- members-discuss - mailman.ripe.net

Re: [members-discuss] [Ticket#2012072401002498] Top 20 LIRs use 33% Allocated PA, 88% of LIRs use 10% Allocated PA.
by Adam Waite 02 Aug '12

02 Aug '12

On Mon, Jul 30, 2012 at 10:12 PM, Rob Golding <rob.golding(a)othellotech.net> wrote: >> If you consume water, you don't pay support ticket, >> you pay liters of water. > > Depends on where you live - that's not the case in all locations or all countries - in the UK most people pay a flat *minimal* fee for their water (mine is £114/year) no matter how much you "consume" > That is because the big consumers of water in the UK have succeeded in externalizing their costs onto smaller consumers. And that's exactly what this discussion thread has been about -- trying to force larger consumers of a finite resource to pay in proportion to their consumption.

3 2

Re: [members-discuss] [Ticket#2012072401002498] Top 20 LIRs use 33% Allocated PA, 88% of LIRs use 10% Allocated PA.
by Sascha Luck 31 Jul '12

31 Jul '12

On Mon, Jul 30, 2012 at 09:12:35PM +0100, Rob Golding wrote: >The post office don???t charge for a postcode (necessary to deliver >your mail in most countries) and the is a finite number of those - they >make their money to pay for the database management of postcode through >other services which are *optional*. I'm not sure I'd want the NCC to sell my address to spamm^Wmarketers, even if that means that membership be free... ;) sncr, Sascha

2 1

Re: [members-discuss] Next RIPE members GM, in A'dam
by Brandon Butterworth 26 Jul '12

26 Jul '12

> To follow up the discussion started at the Ljubjana meeting about Legacy > Resources registration in the RIPE DB, we (LR Holders) are asking forac an > agenda slot (15-20 mn would be sufficient) to have a formal discussion > on this topic based on a policy proposal we're working on. What's the proposal? We have legacy and are happy with it just as it is. Actually I don't see why anyone needs to fiddle with v4 stuff (c.f. the pricing discussion). v4 is done, people should spend their time on v6, do somehting useful brandon

3 3

Re: [members-discuss] Charging scheme discussion
by Max Tulyev 25 Jul '12

25 Jul '12

24.07.12 22:38, Rob Golding написав(ла): > If we're going to "accelerate"the acceptance of IPv6 (which is in all of our > interests) then anyone using PI IPv4 _probably_ needs some PI IPv6 Absolutely sure. The majority of our IPv4 PI customers at least asked us for IPv6 PI. But almost none of them qualify to get it. Others don't want to pay for something can't do the money income. So if RIPE/RIPE NCC really want to push IPv6 - let's do two easy things: change the qualify details for IPv6 PI and provide it free of charge for several years.

6 7

Re: [members-discuss] Charging scheme discussion
by Nick Hilliard 25 Jul '12

25 Jul '12

On 24/07/2012 20:38, Rob Golding wrote: > If we're going to "accelerate"the acceptance of IPv6 (which is in all of our > interests) then anyone using PI IPv4 _probably_ needs some PI IPv6 > > If you're going to charge then for both you put an artificial barrier > (doubling their cost), and they'll never want the v6 ... Have you considered that one hour of engineering time would probably cost a company ~€150 - €300, including overheads? So your argument is that a cost equivalent to 10 minutes of engineering time for a company is too much for the company to bear if they want a provider independent v6 presence on the Internet. I'd respectfully suggest that if an organisation views €50 / annum as a serious barrier to adoption of a technology which is important to their business success, then that organisation has severe priority inversion problems. >> There should be a differentiation in charge depending on assignment size, >> e.g. a /18 is charged more than a /24. > > Why ? > > What more work is there for RIPE to do ? e.g. requests outside an assignment window. Can I suggest you read the RIPE Annual Report for 2011? It gives a breakdown of the annual expenditure for the company, including the registration services costs. >> There is no need to charge for ASNs. > > 16bit ASNs are equally in short supply, and the majority of deployed routers > don't understand longer ASNs Intermediate routers don't need to understand ASN32 because that's handled by clever translation stuff. Edge routers which use an ASN32 as the local asn do need to understand ASN32 natively, but this support has been available on all cisco ios based platforms since 2009 and all juniper platforms since 2008. If you're running software this old on your transit routers, you have bigger problems. Nick

