xanthian@saturn.ADS.COM (Metafont Consultant Account) (02/12/90)
Kent: I think that this discussion needs _much_ wider distribution! Kent: I hacked this out of several articles in the (Bay Area) Kent: newsgroup ba.news.config, changed to "name" style attribution Kent: to make things easier to follow (I hope!), (including my own Kent: stuff, such as this) reflowed the paragraphs to keep the lines Kent: on the screen, and added my uneducated comments. A few Kent: hyphenations that ended up on the same line probably slipped Kent: through; don't blame the original authors, it's just my flawed Kent: proofreading. I hope I got the attributions right. Be Kent: patient, it takes a long time to get to the really scary parts, Kent: but I thought the context was worth having. Followups are Kent: directed to news.groups. Can the folks who follow news.admin Kent: tell us if, and if so how soon, we can expect to see this Kent: enforced? Kent: In ba.news.config article <101264@pyramid.pyramid.com>, Kent: csg@pyramid.pyramid.com (Carl S. Gutekunst) wrote: Carl: In article <267@hplabs.HPL.HP.COM> rodrique@hplabs.HPL.HP.COM Carl: (Mike Rodriquez) writes: Mike: In preparing my previous map update I was struck by the Mike: question of whether one should list their (ex)-uucp neighbors Mike: with whom they now communicate primarily via SMTP. Carl: I assume you mean that you now communicate with them over the Carl: Internet? Carl: I have *very* strong feelings about this, but every time I Carl: bring it up, I get buried in a firestorm; since someone else is Carl: *asking*, maybe it'll be OK this time. :-) Carl: I feel *very* strongly that Internet connections should never Carl: be listed in UUCP map entries. The reason is simple: from the Carl: perspective of UUCP, you can reliably reach any Internet site Carl: from any other Internet site. If I am trying to reach, say, Carl: princeton.edu, and I'm using the UUCP maps for routing, and I Carl: am making any effort at all to pay attention, I will have a Carl: "top level domain" entry for .edu that points to a site that Carl: has agreed to handle all my Internet mail. Now, if hplabs has a Carl: link listed to "princeton," this will take precedence over my Carl: .edu entry, and actually worsen my connectivity. Similarly, if Carl: I have a direct UUCP link to princeton, the hplabs Internet Carl: connection may well be "lower cost," thereby rerouting mail I Carl: intended to go via UUCP over an Internet link, thereby putting Carl: me at an unsuspected risk of violating one of the Internet Carl: usage rules. Carl: The same goes for any fully connected network -- ENet, Bitnet, Carl: etc. Carl: I do recognize the notion of "fully connected network" is Carl: sometimes far less than ideal among the multiple internets. But Carl: generally speaking, the worst Internet path is usually much Carl: better than the best dialup path. Kent: In ba.news.config article <1990Feb8.012051.7798@vicom.com> Kent: lmb@vicom.com (Larry Blair) wrote: Larry: In article <101264@pyramid.pyramid.com> Larry: csg@pyramid.pyramid.com (Carl S. Gutekunst) writes: Carl: be "lower cost," thereby rerouting mail I intended to go via Carl: UUCP over an Internet link, thereby putting me at an Carl: unsuspected risk of violating one of the Internet usage rules. Larry: When we joined the net several years ago, we were told that Larry: uucp mail could not take a short cut through the Internet. Larry: Interestly, the MX system provides for exactly that. Being Larry: one hop off the Internet, and a registered domain, and with Larry: the approval of one of our four Internet neighbors, we began Larry: to use Erik Fair's mkglue script, which provides for automatic Larry: shortcutting. Larry: The questions are: Are there published rules? Has there ever Larry: been any action for this "abuse"? What about when send to Larry: foo.com and they are not an Internet site? Is that an abuse? Larry: How could I tell anyways? Kent: In ba.news.config article <4440@fernwood.MPK.CA.US> Kent: geoff@Fernwood.MPK.CA.US (Geoff Goodfellow) writes: Geoff: lmb@vicom.com (Larry Blair) writes: Larry: The questions are: Are there published rules? Geoff: Perhaps the following will be of help. I was present at the Geoff: meeting when they were approved. Geoff: Geoff Goodfellow Geoff: January 23, 1990 Geoff: FOR IMMEDIATE RELEASE Geoff: FOR MORE INFORMATION: Geoff: Richard Mandelbaum Geoff: (716) 275-2916 Geoff: San Diego, CA, January 