SV: [anti-abuse-wg] DRAFT: RIPE proposal - implementation of an

Hi,

> I think the following should be the "easiest" way of reducing spam:
> 
> 1. Adding that the ISP should have a responsibily to reduce spam to the contract between the ISP and RIPE. (If as only adding "ISP should try to reduce spam" with no other details.)

Yes, but this has to be formulated a way, that the recipient decides
what spam is, so that there has no definition of spam to be included.

So: RIPE has to add to the members contract, that the member is responsible
for receiving abuse reports and reducing the cause of these reports as much as possible.

> 2. Measure how much spam originates from each ISP.

Difficult to formulate this way.

Better would be: RIPE has to start to messure the amount of abuse reports any member receives.
Messurement systems have to include a system run by RIPE NCC and could include
other reliable data from souces not related to RIPE.

> 3. Give RIPE possibility to fine the misbehaving ISP or to cancel the agreement in worst case scenario.

A big yes, but we will have to find consensus here, whats worse
and what punishment could happen, there are a lot of possibilities.

I would prefer to make this step later, after contracts
are changed and a messurement system is introduced.
Having facts, data and results from such a system could
help defining who is  really "bad".

> I saw you had the same or similar idea on http://www.ripe.net/ripe/maillists/archives/anti-spam-wg/2008/msg00056.html
> 
> >From the looks of it im suspecting we cant even get consensus on the first step, unfortunately. :(
> For step number two i think the easiest way would be using spamcop or some similar system. The benefits for using spamcop is that its already an existing system. No cost for developing the system or for teaching users how to report spam. People all around the world already uses it and knows how to report spam. Spamcop has already a system for contacting the correct ISP (I think). RIPE can use the statistics provided by spamcop to see what ISPs misbehaves. (Of course RIPE must be more accepting for the amount of spam than Spamcop. Its already prepared for all other RIRs if they should follow after RIPE has begun. And if RIPE cooperates with spamcop i also assume more users would be reporting the spam there. 

You see, its not that old :o)

> Cons. I dont know if the spamreports on spamcop are reliable or if they can be forged by spammers. I guess their shoud be some way of fixing that problem if that is the case.

The system is well known, and most big provider use it and forget problems
to their customers, the are even opening tickets for these cases, to
not forget them.
spamhaus is another very reliable source.
I bet both and even others would love to share their data with a RIR.

But again, to be really reliable, RIPE needs to have a own system, no judgement
could be done, if RIPE ONLY counts on other resources ...

> Do you think this could work? I will continue to read the archive to see if someone has found a problem with this earlier.

Well, I started this discussion, so I think it would work and would help.
Its only how to formulate this and how details should look like.

There I do still need more input ...


Kidn regards, Frank
--
PHADE Software - PowerWeb                       http://www.powerweb.de
Inh. Dipl.-Inform. Frank Gadegast             mailto:frank at powerweb.de
Schinkelstrasse 17                                fon: +49 33200 52920
14558 Nuthetal OT Rehbruecke, Germany             fax: +49 33200 52921
======================================================================
Public PGP Key available for frank at powerweb.de

> 
> Best Regards
> Martin Tranefjord
>