- lir-wg - mailman.ripe.net

Re: PI vs PA Address Space
by Steven J. Richardson 18 May '95

18 May '95

I agree with Vince.

1 0

Re: PI vs PA Address Space
by Vince Fuller 18 May '95

18 May '95

> Tony Li <tli(a)cisco.com> writes: > > [cc list reduced] Why exclude nanog? Well, I'm the one who requested it. I requested this because it was getting difficult to follow this conversation when multiple copies of the same message (and its followups) were arriving via paths with wildly different propagation delays (NANOG and CIDR appear to be housed on machines with very different available bandwidth, at least relative to where I am). I also feel that this discussion is very CIDR-specific, so anyone interested in it should be on the CIDR list. --Vince

1 0

RE: Privacy of info in IP requests
by Wilfried Woeber, UniVie/ACOnet 18 May '95

18 May '95

Sorry, I seem to have sent a couple of people the wrong way. I didn't want to point fingers to legalese... >Just to remind everyone, ripe-104 says: > > 4. IRs will keep records of correspondence and information > exchanges in conjunction with the registry function for later > review and the resolution of disputes. IRs will hold this > information in strict confidence and use it only to review > requests and in audit procedures or to resolve disputes. I'm aware of that. >The RIPE NCC makes every reasonable effort to keep the information we >store confidential. On the other hand we have to store it electronically >on machines connected -albeit restrictively- to the Internet. > *That's* the things that triggered me! Is it acceptable to have the files stored in a plain office? Do I have to lock the doors of my office when I leave? [ I'm doing that anyway, for various other reasons...] Who is "the Local-IR"? My group? Just a couple of individuals? What's the position of my boss in this? [ I *certainly* don't have a problem with this in our shop ] If I ask the folks in another ACOnet PoP, whether they intend to physically and administratively connect the nets of the applicant, is it a problem telling them who applied for addresses? [ Sure I *do* remove the more specific parts of the application, and when the application is granted, the person and network data is stored in the RIPE-DB, thus becoming public knowledge anyway...] On another aspect, when I receive an application (for a network number or a domain) giving person objects (admin-c's in particular!) for people that I did not talk to personally, should we be double-checking before accepting the application and putting things into the database? [ Right now it is our policy to cc them on the allocation and other correspondence...] You see, I'm not trying to find something that can stand a lawyer, I'm trying to find out what the "common sense" and "state of the art" and the "reasonable effort" and "strict confidence" is. Because we no longer deal exclusively with univiersities and research centres, but also with regional hospitals, security departments of regional government, and the like. But I might, again, be barking at the wrong tree... Thanks for reading that far, Wilfried. PS: Btw, this is just another thing where I think the Local-IR stuff overlaps the (administrative aspects of the) DNS stuff... -------------------------------------------------------------------------- Wilfried Woeber : e-mail: Woeber(a)CC.UniVie.ac.at Computer Center - ACOnet : Vienna University : Tel: +43 1 4065822 355 Universitaetsstrasse 7 : Fax: +43 1 4065822 170 A-1010 Vienna, Austria, Europe : NIC: WW144 --------------------------------------------------------------------------

2 1

Re: Privacy of info in IP requests [was: Re: Draft internic ip allocation doc]
by Dave Morton 18 May '95

18 May '95

> > Dave Morton <Dave.Morton(a)ecrc.de> writes: > > > > I wouldn't bother - paste it out of 104 and put some juicy legal sounding > > terminology around it and if anyone wants one sign it and fax it to them. > > There is nothing really legally complicated that I can see - non-disclosure > > agreeements are common business practice used every day - one doesn't need > > a lawyer to formulate one IMHO. > >That's what I will certainly not do, because it may make things worse than >I expect *if* I get sued. Terms like "inent", "negligence", "gross negligence", >"reasonable precautions", "state-of-the-art precations" start ringing in my >ears. I want to know beforehand what my exposure is and if I must disconnect >my machines from the net and keep them in a locked vault ;-). > Daniel - on second thoughts, and considering all the things that could go wrong, you're probably well advised to get legal advice beforehand. I hate it when that happens with the fingers and the send key but without the brain .... >Daniel Dave PS: I see you've also learned to develop an unhealthy respect for legal beavers recently - life is getting more complicated - not just in Amsterdam.

