Re: Revised 2007-01 set back to Review Phase (Direct Internet Resource Assignments to End Users from the RIPE NCC)
Hi, On Thu, 04 Sep 2008 09:43:03 +0100, Gert Doering gert@localhost wrote:
Actually I'd like to hear a few more voices regarding this latest version of 2007-01 (v4). We have deliberatly set a fairly short discussion phase to be able to get it *done*, finally, before Dubai, so please voice your opinions now.
I support the spirit of this proposal, but I have major issues with imple- mentation. Nick does brush this of as "operational", but that is where the rubber meets the road. Please also note that my concerns are really mostly with the future of PIv6 as PIv4 is really a lost cause in my opinion. At the end of the day my argument is really with barriers, such as fees: I'm not opposed to a recurring fee as such, but it needs to be low enough not to be a significant barrier to entry even to small businesses or pro-bono organisations such as OSS projects or charities. Actually, a case can be made for making the barriers a lot lower for PIv6, if only to speed it on its way to ubiquity ;) I've seen some mention of a "routing slot tax", I think this idea needs to be stepped on. Hard. Now. Besides, whom goes that "tax" to? $VENDOR? After all, everyone with PI space (multihoming assumed) has to carry full tables too, no? And those whose *business* it is to provide transit should surely have calcula- ted upgrades into the price of their service? I don't think that RIPE's func- tion is (or should be) to act as a tax collector for LIRs... I'm aware that the setting of fees is a function of the RIPE NCC and not the community at large, but the process of setting those fees is not transparent to me. OTOH, maybe it is better to leave the setting of fees to others than those who would want to tax the end-users for their incredible cheek in trying to break away from SP lock-in. Another possible barrier is justification. As far as v6 goes there should not be any need for that in the contractual process. IMO, "because I want it" should be all the justification needed for at least the initial PIv6 assignment. Having said all that, I think that keeping track of PI space *is* a good idea and bringing end-users into the RIPE community is an even better one. If the above issues can somehow be addressed, I would support 2007-01. Regards, Sascha Luck SLU3-RIPE
Sascha, Your concerns are possibly reasonable (see details inline, after this introduction). But I have two main issues with your opposition: 1. The current system is, quite frankly, shit. It is embarrassing that the RIPE region has continued to put up with such a poorly managed way to track number resources. Any mechanism to track these resources is better than what we have now (which is NOTHING). 2. No policy is going to make everyone happy. We have worked at this for many months now (the proposal is 2007-01, not 2008-01). At some point we have to recognize that what we have is not perfect, but will ever be perfect, so we should adopt it anyway. To quote Voltaire: Le mieux est l'ennemi du bien (the perfect is the enemy of the good). On Wed, 2008-09-17 at 02:46 +0100, Sascha Luck wrote:
I'm not opposed to a recurring fee as such, but it needs to be low enough not to be a significant barrier to entry even to small businesses or pro-bono organisations such as OSS projects or charities. Actually, a case can be made for making the barriers a lot lower for PIv6, if only to speed it on its way to ubiquity ;)
I think the RIPE community sets policies, but details - like specific fees - are implemented by the RIPE NCC. This is a good model and has worked for a long time. I trust the RIPE NCC will set reasonably low fees. If the "barrier" you want to remove for IPv6 is knowing who is actually using the address space, I think that you are misguided. That "barrier" needs to be firmly in place, before we end up with the same mess in IPv6 as we have in IPv4.
I've seen some mention of a "routing slot tax", I think this idea needs to be stepped on. Hard. Now.
Since this is not a part of the proposal justification or details (I think), we should not have to worry about it. I agree it sounds quite awful though. :)
Another possible barrier is justification. As far as v6 goes there should not be any need for that in the contractual process. IMO, "because I want it" should be all the justification needed for at least the initial PIv6 assignment.
Does this policy mention changing justification at all? I don't think it does, but I might have missed it. If you think justification needs to be changed for IPv6 allocation, then a new proposal is a good idea. Start a thread, we can discuss it! -- Shane
On Wed, Sep 17, 2008 at 12:05:50PM +0200, Shane Kerr wrote:
1. The current system is, quite frankly, shit. It is embarrassing that the RIPE region has continued to put up with such a poorly managed way to track number resources. Any mechanism to track these resources is better than what we have now (which is NOTHING).
Oh, I agree absolutely. Not only is it a good idea to know where the space is, it is also a good idea to bring these end-users into the community in some way or the other. This is after all what the founders of the Internet intended :)
2. No policy is going to make everyone happy. We have worked at this for many months now (the proposal is 2007-01, not 2008-01). At some point we have to recognize that what we have is not perfect, but will ever be perfect, so we should adopt it anyway. To quote Voltaire: Le mieux est l'ennemi du bien (the perfect is the enemy of the good).
Sure, but it is like with laws, if you get it wrong, it's next to impossible to right it after the fact.
If the "barrier" you want to remove for IPv6 is knowing who is actually using the address space, I think that you are misguided. That "barrier" needs to be firmly in place, before we end up with the same mess in IPv6 as we have in IPv4.
That's not my intention at all, see above.
I've seen some mention of a "routing slot tax", I think this idea needs to be stepped on. Hard. Now.
Since this is not a part of the proposal justification or details (I think), we should not have to worry about it.
I agree it sounds quite awful though. :)
It's not part of the proposal but some have suggested setting a high financial barrier to discourage use of PI(v6) in order to avoid "pollution" of the routing table. I would not like to see this essentially useful proposal hi- jacked for that agenda.
Does this policy mention changing justification at all? I don't think it does, but I might have missed it.
It doesn't, there isn't any policy for PIv6 yet, I guess this would have to be discussed in the context of this (new) policy. cheers, Sascha
participants (3)
-
Sascha Luck -
Sascha Luck -
Shane Kerr