9, 1990 - In a move towards the Geoff: establishment of a more coordinated national research and Geoff: education network environment, the Federation of American Geoff: Research Networks (FARNet) has adopted the first in a series Geoff: of guidelines, or FARNet Position Papers (FPP). The two Geoff: documents approved at the just-concluded San Diego conference Geoff: address the following: FPP Development and Approval Process Geoff: (FPP #1) and Guidelines on Acceptable Use and Connection (FPP Geoff: #2). Geoff: FARNet is an organization currently consisting of twenty-five Geoff: regional and state networks, who provide access from local Geoff: networks to the national research and education network Geoff: community (the Internet). The purpose of the Federation is the Geoff: advancement of science and education through the aiding of Geoff: communication among research and educational organizations. Geoff: The Federation endorses the coordination and interconnection Geoff: of regional and backbone networks to encourage the formation Geoff: of a unified network environment, thus providing enhanced Geoff: access to scientific and educational resources, both Geoff: nationally and internationally. Geoff: During the past three years, networks serving the needs of Geoff: research, education, and science have experienced explosive Geoff: growth. The growth has occurred at the campus, local, Geoff: regional, national, and international levels. Technical and Geoff: financial investments by both the public and private sectors Geoff: have been considerable. Utilization of these networks has Geoff: become essential to large segments of the American research Geoff: and academic communities, and continues to grow at a startling Geoff: rate, over 500% in the last 18 months! Guidelines for the Geoff: orderly development and interconnection of these varied Geoff: facilities are essential for the integrity of the networks and Geoff: continued provision of high quality services to educators, Geoff: researchers, scholars, and administrators. For this reason, Geoff: the FARNet Guidelines on Acceptable Use and Connection were Geoff: unanimously approved. Geoff: In summary, the Guidelines govern inter-regional traffic and Geoff: recommend that traffic between the FARnet-Member networks be Geoff: restricted to research or academic purposes, or to direct Geoff: administrative support of such efforts. (Intra-regional Geoff: traffic is governed by the guidelines set by each regional.) Geoff: The position was adopted because the networks represented by Geoff: the members of FARNet are, in many instances, at least Geoff: partially funded by grants from state or federal agencies. Geoff: Activities that are beyond the scope of research or academia Geoff: are not considered acceptable. For example, Richard Geoff: Mandelbaum, FARNet's Chairperson, summarizes from the Geoff: Guidelines, "It is not acceptable to send invoices between two Geoff: commercial entities on different regional networks across a Geoff: national backbone." Geoff: Future FARNet Position Papers are to include such issues as Geoff: network design and engineering, international interaction, Geoff: commercialization of services, network management models, Geoff: value-added services, and methods of more accurately Geoff: addressing the information movement needs of researchers, Geoff: scholars and educators. (For further information, contact Geoff: Richard Mandelbaum (716) 275-2916 or rma@cs.rochester.edu) Geoff: FARnet Position Paper #2: Geoff: FARNET GUIDELINES ON ACCEPTABLE USE AND CONNECTION Geoff: 1.0 Introduction Geoff: During the past three years national regional and local Geoff: networks have experienced exponential growth. The technical Geoff: and financial commitments made by the private and public Geoff: sectors have been varied and considerable. Use of these Geoff: networks is now considered essential by large segments of the Geoff: American research and academic communities. Geoff: Mechanisms for management have been ad hoc and inconsistent. Geoff: Currently there are no published guidelines nor an associated Geoff: method of adjudication addressing the use of network Geoff: resources. Furthermore, inconsistencies exist among regionals Geoff: about what is considered acceptable use of national networks. Geoff: Without effective management of the use of the network, there Geoff: exists potential for severe economic and