- Test - lists.nanog.org

Re: Class "B" forsale (fwd)
by Steven M. Bellovin 10 Mar '97

10 Mar '97

At 08:23 AM 3/10/97 +1000, Geoff Huston wrote: >Of course this scenario breaks down quickly.. > >Listed registrant A "sells" to B who "sells" to C who... > >Now the sucker who "buys" from C has the problem of tracing authenticity >of the "title" C is selling. This is not easy, particularly if B is less >than scrupulous and has "sold" title to a number of folk including >C, or if C is unscrupulous and is "selling" without have concluded >and transaction with B. This can easily be solved in a number of ways. For example, a public key could be publicly associated with each address. A transfer is accomplished by signing a message to that effect. Similarly -- and very importantly -- control over routing of that address is also governed by that private/public key pair. There's more to this protocol than I've described, but it can be made to work.

1 0

RE: measurement
by Jim Browning 10 Mar '97

10 Mar '97

The measurement techniques described thus far focus on performance within the bounds of a single network. While this is of course a challenge, what about efforts to measure performance _across_ networks? I'm not talking about NAP packet loss here, but a true measure of expected customer satisfaction. Most customers do not particularly care why they can't download the latest MSIE/Navigator release quickly, and getting traffic to and through other networks is of equal importance. Network rating systems are starting to emerge, and I think NANOG should participate in their evolution... -- Jim Browning ---------- From: Randy Bush[SMTP:randy@psg.com] Sent: Saturday, March 08, 1997 8:35 PM To: nanog(a)merit.edu Subject: Re: measurement I promised to summarize responses to my query > So who actually measures their network performance and how? As most responses were private, I have removed attribution. Thanks to all constructive respondees. I have proposed a survey panel for the next NANOG if we do not exhaust the subject beforehand. randy

4 3

Re: Cayman Island Scenarios
by Dave O'Shea 10 Mar '97

10 Mar '97

Jim Fleming <JimFleming(a)unety.net> writes > > Question: When companies like MCI and Bellcore get bought, > do they have to turn all of their blocks back into the "NIC" > and start over...;-) I actually went though an amazingly similar situation when we sold off one of our subsidiaries; InterNIC said we could not have new address space because of the B that had been assigned to the newly-sold subsidiary; that we could not "transfer" that block of space. I offered to assume responsibility for the old block, if they would certify me as the authority to have routing turned off for the old subsidiary. In the end, we got new space. All things considered, it would have been nice if it had been a block of addresses that could have been "split" - like a /16 up above 192.x.x.x without confusing the bejeezus out of simple RIP configurations. Would have saved us the "next time I'll just kill myself" task of renumbering, and conserved address space. I think the use of unregistered blocks and better proxy servers is the only way we can avoid this kind of silliness - with a good setup, a Fortune 500 company has only a handful of "visible" IP addresses, with the rest hidden and irrelevant.

1 0

Routing Information guard (was Class "B" forsale )
by Ichiro Mizukoshi 10 Mar '97

10 Mar '97

Chis's Idea is nice. Routing engineers in ISP sometimes misconfigure their gear, so bogus route will be announced. Checking the Routing Information ande sending the warning messages to the ISP, makes the network more stable. And, let me toss an additional idea to it. *) If an allocated addresses does not appear in such table for some period, asking the aloocated guy to return them. Ichiro Mizukoshi e-mail(office):ichiro@TokyoNet.ad.jp Tel:+81-3-3341-6301 Fax:+81-3-3341-6305 From: "Chris A. Icide" <chris(a)nap.net> Subject: Re: Class "B" forsale (fwd) Date: Sat, 08 Mar 1997 04:10:54 -0600 > Now I don't even imagine to propose any comprehensive solution for > this. However let me toss an idea on the table. If in fact ARIN does come > into being as an organization supported and funded by an for the commuity > of IP users, then those IP users could work in cooperation with ARIN (and the > other NIC's) to police the usage of IP's. For example, the NICs could > maintain a database online that could monitor the current BGP tables. > If the NIC determines that an unallocated IP block is in use, they could then > insert a prefix entry into the table, seriously degrading the value of that > IP block. This could also be used in a gross way to evaluate usage of > prefixes, and unauthorized changes to prefixes. > > Again, this is only an idea that I toss on the table, but I do stand by the > fact that unless there is a negetive incentive, there are those who will > do anything that is beneficial to them. > > Chris A. Icide > Sr. Engineer > Nap.Net, L.L.C.

