Hi Michael

Indeed, it was a hard sell, but maybe ANIMA was just too early? Or SDN was 
doing it all wrong? (in the typical deployments)

I am not sure, but certainly incorporating "intelligence" in the network - as 
opposed to leaving it at the ends of it - would require a strong, survivable 
control system, as mentioned below.

The question hence is: should we try to claim that the "agent stuff" is best 
positioned in ANIMA? Would it fly? For network management, I certainly see fit. 
However, most of the current agent discussions rather talk about applications 
(app exchanges on top of HTTPS). While it explains the difficulty, why most of 
it seems completely irrelevant to the IETF (DNS for agent discovery is the most 
relevant thing I have heard - maybe I am wrong), most of those "use cases" 
would hardly be ANIMA-related (would user agents use BRSKI? Pledges? ACP? 
Why???).

So, ANIMA-wise we have a clear path towards "agentic NMS" or "intelligent 
orchestration" as use cases, so far so boring 😊 What else could there be? 
App-to-network interactions? To do what? Service discovery? Transport 
guarantees? Negotiations? Sovereign (source) routing? Positioning? ICN? Should 
we rather talk about "distributed systems" than networks? Should all agents, 
all my data sources, my cloud instances, etc. be registered with "my domain" 
and span an ACP of their own?


Regards
Artur


-----Original Message-----
From: Michael Richardson <[email protected]> 
Sent: Wednesday, July 8, 2026 9:12 PM
To: Artur Hecker <[email protected]>; Anima WG <[email protected]>
Subject: Re: [Anima] Re: I-D Action: draft-eckert-anima-ai4an-00.txt


Artur Hecker <[email protected]> wrote:
    > - if we are serious about ACP/GRASP/BRSKI as foundation, we would need
    > to show that, even in spite of potential misjudgments by LLM or
    > programming errors done by it, ACP would always remain "the last
    > Mohican": ACP should be the last to fail, i.e. in presence of a
    > potentially unverified programs produced at will by the LLM/foundation
    > model, the ACP must act like an implementation of a protected mode in
    > an OS, like an OS kernel if you will, preserving the possibility to
    > repair/remove/terminate misbehaving programs. Such that we can have a
    > bunch of simple yet reliable tools on top of ACP to clean away any
    > nonsense introduced by LLM/foundation models, and, also very important,
    > to be able to isolate the effects of such programs by separating
    > concerns and enforcing "control app manifests" in a distributed system.

+1
I totally agree, but as you said, it's been a very hard sell.

--
Michael Richardson <[email protected]>   . o O ( IPv6 IøT consulting )
           Sandelman Software Works Inc, Ottawa and Worldwide

**       My working hours and your working hours may be different.         **
** Please do not feel obligated to reply outside your normal working hours **




_______________________________________________
Anima mailing list -- [email protected]
To unsubscribe send an email to [email protected]

Reply via email to