Discussion, yes, as a critera for what new work gets chartered (and
therefore whether it has a potential for actual deployment), no.
Typically the discussion goes something like this. At the BOF, Joe Advocate
says that XYZ SDO/industry forum which controls deployment (substitute some
SDO/industry forum like DSL Forum, 3GPP, 3GPP2, etc.) is interested in the
new protocol. When you talk to people in XYZ, however, you find out that
opinion is divided. Nevertheless, the Working Group is chartered anyway,
since the only real criteria for chartering a working group in IETF is that
a "community of interest" exist and that the IESG and IAB are happy with the
technical soundness of the idea. 4 years later, after much good but
essentially useless work, the protocol is complete but nobody wants to
deploy it because XYZ SDO has decided to do something different, usually
modify an existing protocol, since they do not want to deploy a new protocol
due to economic and business considerations.
I believe IETF should not charter work on new protocols until either a)
there is a liason letter from an SDO that can control deployment asking
specifically for the new protocol, or b) there is an industry forum with an
implementation asking for standardization, indicating a committement of
resources towards implementation and deployment.
In contrast maintenace work on old protocols, including cheap hacks such as
the current topic of discussion, should not require any such strong proof of
support, since the amount of work to do them is typically small and
therefore if they fail to get deployed, the resources wasted are less.
It's time for the IETF to admit that the Internet protocol suite is done and
to go into maintenance mode, leaving new developments to the research
community where new and really innovative ideas are much more likely to get
traction.
jak
----- Original Message -----
From: "Jari Arkko" <[EMAIL PROTECTED]>
To: "James Kempf" <[EMAIL PROTECTED]>
Cc: <[EMAIL PROTECTED]>; <[EMAIL PROTECTED]>; <[EMAIL PROTECTED]>
Sent: Saturday, October 13, 2007 12:40 PM
Subject: Business considerations (Was: Re: [Int-area] DCHP-based
authentication for DSL?)
James,
I wanted to respond to this:
Economics (==business considerations) *never* come into consideration
when IETF charters new work in contrast to other sdo's.
This is incorrect. It is true that we are not discussing business
models, prices, etc. However, it is equally true that interest
from major users and vendors is an important input when
chartering new work. Not the only input or an overriding
concern, but an an important input nevertheless.
Jari
_______________________________________________
Int-area mailing list
[email protected]
https://www1.ietf.org/mailman/listinfo/int-area