1 0

Re: Privacy of info in IP requests [was: Re: Draft internic ip allocation doc]
by Dave Morton 18 May '95

18 May '95

> > Dave Morton <Dave.Morton(a)ecrc.de> writes: > > Wilfried - we already have had the situation where we've been asked to > > sign a non-disclosure agreement for the same reasons as stated. > >Was that in your capacity as a lcoal IR? Yes. Mabye I should be thinking of asking the Local IR list to sign the same or a similar agreement before proceeding here :-))))0 > > I'd > > expect it to become more common with more commercial entities appearing. > >Probably. We have had no concrete request to sign anything yet. > > > It might even be an idea for RIPE NCC and local IRs to perhaps offer this > > anyway to organisations seeking space etc. simply to avoid any potential > > legal hassles later. Yes - I know things are treated confidentially but > > it may be better to make it explicit. > >It is explicit in ripe-104 and the request forms. Yes - thanks for pointing this out again - I had forgotten... >Making it more explicit would involve lawyers here, take time and cost money. So would getting sued.... >I can investigate a general nondisclosure statement a sufficient number of >the lcoal-IRs think this is a good idea they want to pay for. I suspoct >however that it will again be a question of national law etc. I wouldn't bother - paste it out of 104 and put some juicy legal sounding terminology around it and if anyone wants one sign it and fax it to them. There is nothing really legally complicated that I can see - non-disclosure agreeements are common business practice used every day - one doesn't need a lawyer to formulate one IMHO. >Daniel > Dave

2 1

Re: Privacy of info in IP requests [was: Re: Draft internic ip allocation doc]
by Dave Morton 18 May '95

18 May '95

Wilfried - we already have had the situation where we've been asked to sign a non-disclosure agreement for the same reasons as stated. I'd expect it to become more common with more commercial entities appearing. It might even be an idea for RIPE NCC and local IRs to perhaps offer this anyway to organisations seeking space etc. simply to avoid any potential legal hassles later. Yes - I know things are treated confidentially but it may be better to make it explicit. Cheers, Dave

2 1

Privacy of info in IP requests [was: Re: Draft internic ip allocation doc]
by Wilfried Woeber, UniVie/ACOnet 18 May '95

18 May '95

Dear Local-IR folks! I don't know how many of us are following the PI vs. PA address discussion on CIDRD and other lists, probably quite a few - apologies to those who now get just another copy - but the message as appended below strung a cord inside me. More so after talking to people from the DE-NIC a while ago, who as I understood worried about security and privacy of the information stored at their premises. I'd be really interested in the point of view of others. I recognize that we, being a SP registry for an R&D community might have have less stringent requirements, but even then the delimiting criteria become more and more blurred these days... Wilfried. PS: Mike, apologies for not asking you before hitting the LOCAL-IR list, but I think it's really something to think about... -------------------------------------------------------------------------- Wilfried Woeber : e-mail: Woeber(a)CC.UniVie.ac.at Computer Center - ACOnet : Vienna University : Tel: +43 1 4065822 355 Universitaetsstrasse 7 : Fax: +43 1 4065822 170 A-1010 Vienna, Austria, Europe : NIC: WW144 -------------------------------------------------------------------------- From: Mark Kosters <markk(a)internic.net> Subject: Re: Draft internic ip allocation doc To: peter(a)agis.net (Peter Kline) Date: Wed, 17 May 1995 13:40:46 -0400 (EDT) CC: markk(a)internic.net, nanog(a)merit.edu, cidrd(a)iepg.org, kimh(a)internic.net > > What's being asked for here is what a lot of businesses might consider > proprietary or trade secret type information. There are few secrets on the > Internet, we all know, once the network is in place, but plans for expansion > and new business ought to be confidential until the day the circuit is > turned on. > > To an extent this problem can be mitigated by private IP numbering, etc., > but more and more I believe people will resist submitting this kind of > information. We are very sensitive to the privacy of this information - to the point that the we are considered within the company to be fanatical on not letting anyone who is not a part of this project be able to touch our net(s) or operations area in any way. I've requested our lawyer to have a non-disclosure agreement ready for those who would like to sign up to one before submitting confidential information. Mark -- Mark Kosters markk(a)internic.net +1 703 742 4795 Software Engineer InterNIC Registration Services --------------------------------------------------------------------------------

