On 9 Sep 2022, at 4:15, Phillip Hallam-Baker wrote:
On Thu, Sep 8, 2022 at 8:31 PM S Moonesamy <sm+ietf@xxxxxxxxxxxx>
wrote:
Hi Phillip,
At 08:53 AM 08-09-2022, Phillip Hallam-Baker wrote:
There are many mistakes made in standards work but one of the most
common is to keep flogging a dead horse. While there is certainly a
possibility that XMPP will somehow manage to Travolta and sweep away
the proprietary messaging systems we are stuck with, that is far
from being a likely outcome.
The IETF recognizing that Jabber has failed to succeed is a positive
step in my view because it clears the way for an approach that has a
better chance of success.
My goal here is to establish an open infrastructure for messaging
that has end-to-end security built in. If I thought Jabber was a
viable vehicle for achieving that, I would have designed my system
around Jabber. As things stand, there is no messaging solution that
is designed as an open infrastructure. Signal has an open standard
but it is not an open service, it is a walled garden.
I read some of your work on mesh. As an comment unrelated to mesh, I
would say that it would be very difficult to compete with the
established messaging services. I doubt that the users [1] would
migrate to a service based on open standards given that their
existing service works for them.
Just throwing another option on the table isn't going to be enough.
This is
not my first rodeo though.
The key control structure here is the contact book. The key pain
points
that are going to drive convergence towards a single open
infrastructure
are user dissatisfaction with having to juggle a dozen different
messaging
apps and the big hammer of government regulation.
Our Services Our Choice — The Mathematical Mesh (mathmesh.com)
<https://www.mathmesh.com/ourservicesourchoice>
The current situation is unsustainable. The problem with walled
gardens is
that they don't work at all or the work so well, they end up as
monopolies.
And if there is one big winner, that means there will be at least
three big
losers. And given the regulatory situation, this is really not a game
anyone should ever hope to win. Not unless they think they can buy off
the
US Congress and the EU Commission at the same time.
The market as I see it today is twenty providers of which at least
half
know they have no chance of winning but they really want to make sure
nobody else can either. So that half is already a constituency that
should
have a serious interest in an open infrastructure.
Adversarial interop is a real thing.
Do we have any data to back up the claim that XMPP is no longer in
active use on the Internet?
I see people on this thread relating their personal experiences, but
does anyone have any actual clue about XMPP use on the Internet?
I found this from 2020.
<https://blog.prosody.im/2020-retrospective/>
A bit dated, but 52000 deployments of a single XMPP server
implementation is nothing to sneeze at.
Anecdotes from a few IETF participants are largely worthless for
determining the usage of a protocol. If we want to make the claim that
XMPP is no longer used some actual data is necessary.
—Andrew
