Re: Agentic Linked Data

Hi.
My two cents: The linked data / sem-web notion of agents is quite 
different from 'trust' and 'legal' notion of agents because the former 
consider 'agency' as in what can 'act', while the latter consider 
'agency' as in who is this representing. Though I haven't seen this 
particular discussion in the WebAgents agenda/minutes so far (caveat: I 
haven't been paying close attention), it would be worthwhile to engage 
with them to see overlaps/distinctions.

In DPV, we have dpv:Representative and dpv:hasRepresentative to model 
one entity representing another (legal entity). We didn't model agents 
so far due to the ambiguity and potential confusion/misinterpretation 
between linked data agent meaning everybody and legal agent meaning 
something different and incompatible. However, we have a planned 
discussion for the next version (Q2 2025) to represent agents: 
https://github.com/w3c/dpv/issues/197 as MachineAgent and LegalAgent - 
including expressing who authorised them, who controls them, and how 
they act (my personal wishlist). This also helps separate 'person' from 
'machine' in context, and our goal is to do this to express 'agents' in 
a way that helps the legal context (e.g. this privacy agent made this 
decision on my behalf).

P.S. odrl:isA is defined as "A set-based operator indicating that a 
given value is an instance of the right operand of the Constraint." - I 
wouldn't use just this relation to have the desired interpretation for 
asserting something is an 'agent' in this context.

