Re: [members-discuss] Input from Membership on RIPE NCC Charging Scheme Model

On Fri, Sep 16, 2016 at 1:25 PM, Daniel Pearson <daniel@privatesystems.net> wrote:
On 09/16/2016 05:57 AM, Prager-IT e.U. wrote:
The plan I envision is quite simple, move from the current Flatrate charging scheme to a system that charges members an amount for the resources they are using. This will ensure that each and every RIPE NCC member will pay their fair share.
1. Charge per allocated IPv4 address with a linear system.
price_per_ipv4 = required_budget / amount_of_addresses_in_use your_yearly_membership_fee = amount_of_ipv4_you_use * price_per_ipv4
Did you happen to read my reply to the thread. This is going to equate to almost no resources left as a /22 will cost roughly 24 EUR per year.
I read everything said so far, to prevent this from happening we also need to increase the setup fee as I already outlined.
This will also apply to Provider Independent(PI) assignemnts, anycasting assignments and Legacy Internet Resources registered via a sponsoring Local Internet Registry(LIR).
You're going to be hard pressed to charge Legacy resource holders, BUT, even if you did, my above figure of a /22 costing 24EUR would probably be cut down to 16 EUR or less
If the resources are sponsored via a sponsoring Local Internet Registry(LIR) I don't see why we should not be able to charge the Legacy resource holders.
IPv6 PI assignments and IPv6 IXP assigments will remain unchanged at 50,-- Euros per year.
Moot point here since the number of IPV6 resources avail.
The goal is to provide a comprehensive plan for the membership to vote on. The way I see this needs to be in addressed as well in order for that to happen.
If we ever reach a point where IPv4 becomes obsolete revert back to a Flatrate System as due to the nature of IPv6 a Flatrate system is a fair choice.
2. Increase the Setup Fee to an amount that reflects the current reality of the transfer market, 6.000,-- Euros.
Read the 2015 Budget, they collected slightly over 5 Million EUR in setup fee's, which got dispersed back to its members as a credit on next years bill, increasing the setup fee will simply act to lower, even further, the yearly cost. In 2015 Each LIR would have received 'roughly' a 415 EUR credit on their bill... So we might as well just not charge for IPv4 upto a /19 , and if you're going to up the setup fee's then we might as well make that a free /18.
I am well aware of the numbers. Under my plan there most likely would never be any sum significant enough to disperse.
3. Introduce an Inter-RIR Transfer Fee for resources that leave the RIPE Region so people don't flee with their resources into a cheaper Region, 3,-- Euros per IPv4 address.
If you do this, then it's a two way street. Not to mention that in the growing global age, you can't tell me every IP being routed to the ripe region is under RIPEs control. By adding additional hurdles folks will simply ignore, and announce their space globally from the originating RIR more so than they currently do now.
The goal of the change is to prevent people from fleeing with their resources allocated by the RIPE NCC from fleeing into cheaper regions after the new charging scheme has been approved. I am confused as to why you are going on about where resources are being announced from.
4. Buy Back IPv4 resources and return them to the free pool with any excess money from members that are willing to give back resources.
With what money are these bought back and at what price? Only to re-assign at a financial loss and then have to buy it back again? Grand scheme of things, everybody here is ignoring the fact that, with basically 21,000,000 EUR budget to disperse between all LIR's in RIPE you aren't going to change the world. Aside from a lot of keyboard warriors pounding away with random ideas that have no factual backings this thread accomplishes nothing.
