Hi Luigi,

Thanks for the paper. It's an interesting approach, although
I don't know how robust the conclusions are given the number
of assumptions and the potential for non-linear behaviour.

One point I want to dispute is in the very first sentence:

"...with the most pressing issue being the super linear growth
of the BGP routing table [3]."

Superlinear with respect to what? The growth may be faster than
linear with respect to time, but time is not a scale metric. As
I've shown [URL below], BGP growth was historically significantly
*sublinear* with respect to the total number of Internet subscribers,
which means that it is also sublinear with respect to total ISP
revenue. That means that BGP-related OPEX has automatically declined
as a fraction of revenue as the subscriber base increased. It may
be that the effects you describe are second-order compared to this,
and therefore not very powerful as (dis)incentives.

  Brian

[See Observed Relationships between Size Measures of the Internet,
ACM SIGCOMM CCR, 39(2) (April 2009) 6-12,
http://www.cs.auckland.ac.nz/~brian/sqlaw.pdf ]




Regards
   Brian Carpenter




On 2010-03-12 23:52, Luigi Iannone wrote:
> Hi All,
> 
> Tapio Leva and myself   worked on some economic aspects of Loc/ID Split. 
> The analysis is pretty simple but has some merits. Attached is the paper. 
> 
> Hope that this work is of some interest for the list. Comments are very 
> welcome.
> 
> Ciao
> 
> Luigi
> 
_______________________________________________
rrg mailing list
[email protected]
http://www.irtf.org/mailman/listinfo/rrg

Reply via email to