Regards, Harsh

On 12/01/2025 21:22, Knut-Olav Hoven wrote:
> How about this:
> 
> _:mybot a foaf:Agent .
> _:me foaf:made _:mybot .
> 
> It doesn't say I created the technology behind "mybot", but I made this 
> resource/thing/configuration and thus inherently control it. Doesn't 
> mean I have to trust everything this agent does...
> 
> And I wouldn't let it write directly into my documents.
> 
> I would then use some vocabulary to express my trust in the statements/ 
> documents it creates. And probably copy selected statements into my own 
> documents.
> 
> --
> Knut-Olav Hoven
> 
> 
> søn. 12. jan. 2025, 20:28 skrev Joshua Cornejo <josh@marketdata.md 
> <mailto:josh@marketdata.md>>:
> 
>     IMHO, ontologists would argue that the definitions for linked data
>     should use the FOAF ontology for the definition of most “general”
>     classes and stick to corresponding definitions …____
> 
>     __ __
> 
>     http://xmlns.com/foaf/spec/#term_Agent <http://xmlns.com/foaf/spec/
>     #term_Agent>____
> 
>     http://xmlns.com/foaf/spec/#term_Person <http://xmlns.com/foaf/spec/
>     #term_Person>____
> 
>     __ __
> 
>     The Person <http://xmlns.com/foaf/spec/#term_Person> class is a sub-
>     class of the Agent <http://xmlns.com/foaf/spec/#term_Agent> class,
>     since all people are considered 'agents' in FOAF.____
> 
>     __ __
> 
>     FOAF also defines Organisation as a type of agent. Other ontologies
>     extend into Automatons/etc. ____
> 
>     __ __
> 
>     Others like DCAT <https://www.w3.org/TR/vocab-dcat-3/> bind Dublin
>     Core and FOAF:____
> 
>     __ __
> 
>     dcat:resource <https://www.w3.org/TR/vocab-dcat-3/
>     #Property:resource_creator> has a property dcmi:creator <https://
>     www.dublincore.org/specifications/dublin-core/dcmi-terms/#creator>
>     with a range of foaf:Agent <http://xmlns.com/foaf/spec/#term_Agent>.____
> 
>     __ __
> 
>     ___________ ____________________________
> 
>     *Joshua Cornejo____*
> 
>     *marketdata <https://www.marketdata.md/>____*
> 
>     smart authorisation management for the AI-era____
> 
>     __ __
> 
>     *From: *Frederick Gibson <frederick@graphmetrix.com
>     <mailto:frederick@graphmetrix.com>>
>     *Reply-To: *<frederick@graphmetrix.com
>     <mailto:frederick@graphmetrix.com>>
>     *Date: *Sunday 12 January 2025 at 18:27
>     *To: *Josh Cornejo <josh@marketdata.md <mailto:josh@marketdata.md>>
>     *Cc: *Melvin Carvalho <melvincarvalho@gmail.com
>     <mailto:melvincarvalho@gmail.com>>, Jesse Wright
>     <jesse.wright@jesus.ox.ac.uk <mailto:jesse.wright@jesus.ox.ac.uk>>,
>     public-solid <public-solid@w3.org <mailto:public-solid@w3.org>>
>     *Subject: *Re: Agentic Linked Data____
> 
>     __ __
> 
>     “Though Tim noted he’d be wary of something calling itself both a|
>     Person|and an|Agent|!”____
> 
>     __ __
> 
>     If we look at the OED definition of agent:____
> 
>     __ __
> 
>     "A person who or thing which acts upon someone or something; one who
>     or that which exerts power; the doer of an action. Sometimes
>     contrasted with the patient (instrument, etc.) undergoing the
>     action."____
> 
>     __ __
> 
>     "A person who acts as a substitute for another; one who undertakes
>     negotiations or transactions on behalf of a superior, employer, or
>     principal; a deputy, steward, representative; (in early use) an
>     ambassador, emissary. Also figurative. Now chiefly in legal
>     contexts."____
> 
>     __ __
> 
>     By definition, a person can be both type person and type agent, just
>     as a person can have countless types depending on the state of a
>     person at a given time.  And given that "thing which..that which" is
>     included in the first definition, it would be consistent that any
>     system could be an agent acting on behalf of another system (whether
>     artificial or organic), not just people as agents.____
> 
>     __ __
> 
>     Fred Gibson____
> 
>     /Founder & CEO/____
> 
>     /mobile: 415.335.8232/____
> 
>     __ __
> 
>     1255 Treat Blvd, <https://www.google.com/maps/
>     search/1255+Treat+Blvd,+Suite+300?entry=gmail&source=g> Suite 300
>     <https://www.google.com/maps/search/1255+Treat+Blvd,+Suite+300?
>     entry=gmail&source=g>
>     PMB#4611____
> 
>     Walnut Creek, CA  94597____
> 
>     /office: 925.940.0741/
> 
>     ____
> 
>     cid:0.28873829620.3205515023989717847.1945bc4a168__inline__img__src____
> 
>     __ __
> 
>     __ __
> 
>     __ __
> 
>     __ __
> 
>     __ __
> 
>     ---- On Sun, 12 Jan 2025 09:25:59 -0800 *Josh Cornejo
>     <josh@marketdata.md <mailto:josh@marketdata.md>>* wrote ---____
> 
>     __ __
> 
>         __ __
> 
>         “Though Tim noted he’d be wary of something calling itself both
>         a|Person|and an|Agent|!”____
> 
>         __ __
> 
>           ODRL has a :isA operator to work around these type of
>         validation/specialisation semantics. (Equivalent to typecasting
>         in programming)____
> 
>         __ __
> 
>         __ __
> 
>         __ __
> 
>         _______________________________________
> 