After the membership has approved the new plan via a majority vote there will be an excess in money available due to the higher setup fee. The price will be determined by availability. Again, I am well aware of the numbers it just seems that you are unable to see the picture yet. Kind Regards, Stefan Prager -- Prager-IT e.U. VAT Number: ATU69773505 Austrian Company Register: 438885w Skype: Prager-IT contact@prager-it.com +43 680 300 99 80 +44 20 376 962 11

Sorry to say but I really don't think you are seeing the picture. I fail to see how dropping yearly fee's, increasing 'setup' fee's will accomplish anything. Perhaps spend some time and do the math out and put an actual outline with some numbers behind it and then match that upto the required budget to operate RIPE. You specifically say this: "After the membership has approved the new plan via a majority vote there will be an excess in money available due to the higher setup fee. The price will be determined by availability." RIPE having an 'excess' of money, will simply mean that money is returned to its members. The price is not determined by availability of resources, it is determined primarily by the operating costs of RIPE its self. Specifically lets look at the Articles of Association that govern RIPE. Article 3 – The Association: Objective The objective of the Association is to perform activities for the benefit of the Members, primarily activities that the Members need to organize as a group. This object can be sub-divided into the following activities: * Registration Activities related to the role of the Association as Regional Internet Registry; * Co-ordination Activities, including the support of the stable operation of the Internet; * Administration Activities, including all regular reports and administrative support as well as all other general administrative tasks which cannot be attributed to a specific activity; * New Activities, including all activities which are necessary to react to the rapidly changing world of the Internet; and to do all that is connected therewith or may be conducive thereto, all this in the widest sense of the word. *Making profit is not an object of the Association.* So if RIPE cannot make a profit, then it is impossible for it to set fee's so high that would make any impact and thus limits the whole game to a 21,000,000 million EUR bubble, which is TINY compared to some of the power house IT companies budgets. On 09/16/2016 06:55 AM, Prager-IT e.U. wrote:
On Fri, Sep 16, 2016 at 1:25 PM, Daniel Pearson <daniel@privatesystems.net <mailto:daniel@privatesystems.net>> wrote:
On 09/16/2016 05:57 AM, Prager-IT e.U. wrote:
The plan I envision is quite simple, move from the current Flatrate charging scheme to a system that charges members an amount for the resources they are using. This will ensure that each and every RIPE NCC member will pay their fair share.
1. Charge per allocated IPv4 address with a linear system.
price_per_ipv4 = required_budget / amount_of_addresses_in_use your_yearly_membership_fee = amount_of_ipv4_you_use * price_per_ipv4
Did you happen to read my reply to the thread. This is going to equate to almost no resources left as a /22 will cost roughly 24 EUR per year.
I read everything said so far, to prevent this from happening we also need to increase the setup fee as I already outlined.
This will also apply to Provider Independent(PI) assignemnts, anycasting assignments and Legacy Internet Resources registered via a sponsoring Local Internet Registry(LIR).
You're going to be hard pressed to charge Legacy resource holders, BUT, even if you did, my above figure of a /22 costing 24EUR would probably be cut down to 16 EUR or less
If the resources are sponsored via a sponsoring Local Internet Registry(LIR) I don't see why we should not be able to charge the Legacy resource holders.
IPv6 PI assignments and IPv6 IXP assigments will remain unchanged at 50,-- Euros per year.
Moot point here since the number of IPV6 resources avail.
The goal is to provide a comprehensive plan for the membership to vote on. The way I see this needs to be in addressed as well in order for that to happen.
If we ever reach a point where IPv4 becomes obsolete revert back to a Flatrate System as due to the nature of IPv6 a Flatrate system is a fair choice.
2. Increase the Setup Fee to an amount that reflects the current reality of the transfer market, 6.000,-- Euros.
Read the 2015 Budget, they collected slightly over 5 Million EUR in setup fee's, which got dispersed back to its members as a credit on next years bill, increasing the setup fee will simply act to lower, even further, the yearly cost. In 2015 Each LIR would have received 'roughly' a 415 EUR credit on their bill... So we might as well just not charge for IPv4 upto a /19 , and if you're going to up the setup fee's then we might as well make that a free /18.
I am well aware of the numbers. Under my plan there most likely would never be any sum significant enough to disperse.
3. Introduce an Inter-RIR Transfer Fee for resources that leave the RIPE Region so people don't flee with their resources into a cheaper Region, 3,-- Euros per IPv4 address.
