject: Re: Ditch OWL:Thing = god? SemWeb 2.0

How is W3C not the international standards body for commercial infrastructure of the use of 'things', from a technical standpoint.

On Thu, 8 Sept 2022, 1:13 am Timothy Holborn, <timothy...@gmail.com> wrote:
oh, overall; whilst the existing modalities are so very well supported - does that give rise to ensuring other modalities are made unable to emerge?

in english, at least, i guess..  

On Thu, 8 Sept 2022 at 01:12, Timothy Holborn <timothy...@gmail.com> wrote:
I've been out tonight, so might not be the best idea to distil the complex concept into a single line... but..

are you nothing more than a thing?

Should we promote this ideology in academic curriculum materials for all young people - so they know they're place (/rights, essentially) in the world...?

Timothy Holborn.

Timothy Holborn.

On Wed, 7 Sept 2022 at 17:17, Paola Di Maio <paola....@gmail.com> wrote:
Timothy, I cc the AI KR CG list, because OWL is described as KR

My understanding (Not an OWL expert myself) is that owl:thing is necessary due to the choice of OWL being based on logical hierarchy, whreby thing is an abstraction for the upmost entity in the upmost category
 and if it makes you feel better, you could probably mask it with another label (such as owl:anything) without breaking it

I dont think there is a case for ditching or breaking OWL,  but rather for understanding itse limitations, there is some good reading pointing to possible workarounds
You could of course look further and start thinking beyond OWL

this seems a good read in that direction

Empowering OWL with Overriding Inheritance, Conflict ...

https://www.aaai.org › Papers › Symposia › Spring
PDF
by S Hosain · 2009 · Cited by 9ing multiple different types of inheritance with overrid- ing, and non-monotonic reasoning
The popularity of OWL for knowledge representation
in the Semantic Web applications makes it an attractive
platform. Although OWL supports some form
of object-oriented features for knowledge structuring
and maintenance, it is significantly weak in capturing
most essential object-oriented features such as single
and multiple inheritance, default class values, methods,
overriding and encapsulation in their true spirits.
It is also weak in extending reasoning support for intelligent
knowledge processing. Such features are becoming
increasingly essential in applications such as
social networks, e-commerce and knowledge rich ontology
for Life Sciences. In this paper, we propose
an extension of OWL toward a more powerful knowledge
structuring language, called OWL++, by supporting
multiple different types of inheritance with overriding,
and non-monotonic reasoning capabilities within
OWL. We demonstrate OWL++’s computability and
implementability by presenting a translational semantics
of OWL++ to OWL, for which we have robust execution
engines while for the reasoning component of
OWL++ we rely on Jena to support rules in OWL.

Also I hear SHACL could be useful used in conjunction with OWL

There is tons of stuff be the looked up I think, as starting points for workarounds
and beyond OWL futures

Let us know if you have the chance of conducting further analysis
and send us a summary of your findings

PDM

On Wed, Sep 7, 2022 at 1:51 PM Timothy Holborn <timothy...@gmail.com> wrote:
Hi all,

I'm still befuddled as to whether my analysis has merit or if there's something basic I'm missing. 

I noted a concept sometime ago about "human centric web" or "human centric AI", etc.  (Early credentials CG work). 

Ontology Dev environments like protege use https://www.w3.org/2002/07/owl therefore anything modelled after it is a subclass of owl:thing

I started on a personhood ontology

Yet, now I'm trying to do some modelling for consciousness & various aspects relating to human agency, where the idea of structuring it all as a subclass of owl:thing, churns my stomach.

I'm thinking about forming a broader upper ontology, and thereafter the implications.

I'm also considerate of DIDs, which, from my point of view was always about ontologies on DLTs (particularly commons), understanding - making tools, isn't necessarily about a particular usecase / implementation structure. 

So, thinking is; if there's a time to break owl:thing (providing diversity) perhaps you hat time is now? 

Or am I missing something simple / fundamental, etc. 

The underlying consideration is impacted by modalities, whereby there may be a lack of diverse options available; if the tools aren't present to do it, distorting the wave function, via "things", perhaps unnecessarily / impactfully,  imo. 

Timothy Holborn.

Timothy Holborn

unread,
Sep 9, 2022, 12:27:26 PMSep 9
to Owen Ambur, paul....@gmail.com, AboutThe...@googlegroups.com, ch...@chriscfox.com, peace-infrastr...@googlegroups.com
Hi Owen,

its late here - i'll follow-up tomorrow, but have cc'd the peace-infrastructure group.  

just quickly - I'm fairly sure StratML has different (likely complimentary) objectives to the modelling i'm working on for natural agents;  who a person is, their conscious experience, the basics - before the relationships linked to work activities, etc...  although linked - as to consider new forms of violence / wrongs, etc. 

Tim.
Reply all
Reply to author
Forward
0 new messages