>         *Joshua Cornejo*____
> 
>         *marketdata <https://www.marketdata.md/>*____
> 
>         smart authorisation management for the AI-era____
> 
>         ____
> 
>         ------------------------------------------------------------------------
> 
>         *From:*Melvin Carvalho <melvincarvalho@gmail.com
>         <mailto:melvincarvalho@gmail.com>>
>         *Sent:* Sunday, January 12, 2025 4:26 pm
>         *To:* Jesse Wright <jesse.wright@jesus.ox.ac.uk
>         <mailto:jesse.wright@jesus.ox.ac.uk>>
>         *Cc:* public-solid <public-solid@w3.org <mailto:public-
>         solid@w3.org>>
>         *Subject:* Re: Agentic Linked Data____
> 
>         ____
> 
>         __ __
> 
>         __ __
> 
>         so 28. 12. 2024 v 12:34 odesílatel Jesse Wright
>         <jesse.wright@jesus.ox.ac.uk
>         <mailto:jesse.wright@jesus.ox.ac.uk>> napsal:____
> 
>             Hi Melvin,____
> 
>             __ __
> 
>             Have you come across the Web Agents CG https://www.w3.org/
>             community/webagents/ <https://www.w3.org/community/
>             webagents/> - there are several people working in that
>             direction there.____
> 
>         __ __
> 
>         Hi Jesse,____
> 
>         Thanks for pointing out the Web Agents CG. Yes, I’ve been a
>         member since its inception—an excellent group!____
> 
>         Quick note, as not all responses made it to the mailing list
>         over the holidays. I used an LLM to summarize some key points:____
> 
>         *Summary of the Discussion on Agentic Linked Data*____
> 
>         *Participants*____
> 
>           * *Melvin Carvalho*: Initiated the discussion.____
>           * *Joshua Cornejo*: Experienced in ODRL and authorization
>             architecture.____
>           * *Eric Jahn*: Interested in agent roles and permissions.____
>           * *Aad Versteden*: Integrates AI and Linked Data in
>             semantic.works.____
>           * *Sindhu Raju*: Focuses on personal data management and
>             consumer use cases.____
>           * *Matt Taylor*: Consultant in ethical tech, cautious about
>             broad agentic access.____
>           * *Ram Mukund Kripa*: Developed privacy agents for consent
>             management.____
>           * *Jesse Wright*: Highlighted the Web Agents CG.____
>           * *Jacob Beauchamp*: Exploring agentic learning and entity
>             data.____
> 
>         There’s clear interest in moving this forward. I also mentioned
>         Agentic Linked Data on the SolidOS call, where the idea (of
>         agents) was well received. For context, TimBL’s “timblbot”
>         illustrates prior work:
>         https://timblbot.inrupt.net/profile/card#me <https://
>         timblbot.inrupt.net/profile/card#me>
>         Classes include:____
> 
>           * |schema:Person|, |schema:SoftwareApplication|, |foaf:Agent|,
>             |foaf:Person|, |prov:SoftwareAgent|.____
> 
>         (Though Tim noted he’d be wary of something calling itself both
>         a |Person|and an |Agent|!)____
> 
>         Next steps: We’re probably too small for a standalone Community
>         Group. A practical option might be to work within the Solid CG
>         for now, with a dedicated chat area for agentic topics. Initial
>         work items could include adapting |Agent|entities to existing
>         Solid tooling, libraries, pods, servers, and apps.____
> 
>         Looking forward to further discussion and collaboration!____
> 
>         Best,
>         Melvin____
> 
>         ____
> 
>             __ __
> 
>             Best,____
> 
>             Jesse ____
> 
>             __ __
> 
>             Get Outlook for Android <https://aka.ms/AAb9ysg>____
> 
>             ------------------------------------------------------------------------
> 
>             *From:*Melvin Carvalho <melvincarvalho@gmail.com
>             <mailto:melvincarvalho@gmail.com>>
>             *Sent:* Saturday, December 28, 2024 11:11:09 AM
>             *To:* public-solid <public-solid@w3.org <mailto:public-
>             solid@w3.org>>
>             *Subject:* Agentic Linked Data____
> 
>             ____
> 
>             Hi All,____
> 
>             When I started this group, it was with the idea that Solid =
>             Social Linked Data. Solid's foundation on WebID ties it to
>             the concept of "Agent," based on FOAF's Agent class (parent
>             of FOAF Person).____
> 
>             With "agentic AI" really taking off right now, I wonder—
>             would anyone be interested in exploring "Agentic Linked
>             Data"? It feels like a natural direction.____
> 
>             Best,____
> 
>             Melvin____
> 
>         __ __
> 
>         _______________________________________
> 
>         *Joshua Cornejo*____
> 
>         *marketdata <https://www.marketdata.md/>*____
> 
>         smart authorisation management for the AI-era____
> 
>     __ __
> 
> 
> 
>     ____
> 

-- 
---
Harshvardhan J. Pandit, Ph.D
Assistant Professor
ADAPT Centre, Dublin City University
https://harshp.com/

Received on Sunday, 12 January 2025 22:02:06 UTC