1 0

RE: since my name was mentioned (was Re: The Mother of all Solutions)
by Jim Fleming 10 Mar '97

10 Mar '97

On Sunday, March 09, 1997 8:00 PM, Paul A Vixie[SMTP:paul@vix.com] wrote: @ > If a LARGE GROUP OF NETWORK PROVIDERS (that's us, btw, nanog), @ > decided TOMORROW that WE will assign address space and route to @ > it, there is no force in the world that will charge for it, or @ > be able to change it. @ @ That's what we already have. IANA is in charge because the people who own @ the physical plant -- that's the multinationals, larger nationals, government, @ military, and other gigantic users -- think IANA is a good solution for the @ time being. IANA delegates its address assignment authority to registries @ (RIPE, APNIC, and ARIN/InterNIC) whose operational guidelines are set by and @ reviewed by open forums made up of the people to whom addresses are allocated, @ with some oversight/assistance from IETF. @ TRANSLATION: The "big boys" support the status quo...that's not surprising... @ If the people who own the physical plant were to somehow jointly decide that @ some other system would work better for them, then that other system would be @ in place (or die trying) pretty much instantaneously, with no relevant @ fighting. (It's worth noting that confusion over the ownership of the @ physical plant is what makes Karl, Eugene, and Jim try to do what they're @ trying to do with ".", but it's probably not worth discussing over again.) @ TRANSLATION: If a whole bunch of average people make a change in unison, then no one, including the "big boys", can stop them... <snip - Root Name Server operational error story> @ > Let me know if I've left something out. @ @ What you've left out is that the model of Internet self governance has been @ in use since before the U.S. Military thought it had allowed such, and is in @ use now even though it looks rather autocratic to someone who does not know @ from whence IANA and RIPE/APNIC/ARIN derive their relevance. @ @ TRANSLATION: The status quo is based on years of momentum... as well as U.S. Government and DOD support... ...again, not surprising... @@@@@@@@ Alternate Viewpoint... The Internet will never be significantly changed by the people in power. Only the collective will of the people can result in changes. To make changes you have to be open to change. Change for the sake of change is not good. Change for the sake of ensuring that governments and small autocratic societies do not dominate the mediums IS good. Change to allow people of all sexes, sizes, shapes, ages, races, religions, sexual preferences, etc. and viewpoints IS good. The U.S. Government's blockade of new participants in the Internet will eventually end and history will show that thousands of hours were spent with one small group supporting the status quo and another large group advocating an open, fair and level playing field. The Internet will be used to reinvent itself and the few "big boys" will play no significant role other than to provide mass for the blockade which will be lifted by the vary government that paid to have it installed...that lifting is happening as NANOG sleeps...;-) -- Jim Fleming Unir Corporation e-mail: JimFleming(a)unety.net JimFleming(a)unety.s0.g0 (EDNS/IPv8)

1 0

re: class B for sale
by Brett L. Hawn 10 Mar '97

10 Mar '97

So that I'm not misunderstood let me say this: 1: I do not neccessarily agree with the sale of IPs, personally, I don't think its a good idea 2: This is a real world economy now, outdated academic practices which are currently being enforced are as wrong as the sale of IPs. 3: Wether you, ARIN, or anyone else likes it or not, IPs are for all intents and purposes a resellable commodity, otherwise ARIN et all can (ala Jim Flemming) be called on as being a Monopoly. 4: The simple fact of the matter is that the RFCs are not at any time, the law of the land. They are at best guidelines and good ideas set down for others to follow, but there is no rule stating that you _must_ follow them. 5: Before you start chasing wild geese selling Class B address space I suggest you go back and check on all those folks that got space long before there were any 'restrictions and justifications'. I have no doubt that there is a veritable feast of IPs sitting unused at MIT, USC, and other such institutions that would be better used elsewhere instead of sitting in a corner like a dusty grad student. 6: Finally and most importantly, stop pretending you still live in the world of happy academia where everyone is willing to follow the rules you set down just because you're the proffessor and they're the student. This just does not work anymore, you may scoff at people like Jim Flemming but for each one you knock down there is another one to learn from his mistakes and take his place. Do not pretend you can sit idle and call people who don't fall in line behind you names so that you can sit back in your dusty chair and pretend nothing is wrong. The internet as a whole is growing at an unthought of pace and your failure to keep up will not be fixed by being tight assed and making it harder on those that follow. Eventually someone else will take the forefront and throw you off your high horse like yesterdays newspaper. You purport to be leaders of the internet, then its about time you acted like it and start to solve the problems instead of trying to make the problems go away by being ignorant of reality. [-] Brett L. Hawn (blh @ nol dot net) [-] [-] Networks On-Line - Houston, Texas [-] [-] 713-467-7100 [-]