5 6

Re: [members-discuss] Charging scheme discussion
by Mark Jones (MK) 25 Jul '12

25 Jul '12

It's also not helped by some vendors cranking up expensive licence fees for companies to be able to use IPV6 on their products.. Mark -----Original Message----- From: members-discuss-bounces(a)ripe.net [mailto:members-discuss-bounces@ripe.net] On Behalf Of Rob Golding Sent: 25 July 2012 04:05 To: members-discuss(a)ripe.net Subject: Re: [members-discuss] Charging scheme discussion >> If we're going to "accelerate"the acceptance of IPv6 (which is in all >> of our >> interests) then anyone using PI IPv4 _probably_ needs some PI IPv6 >> >> If you're going to charge then for both you put an artificial barrier >> (doubling their cost), and they'll never want the v6 ... > > Have you considered that one hour of > engineering time would probably > cost a company ~€150 - €300, including overheads? Regarding IPv6 ... End users don’t know what it is, don’t appear to want it, from take up certainly don’t seem to need it, and basically don’t care about it - they just want their pr0n to turn up quickly. > So your argument is that a cost > equivalent to 10 minutes of engineering time for a company is too much > for the company to bear if they want a provider independent v6 > presence on the Internet. No, I'm saying they don’t want ipv6 at all as they see no requirement for it, then you add making them pay for it in addition to their ipv4 will get you the response (from experience) "no thanks, no-one uses that" >> e.g. a /18 is charged more than a /24. >> Why ? >> What more work is there for RIPE to do ? > e.g. requests outside an assignment window. If my window is a /16 then the /18 is less work, than when your window is /26 and you request a /24 :p > Can I suggest you read the RIPE Annual Report for 2011? Been there, done that, have my own opinions on the expenditure ... > local asn do need to understand ASN32 natively, but this support has > been available on all cisco ios based platforms since 2009 and all > juniper platforms since 2008. > If you're running software this old on your transit routers, you have > bigger problems. You missed the point where I explained that we did all this 10 years ago ! 89% of the worlds desktops are running an OS which does things wrong with IPv6 DNS results when it gets a v6 result from DNS lookups and doesn’t have any v6 routes, so putting a site/service/system on ipv6 only cuts your potential audience by 9/10ths Deliberately putting people off of IPv6 by telling them they'll need expensive tech time (as you say) plus more contributions to ripe fees is not going to help takeup - my experience, YMMV Rob ---- 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/view Click on "Edit my LIR details", under "Subscribed Mailing Lists". From here, you can add or remove addresses. ############################################################################## This communication together with any attachments transmitted with it ("this E-Mail") is intended only for the use of the addressee and may contain information which is privileged and confidential. If the reader of this E-Mail is not the intended recipient or the employee or agent responsible for delivering it to the intended recipient you are hereby notified that any use, dissemination, forwarding, printing or copying of this E-Mail is strictly prohibited. Addressees should check this E-mail for viruses. The Company makes no representations as regards the absence of viruses in this E-Mail. If you have received this E-Mail in error please notify our IT Service Desk immediately by e-mail at abuse.ttb(a)talktalkplc.com Please then immediately delete, erase or otherwise destroy this E-Mail and any copies of it. Any opinions expressed in this E-Mail are those of the author and do not necessarily constitute the views of the Company. Nothing in this E-Mail shall bind the Company in any contract or obligation. For the purposes of this E-Mail "the Company" means TalkTalk Telecom Group PLC and/or any of its subsidiaries. Please feel free to visit our website: www.talktalkgroup.com TalkTalk Telecom Group Plc (Registered in England & Wales No. 7105891) 11 Evesham Street, London W11 4AR ##############################################################################

1 0

Re: [members-discuss] Charging scheme discussion
by Andrea Cocito 25 Jul '12

25 Jul '12

Hi, On Jul 24, 2012, at 9:38 PM, Rob Golding wrote: > "count" of ip's within a block - no, it's back to them being a "commodity" > (altering the tax status) This is wrong and leads to misinformation. Until last year RIPE membership had a cost depending on a class model (large, small, extra large, etc), which on its behalf depended on the number of allocated resources. Until last year (and now, so far) RIPE was classified as non-profit. These are facts. So it IS POSSIBLE to have a membership fee that depends on the number of allocated resources without changing the non-profit status, and an example has been provided. Everybody understands that RIPE cannot "sell" address space, everybody understands that RIPE can have different classes of membership which different fees, where the class of membership depends on the number of allocated resources and the fee is effectively proportional to that, all this maintaining a non-profit status. Continuing to confuse these two issues, which have nothing to do each with the other, sounds like plain misinformation: The "Tax issue" is not an issue, unless further information showing the opposite is presented. Regards, A.