If you do this, then it's a two way street. Not to mention that in the growing global age, you can't tell me every IP being routed to the ripe region is under RIPEs control. By adding additional hurdles folks will simply ignore, and announce their space globally from the originating RIR more so than they currently do now.
The goal of the change is to prevent people from fleeing with their resources allocated by the RIPE NCC from fleeing into cheaper regions after the new charging scheme has been approved.
I am confused as to why you are going on about where resources are being announced from.
4. Buy Back IPv4 resources and return them to the free pool with any excess money from members that are willing to give back resources.
With what money are these bought back and at what price? Only to re-assign at a financial loss and then have to buy it back again? Grand scheme of things, everybody here is ignoring the fact that, with basically 21,000,000 EUR budget to disperse between all LIR's in RIPE you aren't going to change the world. Aside from a lot of keyboard warriors pounding away with random ideas that have no factual backings this thread accomplishes nothing.
After the membership has approved the new plan via a majority vote there will be an excess in money available due to the higher setup fee. The price will be determined by availability.
Again, I am well aware of the numbers it just seems that you are unable to see the picture yet.
Kind Regards, Stefan Prager
-- Prager-IT e.U. VAT Number: ATU69773505 Austrian Company Register: 438885w
Skype: Prager-IT contact@prager-it.com <mailto:contact@prager-it.com> +43 680 300 99 80 <tel:%2B43%20680%20300%2099%2080> +44 20 376 962 11
---- If you don't want to receive emails from the RIPE NCC members-discuss mailing list, please log in to your LIR Portal account and go to the general page: https://lirportal.ripe.net/general/
Click on "Edit my LIR details", under "Subscribed Mailing Lists". From here, you can add or remove addresses.

On Fri, Sep 16, 2016 at 07:07:44AM -0500, Daniel Pearson wrote:
So if RIPE cannot make a profit, then it is impossible for it to set fee's so high that would make any impact and thus limits the whole game to a 21,000,000 million EUR bubble, which is TINY compared to some of the power house IT companies budgets.
Besides, it is *not* the purpose of the NCC to control the resource "market" and use fees, setup or membership, to steer it into whatever direction. This would very quickly lead to accusations of using its market domination (a cool 100% market share inthe service region) to erect barriers to trade. The NCC is the resource registry for its service region and its fees are there to defray the cost of its operation. Any discussion about using it for any other purpose (redistribution or even deterrence of speculation) is fruitless, because it would plainly be *illegal* (at least within the EU, there may be jurisdictions in the SR where this is fine). rgds, Sascha Luck

That is a very good point. Once it begins for-profit operations such as buying back IP space then competition is required by law. Didn't think about it that way as well, thanks for the input! On 09/16/2016 07:39 AM, Sascha Luck [ml] wrote:
On Fri, Sep 16, 2016 at 07:07:44AM -0500, Daniel Pearson wrote:
So if RIPE cannot make a profit, then it is impossible for it to set fee's so high that would make any impact and thus limits the whole game to a 21,000,000 million EUR bubble, which is TINY compared to some of the power house IT companies budgets.
Besides, it is *not* the purpose of the NCC to control the resource "market" and use fees, setup or membership, to steer it into whatever direction. This would very quickly lead to accusations of using its market domination (a cool 100% market share inthe service region) to erect barriers to trade. The NCC is the resource registry for its service region and its fees are there to defray the cost of its operation. Any discussion about using it for any other purpose (redistribution or even deterrence of speculation) is fruitless, because it would plainly be *illegal* (at least within the EU, there may be jurisdictions in the SR where this is fine).
rgds, Sascha Luck
---- If you don't want to receive emails from the RIPE NCC members-discuss mailing list, please log in to your LIR Portal account and go to the general page: https://lirportal.ripe.net/general/
Click on "Edit my LIR details", under "Subscribed Mailing Lists". From here, you can add or remove addresses.

