[145120] in North American Network Operators' Group
Re: "general badness" AS-based reputation system
daemon@ATHENA.MIT.EDU (Serranos)
Wed Sep 28 05:33:17 2011
From: Serranos <serrano.miser@gmail.com>
In-Reply-To: <0EA58FFF-2FBE-4977-BC62-211842FEB52E@merit.edu>
Date: Wed, 28 Sep 2011 10:32:56 +0100
To: Manish Karir <mkarir@merit.edu>
Cc: nanog@nanog.org
Errors-To: nanog-bounces+nanog.discuss=bloom-picayune.mit.edu@nanog.org
On Sep 26, 2011, at 02:23 , Manish Karir wrote:
> We tried to outline some of the challenges of building such a system =
in our NANOG52 presentation:
>=20
> =
http://www.merit.edu/networkresearch/papers/pdf/2011/NANOG52_reputation-na=
nog.pdf
>=20
> In particular see slide 4. where we tried to lay down what we think =
the requirements are for a socially acceptable
> reputation system. =20
>=20
> With a bit of luck we might be able to announce the release of our =
system before the next NANOG mtg, but in=20
> my opinion collating host reputation reports is just a small and the =
easiest part of the effort. The key is in=20
> solving the challenges of allowing (and incentivizing) participation =
and being robust to false information
> injection.
Hi Manish.
As mentioned by Gadi, the maintenance of such tools is not often easy, =
in particular since some datasources may disappear or become obsolete =
over time. For example, to have a global view of the BGP landscape the =
best service I know is RIS from RIPE, but there aren't many =
alternatives. Although this problem may be reduced through an increase =
of the total number of datasources, it is something to be considered. =
Also, since historical data is considered, the fact that some =
datasources may disappear over time can affect the ranking value.
Most importantly, this type of approach is dependent on the level of =
commitment the network community has, which may be mined by not enough =
incentives (the problem mentioned in slide 3). Namely (as stated before) =
the problem of certain customers not being able to reach critical =
systems "just" because that ASN was considered evil, is a strong =
incentive *not* to adhere to the system. This is IMHO THE biggest =
Problem. Also, if you are a transit AS do you think this to be a viable =
approach?
Although I think this philosophy has strong arguments to move forward, =
it also has many challenges that must be dealt with and the biggest ones =
are not technical (what a surprise=85).
Thanks for your valuable contribution.
Regards,
S.
