- Test - lists.nanog.org

Sprint filters (yes really)
by Sean Doran 20 Jun '96

20 Jun '96

-----BEGIN PGP SIGNED MESSAGE----- Please note that the changes to Sprintlink's filtering policies that Marc Hidalgo and I announced at NANOG have been implemented, and should be visible as BGP sessions with our external peers (MCI, ANS, Alternet, PSI et al.) reset over time. It is hoped that the changes won't be flawed, but any oddities would be nice to know about. The two and a half week delay in implementation is my own fault. Note that the policy is now: [allow none of the RFC 1597 et succ. reserved addresses] in 0/8-126/8, allow no subnets of historical As that were not announced in July 1995[*] 127/8-191/8, allow nothing longer than /16 [*] 192/8-205/8, allow nothing longer than /24 206/8-223/8, allow nothing longer than /19 195/8 [RIPE block, 12 dec 95], allow nothing longer than /19 [other RIPE and APNIC blocks that have begun being allocated from since July 1995 will also be filtered on the /19 boundary] The principal change is that we used to disallow /19s in 207/8-223/8 and 195/8, and this was not in line with the minimal allocation units performed by the three primary registries. Please also note that we have also already accumulated a couple of months of successful experience with differential filtering, and will be both tightening up the filtering in the newly unfiltered space and tweaking the currently-used values up and down. Very roughly speaking, the goal is that if you have a very long prefix (like a /24), and it flaps a couple times, you can go home for the day. If you do even the slightest bit of aggregation (into a /23, for example), then you are much less likely to be dampened in the face of minor flapping, and you will be dampened for less time. Once you have aggregated enough addresses to comprise a /18 or shorter prefix, you will be dampened with values very much in line with the default you get from Cisco's "bgp dampening" command. At the moment, you are much more likely to be dampened if you are NOT a Sprint customer. That is, the normal values for "bgp dampening" are in place if you have a customer connection to SprintLink or ICM. However we will be evaluating the engineering and operational effects that a similar policy with respect to SprintLink customers, as an effort to encourage some of our lazier customers to begin doing their own aggregation wherever possible. You may not be surprised that the success of this differential dampening experimentation has played a large role in the easing-up of our filtering policies. Sean. -----BEGIN PGP SIGNATURE----- Version: 2.6.2 Comment: PGP Public Key in ftp://ftp.sprintlink.net/engineer/smd/pgpkey iQCVAwUBMcmkVkSWYarrFs6xAQEblQQAg60neKgLSGpQZ2e0n67MRjVNd+eLDSaJ Ceea+2bypM8V9QD6nTr6zEJ302qLrIK6b9e0lLmgFPpoUs1Rb3KDqv4ruAQKmkBo dDu5NxIAlG9b+qvXTuocPA81y4+zWKMOaFHCPFNksMIzhA19pRYLndiuJqRifiHn CQyAZGorrwQ= =uyTH -----END PGP SIGNATURE-----

1 0

Re: A fourth lawsuit against NSI by a domain name owner to try to keep from losing a domain name
by Carl Oppedahl 20 Jun '96