Hi Stefan and All, How on earth are You, or who has resources to, do the measurement of addresses “in use”? Let me explain: Please explain “in use” means what? a / 30 v4 assigned has 1 address in use for a host, one gateway and technical addresses so it would be 25 % right? a /24 v4 assigned has maximum 253 addresses in use for hosts, one gateway, or maybe 2 you never know, somewhere around 98,4 % these assigned blocks can change many times a year, how do you take that into consideration? How are you going to see behind companies firewalls if the address is really in use…. or behind the border routers of large ISP:s if the space is “free” If you mean registering it in the Database, this can be done…. Basically a new lir needs more help than a 20 year old dino lir, probably covered by the setup fee. A higher setup fee will probably stop new entrepreneurs from starting businesses, and RIPE might get sewed for preventing this. I agree with the flat rate, no need to change. Cheers Nigel ! Rgds, Ray Elisa Corporation. From: members-discuss [mailto:members-discuss-bounces@ripe.net] On Behalf Of Prager-IT e.U. Sent: 16. syyskuuta 2016 14:55 To: members-discuss@ripe.net Subject: Re: [members-discuss] Input from Membership on RIPE NCC Charging Scheme Model On Fri, Sep 16, 2016 at 1:25 PM, Daniel Pearson <daniel@privatesystems.net<mailto:daniel@privatesystems.net>> wrote: On 09/16/2016 05:57 AM, Prager-IT e.U. wrote: The plan I envision is quite simple, move from the current Flatrate charging scheme to a system that charges members an amount for the resources they are using. This will ensure that each and every RIPE NCC member will pay their fair share. 1. Charge per allocated IPv4 address with a linear system. price_per_ipv4 = required_budget / amount_of_addresses_in_use your_yearly_membership_fee = amount_of_ipv4_you_use * price_per_ipv4 Did you happen to read my reply to the thread. This is going to equate to almost no resources left as a /22 will cost roughly 24 EUR per year. I read everything said so far, to prevent this from happening we also need to increase the setup fee as I already outlined. This will also apply to Provider Independent(PI) assignemnts, anycasting assignments and Legacy Internet Resources registered via a sponsoring Local Internet Registry(LIR). You're going to be hard pressed to charge Legacy resource holders, BUT, even if you did, my above figure of a /22 costing 24EUR would probably be cut down to 16 EUR or less If the resources are sponsored via a sponsoring Local Internet Registry(LIR) I don't see why we should not be able to charge the Legacy resource holders. IPv6 PI assignments and IPv6 IXP assigments will remain unchanged at 50,-- Euros per year. Moot point here since the number of IPV6 resources avail. The goal is to provide a comprehensive plan for the membership to vote on. The way I see this needs to be in addressed as well in order for that to happen. If we ever reach a point where IPv4 becomes obsolete revert back to a Flatrate System as due to the nature of IPv6 a Flatrate system is a fair choice. 2. Increase the Setup Fee to an amount that reflects the current reality of the transfer market, 6.000,-- Euros. Read the 2015 Budget, they collected slightly over 5 Million EUR in setup fee's, which got dispersed back to its members as a credit on next years bill, increasing the setup fee will simply act to lower, even further, the yearly cost. In 2015 Each LIR would have received 'roughly' a 415 EUR credit on their bill... So we might as well just not charge for IPv4 upto a /19 , and if you're going to up the setup fee's then we might as well make that a free /18. I am well aware of the numbers. Under my plan there most likely would never be any sum significant enough to disperse. 