3 2

RE: PI VS PA ADDRESS SPACE
by 116780＠ibmmail.com 18 May '95

18 May '95

Daniel Karrenberg <Daniel.Karrenberg(a)ripe.net> wrote: > >I fully agree with you. However we get a significan number of requests >from enterprises who -at this point- do not want "Internet service" but >want to build private internets with other enterprises. This demonstrates >that there is an intrinsic value in the uniqueness of a number even >without "Internet service" with a capital I. I believe that IRs should >continue to provide this uniqueness for reasonable requests. This is very similar to the position we are in - we have obtained registered addresses in the first instance (i.e. as a solution to an immediate problem) to make interconnections to other companies' networks. However, we also wished to have unique addresses for future direct Internet access, and made this clear in our applications. Having gone through the considerable hassle of setting up an enterprise registry and renumbering our networks with registered addresses, we would take a very dim view if we were then told that our addresses were then not routable over the Internet at large! Andrew Cowie uk.royal Extra X400 information begins: Originator Name: COWIEA Org Units: LIVERPOOL Organisation: NHP Domain: GB/IBMX400/ROYINT Node.Userid: IBMX400.116780 Message Id: A70726C5.MAI Sent by Name: COWIEA Org Units: LIVERPOOL Organisation: NHP Domain: GB/IBMX400/ROYINT Node.Userid: IBMX400.116780 Free Fmt Name: COWIE, Andrew / NHP Subject: RE: PI VS PA ADDRESS SPACE Recipients Name: INET INET Domain: GB/IBMX400/IBMMAIL Node.Userid: IBMMAIL.INET Free Fmt Name: INET, INET Reply request: No

1 0

None
by SAEED KHADEMI 18 May '95

18 May '95

subscribe

1 0

Re: PI vs PA Address Space
by woody＠zocalo.net 17 May '95

17 May '95

prue(a)ISI.EDU writes: > In order to charge enough to cause behavior changes the > backbones would have to charge a lot for individual routes. The > cost for internet access is quite high right now so a charge of > say $100/year would be a drop in the bucket compared to the > other costs. I'm entering the conversation in mid-stream, so please excuse me if I reiterate something. I don't know where the discussion of funding the InterNIC after public funding runs out has gone, but it seemed to me as though something exactly like this would be a reasonable means... Rather than making it revenue-neutral and handled by backbone carriers, make it a source of funding for the InterNIC, which would scale as their load increased. A combination of very small fees for domain name registration, and a somewhat larger fee for a route. But not _too_ large a fee. :-) Both intended be be passed down the food chain as far as possible, to keep people on their best behaviour. We're a mid-sized regional, with a partially-populated B, and five or six grandfathered-in Cs, that customers from way back have had. About three months ago, we decided to clean things up, and convinced several of our customers to move from a CIDR block and some of their own Cs to subnets of our B, and then stopped routing the Cs and returned the CIDR block. In order to do this, we had to do quite a bit of persuasion, and the fact that we levied a $100/year fee for routes didn't seem to make an impression on anyone. The Little Garden, another regional here, also implemented a $100/year fee, at the same time, and they're having aboutthe same results. So I think a fee would have to be up more in the $1,000-$1,500/year range, but commensurate with that, the NIC would have to be willing to give out much larger CIDR blocks, rather than lots of little ones as they seem to now. Just my thoughts. -Bill Woodcock ________________________________________________________________________________ bill woodcock woody(a)zocalo.net woody(a)applelink.apple.com user(a)host.domain.com

1 0