Re: [Idr] draft-scholl-idr-advisory

[Dorian Kim <dorian at ntt.net>]

On Thu, Jul 02, 2009 at 11:20:41AM -0400, Christopher Morrow wrote:
> I share the 'overloading bgp yet again is a bad plan' but also

This ship has sailed long ago unfortunately, and at this point
I think the "overloading bgp yet again is a bad plan" cry is reaching
the point of "save the children". 

> 'getting ops to run yet another thing on their hardware and arranging
> that to work properly amongst their peers' is not trivial. (so this
> goes into bgp, please keep it very simple, small payload + limited
> ascii-set)

Given the actual practices of operators today, and the realities of poor 
coordination in most cases, this proposal will provide a net benefit
to day to day operations of the Internet.

Whether the benefit is worth the cost in abstract is unclear to me, but it 
certainly is far more useful than any number of other things people have 
shoehorned into BGP of late. 

I support the adoption of this as a WG item. 

-dorian

> >> However, and I'm sure we will get into this as the WG proceeds, I think
> >> there are fundamental elements of this which are bad.  Notably, I don't
> >> think that bloating BGP with operator-to-operator messages is a good
> >> move to maintain a stable, secure (fwiw) protocol.
> >>
> >> I agree that the problem of op-2-op communication needs more attention,
> >> and I hope with this starting point the WG can forge a reasonable
> >> solution that doesn't increase the fragility and risk of operating an
> >> Internetwork.