8 9

IP Ownership and Domain Names
by Jim Fleming 09 Mar '97

09 Mar '97

On Sunday, March 09, 1997 9:56 PM, Steven M. Bellovin[SMTP:smb@research.att.com] wrote: @ At 08:23 AM 3/10/97 +1000, Geoff Huston wrote: @ >Of course this scenario breaks down quickly.. @ > @ >Listed registrant A "sells" to B who "sells" to C who... @ > @ >Now the sucker who "buys" from C has the problem of tracing authenticity @ >of the "title" C is selling. This is not easy, particularly if B is less @ >than scrupulous and has "sold" title to a number of folk including @ >C, or if C is unscrupulous and is "selling" without have concluded @ >and transaction with B. @ @ This can easily be solved in a number of ways. For example, a public key @ could be publicly associated with each address. A transfer is accomplished @ by signing a message to that effect. Similarly -- and very importantly -- @ control over routing of that address is also governed by that @ private/public key pair. There's more to this protocol than I've @ described, but it can be made to work. @ @ You could also handle it just like "domain names".... Does AT&T "own"....ATT.COM ? Does AT&T "own".... 12.IN-ADDR.ARPA ? and 135.IN-ADDR.ARPA ? How about... 12.IP4.INT ? 135.IP4.INT ? are those registered....? FL 12.0.0.0 AT&T ITS (NET-ATT) FL 135.0.0.0 AT&T ITS (NET-ATT-135-0-0-0-B) -- Jim Fleming Unir Corporation e-mail: JimFleming(a)unety.net JimFleming(a)unety.s0.g0 (EDNS/IPv8)

1 0

Is the InterNIC really screwed in the head?
by Ehud Gavron 09 Mar '97

09 Mar '97

RFC954 specifies WHOIS. Client requests, server sends. Simple, really. $ whois opus4-hst .... Would you like to see registered users of this host? WHAT??? My whois client is supposed to search for this string and then do interactive input??? HuH?? $ whois pa aces (shows a handful of entries) There are 96 more matches.... would you like these displayed? EH? Anyone else think this is not only a gross violation of an RFC, does a disservice to millions of WHOIS-using users out there, and the final leg of the dissolution of the world? Ehud

3 3

RE: The Mother of all Solutions (Was Class B for Sale or Rent)
by Jim Fleming 09 Mar '97

09 Mar '97

On Sunday, March 09, 1997 3:56 PM, Randy Bush[SMTP:randy@psg.com] wrote: @ Emilio Bugatti (sp?), at the time the maker of the finest cars in the world, @ was asked why the brakes on his cars were not as good as they might be. He @ replied "Any fool can make a car stop. It takes a genius to make a car go." @ @ I suggest we focus on the latter. @ After the Top Level Domain debates, someone sent me private mail saying that if General Motors had invented and released the Internet, the public would currently be in every court in the land claiming that GM intentionally forgot to design in the brakes.... I see that you confirm that what some people see as a flaw, you view as a feature...Thanks...;-) P.S. Business people may look at this an conclude that there is a huge market for retrofit brakes... -- Jim Fleming Unir Corporation e-mail: JimFleming(a)unety.net JimFleming(a)unety.s0.g0 (EDNS/IPv8)

1 0

RE: Class "B" forsale (fwd)
by Jim Fleming 09 Mar '97

09 Mar '97

On Sunday, March 09, 1997 4:35 PM, Stephen Sprunk[SMTP:spsprunk@paranet.com] wrote: <snip> @ Did you sign any such contract when you got your IP addresses? Are there @ any laws in your jurisdiction stating the ownership and appropriate use of @ your addresses? @ People who are homesteading may not have "signed" anything. Other people may have signed papers and paid money. @ There is no workable analogy in this case because there are no contracts and @ no laws regarding anything on the Net at this point. Until ARIN makes you @ sign an acceptable-use agreement (and makes pre-1996 "owners" sign it too), @ there can be no enforceable policy other than what the core router owners @ decide. Routability determines address assignment far more definitively @ than a NIC board room full of cigar smoke and $10k fees. @ @ Stephen Sprunk How do you know...."there are no contracts and no laws"...? -- Jim Fleming Unir Corporation e-mail: JimFleming(a)unety.net JimFleming(a)unety.s0.g0 (EDNS/IPv8)

1 0