20 Jun '96

At 06:07 PM 06/19/96 -0700, Gene Ingram wrote: >Carl Oppedahl wrote: >> There is now a fourth case in which NSI wrote to a domain name owner >> saying that their domain name would get cut off in 30 days, and in >> which the domain name owner says it wasn't infringing anybody's >> trademarks (which is of course no defense under the present NSI >> policy), and finds itself suing NSI to avert loss of the domain name. [roadrunner.com] [dci.com] [ty.com] [clue.com] >Based on the fourth case, I'd say we are in serious trouble. >The word "clue" is a word in a dictionary, and the company is >entitled to call their firm "Clue" on a local level provided >they have a business license and have filed a fictitious name >statement. In my area we also need to renew our fictitious name >with some government agency (can't remember offhand which one) >downtown, every 5-7 years. Provided you do that, and your >business is industry-specific (does not violate trademark laws) >then there is nothing anyone can do to stop you. Well, of course you are right that "clue" is in a dictionary. It's not difficult to imagine that someone could pick the name "Clue" for their business for perfectly innocent reasons. But then the word "road runner" is in the dictionary too, and is also the state bird of New Mexico. What can be said about "dci"? Probably what can be said is that it is three initials, and there are probably hundreds or thousands of companies with those initials. And "ty"? It is a person's name. Short for Tyrone, I suppose. It seems to me that in each of these cases it is not difficult to imagine that the name could have been chosen for perfectly innocent reasons. >What if the company wins and gets "clue.com" but then >someone in an entirely *different* industry with a trademark >for use in *their industry* under the name "clue," comes >along and shows they got their trademark first? Let's say >the company that is suing Clue wins because they got a >trademark in their respective industry in say, 1970. Now >let's say that someone who *also* owns the trademark "clue" >in an entirely separate industry, comes along and sues for >``clue.com'' because their trademark was obtained in 1969, >one year earlier? Will NSI then suspend the company's >domain ``clue.com'' (the company that won the right to >``clue.com'' through litigation), pending outcome of the >third company who says their trademark in a separate >industry (let's say, it's ``Clue Lingerie'' and they >registered the Clue trademark for women's lingerie in >1969, while the game manufacturer registered their Clue >trademark for games in 1970). > >Now is someone going to tell me that the process will start >all over again in court, with the lingerie company perhaps >being awarded the ``clue.com'' domain that was won with >costly litigation by the game company? Where does this >legal madness end? Keep in mind that it's not just US trademarks, but also trademarks of countries outside the US, that under present NSI policy permit taking away a domain name from somebody. I just did a quick check and found more than half a dozen "clue" trademark registrations in Europe. Each of these European trademark owners could, one after the other, bring an NSI challenge against whoever happened to own "clue.com". Of course, by definition all but one of them will have to somehow find a way to go on in life without possessing "clue.com". Notwithstanding that, each of them could initiate a challenge according to NSI policy. In the US alone there are over a dozen "roadrunner" and "road runner" trademark registrations. In the US alone there are almost two dozen "dci" trademark registrations. In the US alone there are about half a dozen "ty" trademark registrations. Let's look at SAIC, the parent company of NSI. As you may know, their domain name is saic.com. Well, there is some company in Italy with a trademark registration of "saic". They make meat, I seem to recall. This Italian company could simply spend the cost of a postage stamp and send a copy of their trademark to NSI, and NSI would have to commence a dispute proceeding with its own parent company. If SAIC failed to provide the precious trademark certificate within 30 days, NSI would have to cut off the domain name of its own parent company. Or if SAIC did manage to provide a trademark certificate within the required 30 days, NSI would have to ask SAIC to sign the well-known "indemnification agreement" and perhaps ask it to post a bond. If SAIC failed to sign the indemnification agreement within 14 days, then NSI would have to cut off the domain name of its own parent company. Of course, SAIC might feel that it was unreasonable to have to sign the "indemnification agreement", in which case SAIC would presumably have to sue NSI for a court order to restrain NSI from cutting off the saic.com domain name. This all assumes that NSI would carry out its policy with respect to its parent company in the same way that it carries out the policy in other cases that are publicly known. Since NSI conducts its proceedings in secret, one would have no way to check if this were so. --- Carl Oppedahl, oppedahl(a)patents.com Oppedahl & Larson, patent law firm http://www.patents.com/ is a web server with frequently asked questions and answers on patent law and other intellectual property subjects

3 2

Are you all aware of Juno Mail? (http://www.juno.com/)
by Steve Goldstein 20 Jun '96

