[Info-vax] Rethinking DECNET ?

[Bill Gunshannon]

In article <e648ff3f-2092-4b81-bbe1-4c72d112c8e0 at googlegroups.com>,
	johnwallace4 at yahoo.co.uk writes:
> On Sunday, 31 August 2014 13:52:10 UTC+1, Simon Clubley  wrote:
>> On 2014-08-31, VAXman-  @SendSpamHere.ORG <VAXman- at SendSpamHere.ORG> wrote:
>> 
>> > In article <ltscu4$1le$2 at dont-email.me>, Simon Clubley <clubley at remove_me.eisner.decus.org-Earth.UFP> writes:
>> 
>> >>On 2014-08-30, JF Mezei <jfmezei.spamnot at vaxination.ca> wrote:
>> 
>> >>> DECNET offers some neat stuff and security.
>> 
>> >>>
>> 
>> >>
>> 
>> >>DECnet most certainly does _not_ offer security - it's an unencrypted
>> 
>> >>data stream.
>> 
>> >
>> 
>> > Just like TCP/IP.
>> 
>> >
>> 
>> 
>> 
>> DECnet is a whole range of protocols just like TCP/IP is and while
>> 
>> the basic TCP and IP layers do not have encryption built in, there
>> 
>> are other layers, within the protocol stack, which do give you that
>> 
>> capability. You can even choose per-site or per-session encryption
>> 
>> as you wish.
>> 
>> 
>> 
>> DECnet has no such equivalent to those protocol layers.
>> 
> 
> DEC bet the farm on OSI. At the time it looked (imho) like a good bet,

One of many mistakes!!

> with loads of corporate backing in the late 1980s and beyond, when
> IPv4, POP/SMTP, DNS, etc, was already looking rather dated in a number
> of respects but clearly wasn't going to go away overnight, hence DEC's
> considered integration of DECnet Phase IV, OSI, and IP networks.
> 
> Then a few years later there was the emergence of the business and
> consumer Internet, with lots of people wanting connectivity and none
> really wanting to pay much for it, or realising the implications of
> doing it badly/on the cheap. What's the answer back then, from an ISP
> point of view? RFCs, zero-cost software (largely starting with zero
> trust too) 

There was a lot of trust.  Some would say way too much.  Systems without
passwords.  The BSD R-commands.  That was a social consideration, not a
technical one.

>            and hence we now have an IP-centric world full of spam
> and other such delights.

IP has nothing to do with SPAM otehr than the inertia one would have
to buck to change it.  I have said before that there is an even older
protocol for doing email that could virtually elimiunate SPAM and it
is 100% compatablie with the current networking technology.  Not only
that, but if you have seen the recent comments from (I think) DEFCON
about the need to encrypt meta-data the protocol I am thinking of is
already postioned to do that leaving only two pieces exposed during
transit source host and destination host, the two pieces absolutely
needed to move the email.  Everything else can be encrypted including
From:, To: and Subject:.

But only salmon succeed in swimming upstream against that strong a current.

> 
> If OSI had caught on properly, we'd have (for example) rather less spam,

Exactly how would that be?  It isn't the protocol that is the propblem, it
is the application.  DECNET has them, too.

> fewer opportunities for phishing, genuine interoperable multimedia email,
> and so on. X.400 email etc would have sorted those, from the ground up
> rather than by means of near-universal but barely-functional elastoplasts
> and band-aids. With equivalents in other areas as needed (e.g. FTAM
> rather than ftp).
> 
> Didn't happen, pointless worrying about it now.
> 
> Meanwhile, IPv6 for Joe Public is still a few years away, just like it
> was a decade and a half ago when I did my first IPv6 course.  And it
> has the same issues to address as OSI had to address maybe three
> decades ago.

And, for the same reason.  inertia.

> 
> "Microsoft: Bringing you 1980s technology to a 21st century world"
> 
> Not just Microsoft - but they're a big part of it. Can't exclusively
> blame them for the near-universal dominance of POP/SMTP though; you
> don't see many Microsoft-based ISPs.

bill

-- 
Bill Gunshannon          |  de-moc-ra-cy (di mok' ra see) n.  Three wolves
billg999 at cs.scranton.edu |  and a sheep voting on what's for dinner.
University of Scranton   |
Scranton, Pennsylvania   |         #include <std.disclaimer.h>