Re: [arch-d] Layer violations considered harmful?

["Bless, Roland (TM)" <[e-mail]>]

Hi Brian,

Am 06.06.19 um 01:37 schrieb Brian E Carpenter:
> For something I'm currently working on, I was looking for a citation for "layer violations considered harmful" or something similar.
> 
> A little to my surprise, I didn't find it anywhere.
> 
> Zimmerman's 1978 paper on the OSI model describes the layered model, but doesn't discuss layer violation as far as I can see.
> 
> Dave Clark's 1988 "The design philosophy of the DARPA internet protocols" doesn't either.
> 
> RFC1958, which I edited, doesn't either, and that was an explicit choice. Yakov Rekhter wrote to me in 1996: 'In other words, "layer violation" is theory, but when making pragmatic decisions we coul violate the theory...' so we left it out.
> 
> So did anyone ever analyse and justify this shibboleth?

RFC 3439 https://tools.ietf.org/html/rfc3439
contains a discussion on Layering, but rather in the
purport of "Layering Considered Harmful"
https://tools.ietf.org/html/rfc3439#section-3
discussing the problem of redundant and repeated
functionality in the layers as well as increased
complexity by increased layering.

However, Section 3.1 "Optimization Considered Harmful"
mentions that optimization across layers may cause
also problems (increased complexity, tighter coupling,
lower reliability).

RFC 1122 also discusses "strict layering" a bit:

1.3.1  Organization

         Protocol layering, which is generally used as an organizing
         principle in implementing network software, has also been used
         to organize this document.  In describing the rules, we assume
         that an implementation does strictly mirror the layering of the
         protocols.  Thus, the following three major sections specify
         the requirements for the link layer, the internet layer, and
         the transport layer, respectively.  A companion RFC [INTRO:1]
         covers application level software.  This layerist organization
         was chosen for simplicity and clarity.

         However, strict layering is an imperfect model, both for the
         protocol suite and for recommended implementation approaches.
         Protocols in different layers interact in complex and sometimes
         subtle ways, and particular functions often involve multiple
         layers.  There are many design choices in an implementation,
         many of which involve creative "breaking" of strict layering.
         Every implementor is urged to read references [INTRO:7] and
         [INTRO:8].

Even RFC 817 https://tools.ietf.org/html/rfc817 mentions:

The way to achieve this goal of packet sharing is to break down the
barrier between the layers of the protocols, in a  very  restrained  and
careful  manner, so that a limited amount of information can leak across
the barrier to enable one layer to optimize its behavior with respect to
the desires of the layers above and below it.

So this is probably a starting point.

Regards
 Roland