20 Jun '96

Re a recent remark about support of network communications ventures by advertising, Juno is supported by advertisements only. There is another similar service in the launch phase: http://www.FreeMark.COM:80/256color/index.html Both are free to the end-user, modulo extra connect time to download the unobtrusive advertisements that appear in the upper right portion of the Juno mailreader screen (I do not know about FreeMark). Juno supports only Internet-style e-mail, and only via dial-up (including 1-800 numbers for those subscribers not yet served by local toll-free numbers). No web, FTP, chat, etc. But, it is a great adjunct for people who want to keep personal mail separate from the official mail that they do on their employers' systems. Or for folks who just want e-mail. A recent message to subscribers notes that Juno is doing its best to keep up with burgeoning demand. Regards, Steve G. From: charlesa(a)juno.com (Charles Ardai) Reply-To: feedback(a)juno.com Subject: Service Update Message-ID: <update.1996.06.19(a)juno.com> Dear Juno Member, It has been more than a month since we last sent a general update to all members about the status of our launch and almost two months since the launch itself; in light of exceptional recent growth (and some related growing pains) we thought a note was in order. Juno currently has more than 80,000 accounts, and we're getting new requests for copies of our software as fast as we can send disks out. On average, we're seeing at least one new account created every minute, and sometimes see several per minute during our busiest hours. Needless to say, this overwhelming interest in Juno (and the overwhelmingly positive feedback we've received from you about the service) is very gratifying. It's also educational; we're still shaking the last few bugs out of the system and are learning what we need to know to ensure that our telephone and computer networks have adequate capacity to handle the growing demand they need to support. It is critically important to us that our members find Juno a reliable and convenient service, one unhampered by the sorts of problems (such as busy signals when you try to connect or disconnections in the middle of a transmission) that have sometimes plagued other online services during periods of rapid expansion. I am glad to say that our network has performed very well generally and that most of you have not reported any difficulty using the service. However, we know that some of you have had difficulties connecting to Juno over the past few weeks, especially during our busiest hours. Please accept our apologies for any inconvenience, and rest assured that we are taking steps to increase our capacity (that is, the number of computers and telephone lines available for our members to dial into) and to eliminate the current strain on the network by adding the other hardware we need. For the short term, if you do experience difficulty connecting to Juno, please try connecting before or after peak hours (9:00PM to midnight, Eastern Time). You need not be concerned that there is any problem with your account; none of your mail will be lost. We hope to alleviate the current overload within a week or two. In the meantime, many thanks for your patience and for the helpful feedback and comments many of you have passed along. Growing pains are probably inevitable when a service grows at Juno's pace, but by working together we can make sure that they are short-lived. Sincerely, Charles Ardai President Juno Online Services, L.P.

1 0

Re: critique of NANOG meeting
by Scott Bradner 20 Jun '96

20 Jun '96

> Actually i don't understand what his motivation is. I think the answer to that is simple - he wants the net to be seen as working to the users and fears that is not now the case (in some places) and that the "not the case" will spread before it gets better. his is a case where I think he is doing what he thinks is the right thing (not always the case with some of the commenters about the net :-) ) Scott

2 1

Re: critique of NANOG meeting
by Per Gregers Bilse 20 Jun '96

20 Jun '96

It's straightforward, really: Bob's got a column to do. He's merely doing his job, which is to produce snappy, juicy tidbits, gossip, biting remarks, whatever else makes readers happy. Just like some journalists do film reviews, TV reviews, etc, Bob is doing "Internet reviews", and I have no doubt that he receives vast amounts of supportive mail. It's a psychological thing, Joe Public liking to see somebody telling somebody some "truths" on their behalf -- "Yes, you tell them Bob! Good work!" -- and is merely a result of the Internet becoming part of Joe Public's life. It can't be any other way, and if it wasn't Bob doing it, it would be somebody else. And there isn't really any reason to be too concerned about it; it's an achievement of sorts to have people spending time criticising you. Some people believe that Bob's past somehow should influence what he's doing now. It doesn't, it won't, and it never will, because he's not doing the things now that he did many years ago. He's doing something completely different. Bob is actually quite funny at times, I liked the thing about wearing a tie. Can we get somebody to send a reminder on the IETF announcement list, for people to bring ties and testboot their laptops before turning up in Montreal? -- ------ ___ --- Per G. Bilse, Mgr Network Operations Ctr ----- / / / __ ___ _/_ ---- EUnet Communications Services B.V. ---- /--- / / / / /__/ / ----- Singel 540, 1017 AZ Amsterdam, NL --- /___ /__/ / / /__ / ------ tel: +31 20 6233803, fax: +31 20 6224657 --- ------- 24hr emergency number: +31 20 421 0865 --- Connecting Europe since 1982 --- http://www.EU.net e-mail: bilse(a)EU.net