political problems. Geoff: Regional networks and the national backbones receive a Geoff: considerable amount of federal funding. This subsidy requires Geoff: accountability, a means to demonstrate that the federal funds Geoff: are being properly applied. Given the strategic importance Geoff: that the networks have assumed for national research and Geoff: development, it is vital that the integrity of the resource be Geoff: maintained. Geoff: 2.0 Intent Geoff: The intent of this document is to suggest policies and Geoff: mechanisms for determining appropriate use of and connection Geoff: to networking resources. The networking environment model is Geoff: assumed to be a three-tiered hierarchy consisting of a set of Geoff: national backbone nets (such as NSFnet and NSN), campus and Geoff: corporate networks (such as a campus-wide university network Geoff: or a corporate site LAN) and, connecting these components, Geoff: mid-level networks that offer sites in states or geographic Geoff: regions access to national nets. It should be noted that Geoff: mid-level networks may in turn be made up of several layers of Geoff: state and regional networks. Geoff: This document specifically addresses traffic that is exchanged Geoff: among mid-level networks that are members of FARnet, whether Geoff: across a national backbone or on a publicly subsidized direct Geoff: regional connection. It does not preclude additional Geoff: requirements that a national backbone might establish. This Geoff: document may also serve as a basis for acceptable use policies Geoff: within a mid-level network. Geoff: 3.0 Definition of Terms Geoff: Appropriate use refers to whether the use of the network is Geoff: consistent with the guidelines for each network that the Geoff: traffic traverses. This applies both to standard applications Geoff: (e.g., electronic mail, file transfers, and remote login) and Geoff: nonstandard uses (chat, experimental protocols, etc) Geoff: Acceptable connection refers to the specific authority and Geoff: terms by which a user accesses the network. Issues that are Geoff: addressed here include restrictions on access (for security Geoff: purposes), resale of connectivity, etc. Acceptable use and Geoff: acceptable connection, while related, are separate issues. It Geoff: is possible for acceptable connections to be used for Geoff: unacceptable use, and for acceptable use to be performed on an Geoff: unacceptable connection. Geoff: 4.0 Acceptable Use Policy Geoff: Given both the volatile nature of the technology employed and Geoff: the demand that users make of the network, determining Geoff: acceptable use is a dynamic and iterative process. In Geoff: evaluating whether a particular use of the network is Geoff: appropriate, several factors should be considered: Geoff: Traffic between mid-levels should be restricted to research Geoff: or academic purposes, or to direct administrative support Geoff: of such efforts. Organizations whose connection to the Geoff: internet is sponsored by a FRICC agency can use the network Geoff: in support of the sponsored activities. Traffic whose Geoff: content is solely commercial is not acceptable. Malicious Geoff: use is not acceptable. Use should be consistent with Geoff: guiding ethical statements and accepted community Geoff: standards. Use of the internet in a manner that precludes Geoff: or significantly hampers the use by others should not be Geoff: allowed. Geoff: Each mid-level network should establish a regional acceptable Geoff: use policy that permits, at a minimum, the transit of any Geoff: traffic that is acceptable to an attached national backbone. Geoff: Mid-level networks may establish additional requirements as Geoff: are appropriate to the regional mission. Geoff: FARnet recommends that each regional accept traffic from other Geoff: regionals if the use was determined to be acceptable under Geoff: these guidelines by the originating network. Geoff: Decisions made by mid-level networks or backbone providers Geoff: regarding specific instances of acceptable and unacceptable Geoff: use should be widely circulated to encourage consistency. Geoff: FARnet can and will act as a vehicle for the distribution and Geoff: maintenance of such information. Each mid-level network Geoff: should designate an individual to participate in the exchange Geoff: of this information. Geoff: 5.0 Acceptable connection Geoff: Mid-level networks should