Re: Single vs many solution(s)

Summarizing a bit:

Monday, May 12, 2003, 8:36:43 AM, Yakov Rekhter wrote:
> To avoid disappointing you the WG immediately needs to start the
> discussion on whether to use L2TPv3 encapsulation or MPLS or MPLS
> over GRE, or MPLS over IPSec.

> Likewise the WG immediately needs to start the discussion on L2TPv3
> vs LDP based signaling.

> Likewise, the WG immediately needs to start the discussion on BGP
> vs Radius based auto-discovery.

If the goal of the WG is to produce the IETF Standards for L2 VPNs, I
think the WG should indeed be discussing which mechanisms should be
chosen as mandatory to implement. Not requiring a mandatory one will
affect interoperability, and more than one mandatory mechanism per
function are just not needed--one is enough. If folks believe other
(optional to implement) mechanisms would really add value, I don't
think anyone would object. The key point here is if we want
implementations of the IETF standard to interoperate, we need to
define the minimal function set to be supported.

Monday, May 12, 2003, 10:08:22 AM, Yakov Rekhter wrote:
> Agreed. In fact, I don't quite understand why some folks on this
> list insist that the IETF should be in the business of making
> business decisions...

I don't think anyone is trying to insist on that, quite on the
contrary. I think there's a general agreement that the IETF should NOT
be making business decisions, and the disagreement is in which
decisions are business and which are technical.

If we come here believing that the IETF is nothing but a market
battlefield, then we can call any situation where more than one
existing implementation is available a "business issue", and agree
that trying to resolve it would cost too much blood, so "the market"
should pick the one that they like more.

If instead we believe that the IETF is an engineering organization,
where we come to work together on something we can later use to
interoperate in SP networks (even though we may have our own existing
implementations), I think we should be ready to sit down, talk to each
other and try to build consensus on what should that IETF Standard be.

Which mindset do we choose?

Alex