2 1

Re: NSI Domain Name Dispute Policy Statement
by Carl Oppedahl 20 Jun '96

20 Jun '96

At 08:18 AM 06/20/96 -0500, E. Gregory Higgins wrote: >Let me see if I understand the current intrepretation of >paragraphs 6(c)(3) and 6(c)(4) of the NSI Domain Name Dispute >Policy Statement: > >I can be using a domain name for years, but as long as someone somewhere >in the world receives a trademark registration which claims a first use date >prior to the day I first started using my domain, InterNIC will give >them preference for my domain name and will forceably interrupt >service to my self and others dependent upon the existance of my domain. > >Is this essentially correct? Yes, that's essentially right. See http://www.patents.com/nylj6.sht and http://www.patents.com/pubs/nw1.sht . And indeed even if you were able to prove to NSI that you weren't infringing that trademark, it would be no defense. Your domain name gets cut off in 30 days. But then read the policy even more closely, and you will see that actually the dates don't matter. Even if the trademark (from anywhere in the world) is *newer* than your domain name, you still lose the domain name. Interestingly, the time period is shorter: 14 days. Here's how it works, according to the NSI policy. The trademark owner (TMO) sends the trademark certificate to NSI. Then NSI sends you an "indemnification agreement" (see <http://www.patents.com/fogbelt/ia.sht>) to sign, stating that if you don't sign it within 14 days NSI will cut off your domain name. You show it to your lawyer, your lawyer says "you can't sign this!". 14 days later, your service (and that of others dependent on the domain name) is forceably interrupted. --- Carl Oppedahl, oppedahl(a)patents.com Oppedahl & Larson, patent law firm http://www.patents.com/ is a web server with frequently asked questions and answers on patent law and other intellectual property subjects

1 0

Re: critique of NANOG meeting
by Scott Bradner 20 Jun '96

20 Jun '96

> Let's get real while I agree that the solution suggested is tried & untrue I was responding to the specific question of what is on his mind and I think he is being honest. Scott

1 0

Re: Address allocation stats
by Havard.Eidnes＠runit.sintef.no 20 Jun '96

20 Jun '96

> I've just updated the graphs for the data through the end of May. > These'll be the ones I'll be bringing to Montreal and are very > similar to the ones from LA -- assuming no other changes. > > The postscript files can also be copied from: > ftp://research.ftp.com/pub/cidr/current > > If I get a chance this week, I'll also look for a .ps->.gif converter.. You've just found one. Me. ;-) Seriously, you need GhostScript 2.6.1 with the following modification to gdevgif.c: /* #define TABLE_SIZE 5123 /* this is max for 12-bit codes */ #define TABLE_SIZE 5119 /* this is max for 12-bit codes (and prime) */ 5123 is nice, but not prime, and this will in certain circumstances produce an infinite CPU-consuming loop. Second, uncomment the "30000 VM?" test in the PS files. Third you will need the ps-to-gif script (a one-liner in the shell given the right gs) and the landscape.ps which is available together with the already converted (gif) files in ftp://trane.uninett.no/tmp/pier/ Note that a newer version of GhostScript won't do, since they removed the GIF driver in subsequent versions to 2.6.1 (you probably know why...). Regards, - Havard

3 2

Netcom Issues
by Alan Hannan 20 Jun '96

20 Jun '96

Hi. I'm seeing alot of problems with Netcom, and I can't seem to get any info from them, the net, or the normal grapevine. Anyone have any info to share? -alan

3 2

Re: critique of NANOG meeting
by avg＠postman.ncube.com 19 Jun '96

19 Jun '96

Jeff Young wrote: >metcalfe understands the issues only too well. only the truly >blessed can blindly exploit half-truths. he is not blessed and >he is not blind. >i have no doubt that his campaign will succeed. he is bright, has >built companies, used to be a technically saavy engineer. the trick >will be to bide our time, he will go away in the end. hopefully his >winnings will be modest this time. You made him to look cunning and evil. I'd like to make him to look ridiculous. That's harmless at least and there's a chance that this will make him to go away. Actually i don't understand what his motivation is. It appears that it is not material; and any decent person wouldn't just go and use somebody else's problems as a ladder to do social climbing. --vadim

3 2