On Sat, 2021-12-18 at 15:13 +0100, Alexey Shpakovsky via mailop wrote:
> On Sat, December 18, 2021 13:50, yuv via mailop wrote:
> > What makes the difference between [the smoothly running messaging
> > systems] and internet email?
> 
> I believe answer is centralization and to some extent lack of
> backwards compatibility requirement.

what is it that centralization brings to those systems?  after all,
they also consists of numerous independent parties communicating with
one another over electronic devices, exactly like internet email.


> When you're one company controlling both backend and all frontend

This is the undesirable feature of centralization, I think we can all
agree on that.  But what are the desirable features of centralization,
and can they be breaken out and applied to a decentralized world?

Let's start:

Front end:
* undesirable: dictated device, sometimes only available from selected
vendors on selected platform (the walled garden, though for
Apple/iMessage it has apparently already achieved sufficient network
effect for the company to start allowing its captives to invite non-
Apple devices to the conversation)
* desirable: implement ONE SET OF RIGID (but can change over time),
INTEROPERABLE, SIMPLE, SPECIFICATIONS.

Back end:
* undesirable: single controlling entity
* desirable: ONE SET OF RIGID (but can change over time) INTEROPERABLE,
SIMPLE, SPECIFICATIONS.

Ecosystem:
* desirable: CLOSED TO ABUSE SOURCES (and no one should be big enough
to tolerate abuse from their network, including Google, Microsoft,
Amazon, and any other cloud provider) 
* undesirable: CLOSED TO ALTERNATIVE Front and Back ends. 
Uncompetitive
* desirable: OPEN TO ALTERNATIVE as long as they implement the exact
rigid specification. (Competitive, without allowing for uncontrolled
feature creep, or in other words no Embrace, Extend, Extinguish)
* desirable: CLOSED PROTOCOL SPECIFICATION, subject to strict change
control 

Care to continue?


> And who on this list haven't thought: "oh, I wish this part of email
> spec was different"?!


The thinking is legit.  It is the process of addressing that thinking
that has failed.  Too often, change is adopted for the sake of change. 
See the recently proposal for DKIM-QR on this list.  Just because QR
codes have been made popular by vaccine passport, it does not mean that
we have to rush to implement them everywhere and for everything.  Same
with blockchain previously.  But fashion is irresistible and when I
have a hammer, every problem looks like a nail.


> Email openness is both blessing (when any person can implement an
> email client however they like) and a curse (when any spammer can
> implement an email client...).

Because the protocol's implementation is not the appropriate place to
deal with spammer, and this is one of the major driving forces that has
pushed internet email to the point of breakage.  The protocol should be
open to read, but under strict change control.  Implementation of the
protocol should be open to anyone.  Bad actors should be kept out not
by preventing them to implement the protocol, but by preventing them
from joining the network.  The network must be open only to actors that
adhere strictily to the rules.  This includes operators AND end-users.


> Worth mention however, that I've seen spam on other messaging
> platforms, too, and a black market for Telegram accounts being
> mentioned, and people developing anti-spam solutions for not-so-big
> public Telegram groups...

Of course, if one thing is not changing, it is human nature, and any
communication platform large enough will attract scumbags.  Keeping
scumbugs at bay must be central to the protocol design (or evolution). 
SPF/DKIM/DMARC have all failed at that.  They have made life more
difficult for the legitimate user and they are better mastered by the
spammers than by the legitimate users.  The platform has been overtaken
by marketers.  I am disgusted when I see a recommendation (by
Microsoft) to craft emails in HTML format to make them more
deliverable.  To me, HTML is a hallmark of marketing email that I do
not care a damn about, and the messengers protocols are much better at
that.  To me, internet email should not try to become like the
messenger protocols: it is a losing competition, partly because of the
legacy baggage.  To me, internet email should refocus on what it was
meant to be in its original design: a way to send/receive PASSIVE
CONTENT, without the tracking/spying bugs and other active/dynamic
elements that advertisers so covet and spammers abuse.


On Sat, 2021-12-18 at 15:42 +0100, Jaroslaw Rafa via mailop wrote:
> Dnia 18.12.2021 o godz. 07:50:19 yuv via mailop pisze:
> > 
> Sadly, in recent years - mostly because of the actions of the "big
> guys", I will hold on to my opinion on that - email has steered
> towards becoming a similar group of independent, non-interoperable
> services

It may be the "big guys" that have poured energy into Embrace, Extend,
Extinguish.  However, the way I see it is that internet email is
suffering death by committee.  It is in the RFCs, and the long list of
names contributing to them.  The "big guys" are part of it, but they
are small in number compared to everyone else.


> As I understand this, we here on this list are working to keep up the
> interoperability of email, but this becomes more and more difficult

I wish your understanding is correct.  I am cynical enough to suspect
that many here on this list are working for their own
interest/agenda.  Spammers are the ones that implement best the
alphabet soup of DKIM/DMARC/whatever else.  The "big guys" that are
such a lightning rod for you are here for their business, not for
interoperability, and their anti-competitive practices are obvious to
many, just not enough present to legislators and competition /
antitrust authorities in the relevant countries.


> It will probably sooner or later end so that you
> would have to be on Gmail to communicate with Gmail users and the
> same for other services.

Yes, this is the trajectory, sadly.  And as one wise man (whose
political ideas I do not share) once said (and I herewith adapt to the
circumstances): if I want to talk to you, I will speak your
language/protocol ABER wenn Sie mit mir sprechen wollen, muessen Sie
Deutsch/my protocol sprechen.

--
Yuval Levy, JD, MBA, CFA
Ontario-licensed lawyer


_______________________________________________
mailop mailing list
mailop@mailop.org
https://list.mailop.org/listinfo/mailop

Reply via email to