3. Introduce an Inter-RIR Transfer Fee for resources that leave the RIPE Region so people don't flee with their resources into a cheaper Region, 3,-- Euros per IPv4 address. If you do this, then it's a two way street. Not to mention that in the growing global age, you can't tell me every IP being routed to the ripe region is under RIPEs control. By adding additional hurdles folks will simply ignore, and announce their space globally from the originating RIR more so than they currently do now. The goal of the change is to prevent people from fleeing with their resources allocated by the RIPE NCC from fleeing into cheaper regions after the new charging scheme has been approved. I am confused as to why you are going on about where resources are being announced from. 4. Buy Back IPv4 resources and return them to the free pool with any excess money from members that are willing to give back resources. With what money are these bought back and at what price? Only to re-assign at a financial loss and then have to buy it back again? Grand scheme of things, everybody here is ignoring the fact that, with basically 21,000,000 EUR budget to disperse between all LIR's in RIPE you aren't going to change the world. Aside from a lot of keyboard warriors pounding away with random ideas that have no factual backings this thread accomplishes nothing. After the membership has approved the new plan via a majority vote there will be an excess in money available due to the higher setup fee. The price will be determined by availability. Again, I am well aware of the numbers it just seems that you are unable to see the picture yet. Kind Regards, Stefan Prager -- Prager-IT e.U. VAT Number: ATU69773505 Austrian Company Register: 438885w Skype: Prager-IT contact@prager-it.com<mailto:contact@prager-it.com> +43 680 300 99 80<tel:%2B43%20680%20300%2099%2080> +44 20 376 962 11

The measurement of addresses should be purely based upon allocation blocks (of say a /24 equivalent), not whether they are in use. It’s a relatively simple calculation to perform based upon the RIPE database. If a member hands back IP’s, the calculations that run in the following year would take this into account. Keeping the numbers simple. Say total running costs for RIPE are $100 10% of IP total RIPE space is unallocated 90% of IP total RIPE space is allocated LIR1: 30% : 100/90 * 30 :- $33.33 LIR2: 20% : 100/90 * 20 :- $22.22 LIR3: 40% : 100/90 * 40 :- $44.44 If you want to take account of new LIR setup fees, then just do $100 – (projected new LIR member count * LIR setup fee) And use that as the input to the initial running cost calculation. However it would be much easier to consider any setup fees generated from new LIR’s be put into a surplus funds pot, and used in the following year to discount the fees that everyone pays for their allocations. Also, a number of people have been making the point about RIPE making a profit if this charging model was adopted. Yes, there’s a risk of RIPE ending up in surplus, they’ve previously run a surplus in prior years, I would actually say this a wise thing to so long as it’s a reasonable figure – say 5%-10%? I’d agree with other comments about increasing costs to new LIR’s, it’s a sensible argument to reduce the potential risk of LIR’s registering to gain additional /22’s. Chris From: members-discuss [mailto:members-discuss-bounces@ripe.net] On Behalf Of Jetten Raymond Sent: 16 September 2016 1:24 PM To: Prager-IT e.U.; members-discuss@ripe.net Subject: Re: [members-discuss] Input from Membership on RIPE NCC Charging Scheme Model Hi Stefan and All, How on earth are You, or who has resources to, do the measurement of addresses “in use”? Let me explain: Please explain “in use” means what? a / 30 v4 assigned has 1 address in use for a host, one gateway and technical addresses so it would be 25 % right? a /24 v4 assigned has maximum 253 addresses in use for hosts, one gateway, or maybe 2 you never know, somewhere around 98,4 % these assigned blocks can change many times a year, how do you take that into consideration? How are you going to see behind companies firewalls if the address is really in use…. or behind the border routers of large ISP:s if the space is “free” If you mean registering it in the Database, this can be done…. Basically a new lir needs more help than a 20 year old dino lir, probably covered by the setup fee. A higher setup fee will probably stop new entrepreneurs from starting businesses, and RIPE might get sewed for preventing this. I agree with the flat rate, no need to change. Cheers Nigel ! Rgds, Ray Elisa Corporation. From: members-discuss [mailto:members-discuss-bounces@ripe.net] On Behalf Of Prager-IT e.U. Sent: 16. syyskuuta 2016 14:55 To: members-discuss@ripe.net Subject: Re: [members-discuss] Input from Membership on RIPE NCC Charging Scheme Model On Fri, Sep 16, 2016 at 1:25 PM, Daniel Pearson <daniel@privatesystems.net> wrote: On 09/16/2016 05:57 AM, Prager-IT e.U. wrote: The plan I envision is quite simple, move from the current Flatrate charging scheme to a system that charges members an amount for the resources they are using. This will ensure that each and every RIPE NCC member will pay their fair share. 