4 4

Re: [members-discuss] Charging scheme discussion
by Nick Hilliard 25 Jul '12

25 Jul '12

On 25/07/2012 04:05, Rob Golding wrote: > End users don’t know what it is, don’t appear to want it, from take up > certainly don’t seem to need it, and basically don’t care about it - > they just want their pr0n to turn up quickly. Certainly desktop users are most interested in just doing whatever they want to do and don't care too much about the underlying protocols. But I'm getting the impression that you're mixing up what an End User means here. If you're the sort of end-user who requires provider independent addressing, you're probably a small-medium sized content provider, or perhaps a very large large commercial customer. For customers of this form, €50 / annum is an insignificant financial outlay. If you're the sort of end user who just wants to browse stuff on your desktop, ipv6 PI is completely irrelevant to you. Your service provider can assign you some PA space and you'll probably never even notice. Look, you clearly feel very strongly that the equivalent cost of 10 minutes of engineering time is an insurmountable obstacle for the sort of organisation which is going to need IPv6 PI address space, even though an IPv6 rollout may take anything from weeks to years to implement fully, with proportional total costs of ownership. So let's agree to disagree on this point. >> local asn do need to understand ASN32 >> natively, but this support has been >> available on all cisco ios based platforms >> since 2009 and all juniper >> platforms since 2008. >> If you're running software this old on your >> transit >> routers, you have bigger problems. > > You missed the point where I explained that we did all this 10 years ago ! Nope, didn't miss it. I've no doubt that you had servers dual-stacked 10 years ago. I've even less doubt that ipv6 was nowhere near ready for large-scale deployment 10 years ago. Dual-stacking some servers is easy. Rolling v6 services out to eyeballs, not so much. > 89% of the worlds desktops are running an OS which does things wrong > with IPv6 DNS results You may be confusing ipv6 with asn32s here? I was talking about asn32s. > when it gets a v6 result from DNS lookups and doesn’t have any v6 > routes, so putting a site/service/system on ipv6 only cuts your > potential audience by 9/10ths I suspect your understanding of global ipv6 deployment problems may be substantially different to mine. Let's just agree that this isn't really relevant to members-discuss@ and go back to discussing the charging scheme. Nick

1 0

Re: [members-discuss] Charging scheme discussion
by LIR 25 Jul '12

25 Jul '12

Il 25/07/2012 04:07, Rob Golding ha scritto: >>> "count" of ip's within a block - no, it's back >>> to them being a "commodity" >>> (altering the tax status) >> This is wrong and leads to misinformation. >> Until last year RIPE membership had a cost >> depending on a class model >> (large, small, extra large, etc), which on >> its behalf depended on the >> number of allocated resources. > *number* of resources, not *size* of resources > > So thank you for confirming my point:p Actual calculation is based on number and size of resources, which the exception of age, where old users pay much less than new users. What should be eliminated is the NOT FAIR gift made to old users. Just eliminate the discount based on age and scale proportionally new prices to cover RIPE costs. Regards, Tonino

1 0

Re: [members-discuss] Charging scheme discussion
by hostmaster＠freethought-internet.co.uk 25 Jul '12

25 Jul '12

----- Original Message ----- > From: "Rob Golding" <rob.golding(a)othellotech.net> > To: members-discuss(a)ripe.net > Sent: Wednesday, 25 July, 2012 3:07:42 AM > Subject: Re: [members-discuss] Charging scheme discussion > > >> "count" of ip's within a block - no, it's back > >> to them being a "commodity" > >> (altering the tax status) > > This is wrong and leads to misinformation. > > Until last year RIPE membership had a cost > > depending on a class model > > (large, small, extra large, etc), which on > > its behalf depended on the > > number of allocated resources. > > *number* of resources, not *size* of resources > > So thank you for confirming my point:p > > If the charging switches to per-IP then they become "product" or > "stock" or > "items" or similar, and have a value, and that WILL change the > tax-status of > RIPE - which at current levels, will mean 25% tax which *WILL* have > to be > taken from members in fees > > So, hands up who wants a 25% price hike ? > > Rob > No, there are five different billing categories for ranges of scores and each billing category has a price associated with it. The scores are calculated from a formula which includes the time of allocation relative to 1992 and the size of the allocation from a /19 to a /22 in IPv4 or a /30 to a /33 in IPv6. The scoring units increase with the number of IP addresses in an allocation, so someone with a /20 allocated in 2012 has a score double that of someone with a /21 allocated in 2012 (40 and 20 respectively). There is a clear emphasis on the size of resources allocated. Edward Dore Freethought Internet

2 1