[ih] Internet without entrenched factions?

[Brian Carpenter]

SHIM6 did that, Happy Eyeballs does some of it. It is the real reason for
having a session layer.

(via tiny screen & keyboard)
Regards,
        Brian Carpenter

On Sat, 16 May 2026, 12:44 Dave Crocker via Internet-history, <
internet-history at elists.isoc.org> wrote:

> On 5/15/2026 1:19 PM, Noel Chiappa via Internet-history wrote:
> >      > From: Brian E Carpenter
> >      > Well, there's one area that stills needs to be fixed ... We don't
> know
> >      > how to support site multihoming in a competitive environment for
> tens
> >      > of millions of small and medium enterprises.
> >
> > ...
> >
> > Basically, TANSTAAFL. Site multi-homing gives benefits; benefits usually
> > aren't free; somebody has to pay.
>
>
> Since I found myself with an operational activity from my home and my
> ISP went out for some day and my cell-phone fallback was not up to the
> task, I bought an OTC router that permits multi-homing and got an
> account with a second ISP.  (And per my insight of some decades ago, I
> have both be active, rather than have the second be on hot standby and
> hopefully start performing when it is needed.)
>
> The multi-homed router does NAT, so my internal hosts don't see any
> functional -- or even addressing -- difference depending on which ISP is
> used for a connection.
>
> This does provide simple, cheap, useful operational redundancy.
>
> What it does not do is maintain a connection across an ISP failure.
>
> Back when this topic was active 25-30 years ago, I suggested putting a
> shim between the app and TCP, providing an appearance of a TCP
> connection to the app, but mapping it down to multiple, alternative
> connections over different ISPs. This could then be robust and maintain
> a TCP connection across ISP failures.
>
> When I was researching this, I discovered that the design of TLS is
> actually of an extensible session layer, which probably would make
> possible to do a design that could be deployed as just an upgrade to the
> TLS software.
>
> Never got any interest in the approach.
>
> And from what I recall, neither did any other approach.  This suggests
> that the Internet is simply much too reliable to motivate people for an
> additional robustness layer...
>
> d/
>
> --
> Dave Crocker
>
> dhc at dcrocker.net
> bluesky: @dcrocker.bsky.social
> mast: @dcrocker at mastodon.social
> +1.408.329.0791
>
> Volunteer, Silicon Valley Chapter
> Northern California Coastal Region
> Information & Planning Coordinator
> American Red Cross
> dave.crocker2 at redcross.org
>
> --
> Internet-history mailing list
> Internet-history at elists.isoc.org
> https://elists.isoc.org/mailman/listinfo/internet-history
> -
> Unsubscribe:
> https://app.smartsheet.com/b/form/9b6ef0621638436ab0a9b23cb0668b0b?The%20list%20to%20be%20unsubscribed%20from=Internet-history
>