[ih] A revolution in Internet point-of-view - Was Re: Internet analyses (Was Re: IPv8...)

[Joly MacFie]

Since this thread is now up to 48 messages and I am having trouble keeping
up, I had ChatCPT do a summary:

*[ih] A revolution in Internet point-of-view – Internet-history mailing
list discussion (April 28–30, 2026)*
Participants

Karl Auerbach; John Day; Vint Cerf; Hesham ElBakoury; Miles Fidelman; Steve
Crocker; Andrew Sullivan; Dave Crocker; Jack Haverty; Bob Purvy; Brian E.
Carpenter
Moderator

Mailing list discussion (Internet-history at ISOC)
Context and Framing

This extensive mailing list exchange begins with Karl Auerbach’s
observation that the Internet is undergoing a “revolution in perspective,”
and develops into a layered discussion combining technical history,
governance models, and future concerns about fragmentation, centralization,
and user understanding.
Karl Auerbach — The Internet as Applications, Not Infrastructure

Karl Auerbach argues that the dominant mental model of the Internet has
shifted:

   -

   Early practitioners focused on *packet transport, protocols, and
   end-to-end connectivity*
   -

   Modern users see the Internet as a *set of interoperating applications*
   (messaging, social media, video, maps)

This abstraction shift means:

   -

   The underlying network has become an *invisible utility*
   -

   Users are largely indifferent to core technologies like routing, DNS, or
   transport protocols

He connects this shift to a broader structural concern:

   -

   The original vision of a *single, seamless global network* is being
   eroded by
   -

      National and regional policies
      -

      Security and control mechanisms
      -

      Cultural and political fragmentation

This creates conditions for *localized or alternative networking models*,
potentially diverging from the traditional Internet architecture.
John Day — Infrastructure Still Matters and Education Gaps

John Day supports Auerbach’s analysis but stresses that infrastructure
remains decisive:

   -

   Even if users ignore it, *network design shapes application behavior*
   -

   Trends like CDNs and edge computing reinforce *regionalization at the
   infrastructure layer*

He raises a key long-term issue:

   -

   How should new generations be taught?
   -

      As operators of existing systems
      -

      Or as thinkers who understand *design tradeoffs and past mistakes*

This highlights a risk that institutional memory is fading as abstraction
increases.
Vint Cerf — Internet Literacy and the “Driver’s License”

Vint Cerf reframes the discussion around user knowledge:

   -

   Proposes the idea of an *“Internet Driver’s License”*
   -

   Asks what the *average user should understand* before navigating the
   digital ecosystem

The exchange distinguishes between:

   -

   Everyday users
   -

   Infrastructure engineers
   -

   Researchers and developers

It underscores the growing *gap between system complexity and user
comprehension*.
Miles Fidelman — Internet vs Web Misconception

Miles Fidelman notes a persistent conceptual confusion:

   -

   The Internet is often conflated with the Web
   -

   In reality, the Internet supports multiple functions:
   -

      Email
      -

      Voice and messaging
      -

      Video conferencing
      -

      Commerce and services

He suggests that operators and engineers may have failed to adequately
*communicate
the breadth and importance of underlying network services*.
Andrew Sullivan — Governance, Centralization, and Alternative Histories

Andrew Sullivan expands the discussion into governance and history:

   -

   Revisits the contrast between *Internet (TCP/IP)* and *OSI models*
   -

   Highlights arguments that OSI’s more inclusive governance was *slower
   and less effective*

He raises broader questions:

   -

   Could today’s Internet have emerged in a different *political or
   economic environment*?
   -

   Would alternative architectures have been more resistant to *centralization
   and consolidation*?

He also links current issues to the *rise of the Web*, which he suggests
has inherently *centralizing tendencies* compared to the distributed
Internet architecture.
Steve Crocker — ARPANET Origins and Open Model

Steve Crocker provides historical grounding:

   -

   ARPANET development was funded by *ARPA/DARPA*, but
   -

   The *primary users were the researchers themselves*

Key characteristics of the early model:

   -

   Open sharing of results
   -

   No strong intellectual property constraints
   -

   A culture of *collaboration and experimentation*

He emphasizes that openness was not a later decision — it was *built into
the system from the beginning*, enabling widespread adoption.
Dave Crocker — IETF vs OSI: Process and Outcomes

Dave Crocker analyzes differences between standards processes:

   -

   *IETF model*:
   -

      Rough consensus
      -

      Open participation
      -

      Iterative development with running code
      -

      Fast, pragmatic progress
      -

   *OSI/ISO model*:
   -

      Formal representation of stakeholders
      -

      Consensus via structured deliberation
      -

      Slower, more bureaucratic processes

He characterizes the distinction as:

   -

   Internet approach: *intersection of workable ideas*
   -

   OSI approach: *union of all stakeholder requirements*

This difference significantly affected *speed, scalability, and adoption*.
Jack Haverty — Licklider’s Vision and Human-Centric Computing

Jack Haverty brings in deeper historical context via J.C.R. Licklider:

   -

   Licklider envisioned *“man-computer symbiosis”*
   -

   The Internet evolved as a tool to support *human communication and
   collaboration*

He notes that:

   -

   Today’s user perception aligns more with Licklider’s vision than with
   protocol-level thinking
   -

   The Internet’s success stems from meeting broad *human and
   organizational needs*, including military and corporate C3I functions

This reinforces the idea that the Internet was always about *human activity*,
not just technical infrastructure.
Ongoing Debate — Standards, Power, and Technical Choices

Later exchanges become more pointed, especially involving John Day and
others:

   -

   Debate over whether IETF decisions were:
   -

      Technically optimal
      -

      Politically influenced
      -

      Or overly conservative

Examples discussed include:

   -

   TCP/IP vs alternatives
   -

   SNMP vs HEMS
   -

   IPv6 design choices

Responses from contributors like Brian Carpenter and Dave Crocker emphasize:

   -

   Decisions were largely driven by *rough consensus and practical
   deployment experience*
   -

   External political pressure existed but was not dominant

The exchange highlights enduring tensions between:

   -

   Elegance vs deployability
   -

   Theory vs practice
   -

   Central direction vs community consensus

Key Themes Emerging

Across the thread, several major themes recur:

   -

   *Abstraction shift*: Users relate to applications, not infrastructure
   -

   *Fragmentation risk*: Political, technical, and economic forces may
   splinter the Internet
   -

   *Governance tradeoffs*: Speed vs inclusiveness in standards development
   -

   *Centralization concerns*: Especially linked to the Web and platform
   dominance
   -

   *Historical openness*: A key factor in the Internet’s success
   -

   *Education gap*: Growing disconnect between users and underlying systems

Overall, the discussion reflects a community grappling with how a system
originally built as an open, experimental, and global network is evolving
into a *complex, contested, and partially fragmented digital ecosystem*.

-- 
--------------------------------------
Joly MacFie  +12185659365
--------------------------------------
-