Hi Christopher, Looking at the current (and past) charging fees, only 1% of all registered LIRs are category extra large, and they already pay 'peanuts' compared to their size: 5750 euro, not even the double of a medium size category LIR (2750 euro). I doubt those extra large companies would pay considerably less with this proposed scheme, as most of them have numerous registered objects. Just run an inverse lookup at the RIPE database for some of maintainer objects of the extra large category members to see big amounts of inetnum objects as result. https://www.ripe.net/ripe/docs/ripe-499 Some arguments in favour for this new scheme would be: - Ease of administration (and less costs for NCC); - Fair and transparent for every LIR, without complicated formula like the current scheme; - Future proof, not based on ipv4 address count; Please take a look at the current numbers or wait for the proper calculations before jumping to conclusions about this proposal. With kind regards, Michiel Klaver At Wed, 5 Oct 2011 15:28:01 +0200, "Christopher Kunz (Filoo GmbH)" <chris@filoo.de> wrote: Err... Great idea... or is it? Really, cui bono? This scheme has been proposed numerous times, IIRC even in this very discussion. In my perception, Small LIRs will not really profit from it, as they typically have a small network (1 or 2 allocations, 1 ASN, X-SMALL category). They'll pay roughly the same because the annual membership fee will have to be adjusted to account for the losses accrued by your idea. Medium LIRs might pay slightly less, I guess... There's probably a sweet spot somewhere (and I'll assume you calculated it so you're in it ;) ). Big LIRs however, with their multiple 12's, large AS sets and network allocations would probably pay drastically less than before. This would shift the weight off those who massively profit from the resources administrated by RIPE NCC onto those who take up less resources. I don't consider this to be very fair. On the contrary: Those who use up large portions of the address space should damn well pay large portions of the NCC's bills. In addition, charging per resource allocation, be it one address or 65K, looks like "per address" charging to the tax authorities. And this is specifically what RIPE is looking to avoid. Gruß, --ck PS: To the colleagues at Telekom and PrivateLayer: GET YOUR DAMN TICKET SYSTEMS OFF THIS LIST!