Re: RE: AD request / L2 Triggers Charter Statement

> From: James Carlson <[email protected]>

> Sure.  But it's still a system design issue, and not a TCP/IP protocol
> issue.  The former is certainly a fascinating and useful area, and
> obviously very important for many (or most) TCP/IP implementors, but
> it's just not the focus of the IETF.  IETF documents tend to tell
> people what the system must do in order to interoperate, and quite
> intentionally leave out *how* it must be done.
>
> A revised version of DLPIv2 that covers these signaling mechanisms in
> detail would probably make a number of people happy.

I bet that besides being unfamiliar with the unhappy history of WAP,
the WAP Forum, and whatever the WAP Forum recently changed into (unhappy
at least from the prospective of people who know that TCP worked always
fine on 9600 bit/sec modems), some people are not aware that "DLPI"
stands for "Data Link Provider Interface."  DLPI already includes
"packets or meta packets" sent among L2 and L3 modules.

See http://www.google.com/search?q=dlpi

There have been repeated demands in this thread that the IETF "work
with" other standards bodies.  The first rule for any bureaucratic
outfit that hopes to work with others is not stomp on other outfits'
turf (at least not accidentally).  For the IETF to write standards in
the Open Group's territory would be seen as an act of war by the Open
Group.  The IETF could say "these are the vague sorts of things that
something like DLPI might communicate," but the actual "meta packets"
must be defined by the committees that own that turf.

And no, an individual's difficulties getting an idea heard by the Open
Group would not be a reason for the IETF to go to war with the Open
Group....or the ITU, IEEE, ANSI, or any other standards group.


>              ...  (If the idea can't survive outside of the standards
> track, then I'm afraid I must not completely understand the issue.)

No, that an idea cannot survive outside of the standards track only
shows that the idea is seen as wrong or useless by the people who
matter.  The vast majority of the people who matter do not participate
in the IETF, but are the people who read (or don't read) RFCs and
make (non-)compliant products.  The standards track is not about
laws or compelling programmers and hardware designers to write C.
The IETF is powerless to order anyone to do anything and has little
power to persuade.  If an idea is not innately convincing, it's
dead on delivery in an RFC.

The difference between Informational and Standards Track RFCs is
that the former document ideas while the latter document the
consensus of the IETF about an idea the consensus says is good and
also the business of the IETF.  There could be an RFC about world
peace and hunger, but it should not be on the standards track.


Vernon Schryver    [email protected]

_______________________________________________
pilc mailing list
[email protected]
https://www1.ietf.org/mailman/listinfo/pilc
http://www.ietf.org/html.charters/pilc-charter.html
http://pilc.grc.nasa.gov/