1. Charge per allocated IPv4 address with a linear system. price_per_ipv4 = required_budget / amount_of_addresses_in_use your_yearly_membership_fee = amount_of_ipv4_you_use * price_per_ipv4 Did you happen to read my reply to the thread. This is going to equate to almost no resources left as a /22 will cost roughly 24 EUR per year. I read everything said so far, to prevent this from happening we also need to increase the setup fee as I already outlined. This will also apply to Provider Independent(PI) assignemnts, anycasting assignments and Legacy Internet Resources registered via a sponsoring Local Internet Registry(LIR). You're going to be hard pressed to charge Legacy resource holders, BUT, even if you did, my above figure of a /22 costing 24EUR would probably be cut down to 16 EUR or less If the resources are sponsored via a sponsoring Local Internet Registry(LIR) I don't see why we should not be able to charge the Legacy resource holders. IPv6 PI assignments and IPv6 IXP assigments will remain unchanged at 50,-- Euros per year. Moot point here since the number of IPV6 resources avail. The goal is to provide a comprehensive plan for the membership to vote on. The way I see this needs to be in addressed as well in order for that to happen. If we ever reach a point where IPv4 becomes obsolete revert back to a Flatrate System as due to the nature of IPv6 a Flatrate system is a fair choice. 2. Increase the Setup Fee to an amount that reflects the current reality of the transfer market, 6.000,-- Euros. Read the 2015 Budget, they collected slightly over 5 Million EUR in setup fee's, which got dispersed back to its members as a credit on next years bill, increasing the setup fee will simply act to lower, even further, the yearly cost. In 2015 Each LIR would have received 'roughly' a 415 EUR credit on their bill... So we might as well just not charge for IPv4 upto a /19 , and if you're going to up the setup fee's then we might as well make that a free /18. I am well aware of the numbers. Under my plan there most likely would never be any sum significant enough to disperse. 3. Introduce an Inter-RIR Transfer Fee for resources that leave the RIPE Region so people don't flee with their resources into a cheaper Region, 3,-- Euros per IPv4 address. If you do this, then it's a two way street. Not to mention that in the growing global age, you can't tell me every IP being routed to the ripe region is under RIPEs control. By adding additional hurdles folks will simply ignore, and announce their space globally from the originating RIR more so than they currently do now. The goal of the change is to prevent people from fleeing with their resources allocated by the RIPE NCC from fleeing into cheaper regions after the new charging scheme has been approved. I am confused as to why you are going on about where resources are being announced from. 4. Buy Back IPv4 resources and return them to the free pool with any excess money from members that are willing to give back resources. With what money are these bought back and at what price? Only to re-assign at a financial loss and then have to buy it back again? Grand scheme of things, everybody here is ignoring the fact that, with basically 21,000,000 EUR budget to disperse between all LIR's in RIPE you aren't going to change the world. Aside from a lot of keyboard warriors pounding away with random ideas that have no factual backings this thread accomplishes nothing. After the membership has approved the new plan via a majority vote there will be an excess in money available due to the higher setup fee. The price will be determined by availability. Again, I am well aware of the numbers it just seems that you are unable to see the picture yet. Kind Regards, Stefan Prager -- Prager-IT e.U. VAT Number: ATU69773505 Austrian Company Register: 438885w Skype: Prager-IT contact@prager-it.com +43 680 300 99 80 <tel:%2B43%20680%20300%2099%2080> +44 20 376 962 11
participants (5)
-
Chris Smith
-
Daniel Pearson
-
Jetten Raymond
-
Prager-IT e.U.
-
Sascha Luck [ml]