RE: [Asrg] seeking comments on new RMX article

["Eric D. Williams" <[e-mail]>]

On Monday, May 05, 2003 1:30 PM, Vernon Schryver 
[SMTP:[e-mail]] wrote:
> > From: "Eric D. Williams" <[e-mail]>
>
> > Is RMX a proposed change to existing Internet services OR, 1) a proposed
> > extension to/of existing services? 2) an additional use of existing
> > services?
> > 3) a new protocol?
> >
> > My read is that it is #2 above.  The RMX proposal do not seem to constitute
> > a
> > change to existing services but rather a leveraging of those services to
> > achieve a specific goal.  It also seems to be application specific as
> > opposed
> > to a networking protocol.
>
> Yes, but what is your point?
>   - Have you noticed much use of DNSSEC?
>   - How about some of the major improvements in the not so recent
>      update to the HTTP protocol?
>   - SMTP-TLS has been available for years, but this mailing list is not
>      using it.  Do the SMTP servers or clients near you use it?
>
> Then there are RIPv2, IPv6, TCP-SACK, and TCP-ENS.  Even TCP
> slow-start took most of 10 years, and that was in the old days when
> the net was tiny and well controlled.

No internet protocol started with ubiquity AFAIK, some have not been adopted 
after being 'standardized' some evolved into de-facto standards before IETF/ISO 
(insert acronym) work was completed.  I do not think that these are valid 
measures of the viability of a proposal.  Also successful adoption IMHO is 
based on whether something 'gives the people what they want' and not 
necessarily 'what they need' or 'what they asked for'.

The paragraph you cite was inserted based upon Dave's post concerning changes 
to Internet services.  My point is I don't see the RMX proposal as a change to 
Internet services.  DNSSEC and SMTP-TLS are along the same lines.  Recall my 
opinion is that it may be 'what people need' or 'what they asked for' but is it 
'what they want'?  I think implementation difficulty also plays into the mix 
(on that I think we all agree).

> As the saying goes, "In theory there is no difference between theory
> and practice, but in practice there is."
>
> Would those who assure us that deploying RMX or any anti-spam scheme
> could be done in fewer than 10 years tell us about their credentials
> for making their authoritative sounding statements?  What protocol
> have they been involved in that was deployed as fast as their spam
> system would be?

I don't think any of us could boldly make such a claim as an individual, but 
surely that is not a criteria for evaluation of a proposal.  Achieving 
convergence and interoperation of a specification can be onerous (LDAP) or 
trivial (RFC822).

-e
_______________________________________________
Asrg mailing list
[e-mail]
https://www1.ietf.org/mailman/listinfo/asrg