insure that the connections made to Geoff: them are consistent with the effective use and protection of a Geoff: shared resource. The mid-levels should know what networks are Geoff: connected and what use is being made of the network. Geoff: Mid-level networks should instruct members on current Geoff: guidelines for acceptable use. Access to the internet should Geoff: be protected through the use of prudent security measures. Geoff: Unauthorized connections to the internet should not be Geoff: permitted. "Third party" connections (such as internet access Geoff: being provided by research parks or through resale by a Geoff: mid-level subscriber) should be done only with the approval of Geoff: the mid-level networks. Connections which create routing Geoff: patterns that are inconsistent with the effective and shared Geoff: use of the network should not be established. Geoff: 6.0 Adjudication Geoff: Mid-level networks should distribute this statement to member Geoff: institutions and request members to inform their communities Geoff: about these issues. Geoff: Responsibility for the determination of whether a proposed use Geoff: of the network is acceptable begins with the initiating user. Geoff: If the user is uncertain, the associated connecting authority Geoff: or mid-level should be contacted. Geoff: Mid-level networks should consult with backbone providers and Geoff: FARnet as needed to determine if an intended use of a backbone Geoff: is consistent with the policies of the provider. The results Geoff: of these deliberations should be distributed among the Geoff: mid-level networks to encourage consistent policy. FARnet Geoff: should be active in implementing this process. Geoff: If disagreements arise among mid-level networks concerning Geoff: their direct connections, FARnet should attempt to act as a Geoff: reconciliatory agent. Geoff: 7.0 Enforcement Geoff: In instances where particular traffic is determined to be an Geoff: abuse, the mid-level network that originated the traffic will Geoff: be held responsible for both admonishing the perpetrator and Geoff: preventing further abuse. It is assumed that the mid-level Geoff: network will, in turn, place similar responsibilities upon its Geoff: members. Geoff: Mid-level networks should make a good faith effort to enforce Geoff: the decisions that emerge from the adjudication process Geoff: undertaken by FARnet. Kent: In ba.news.config article <1990Feb11.225848.23276@vicom.com> Kent: lmb@vicom.com (Larry Blair) writes: Kent: [Some previously included stuff omitted; there was quite a bit Kent: of it] Larry: Are these in any way binding? If I read these correctly, most Larry: of the .COM domain does not qualify. The MX system for uucp Larry: .COM sites would have to be eliminated or severly restricted. Larry: The Internet has evolved over the years. While it may have Larry: started as a DARPA project to link universities and defense Larry: contractors, it now provides the communication backbone for Larry: hundreds of thousands of machines throughout the world, most Larry: of which have nothing to do with research, education, or Larry: defense work. If I'm reading the guidelines correctly, the Larry: majority of the "trash" on netnews would not qualify. One Larry: .COM could not talk to another, even if both are directly on Larry: the Internet, unless the communication satisfied a very narrow Larry: range of allowable topics. Larry: Does FARNet really run the second level interconnect? Could Larry: this spell a return to the USENET of several years ago, with Larry: two day delivery times? Are they going to kick all of the Larry: .COM sites off the Internet? Or just resrict them to their Larry: regional net? Kent: I don't think a single one of the many (>40) newsgroups I read Kent: from USENet could be copied across to the Internet under the Kent: listed rules. In fact, most traffic of _all_ groups I've ever Kent: read (used to be more than 130) would probably not qualify. Kent: Even in such staid groups as sci.math, most of the postings Kent: would be lucky to qualify as "recreational" in nature, vice Kent: "arrant nonsense" for the plurality of the rest. The essence Kent: of communication pathways normal humans are willing to tolerate Kent: includes concepts such as "fun", "anger", "errors", "insults", Kent: "blundering around lost", and "misunderstandings". If "you" Kent: try to force the signal to noise ratio too high (> 10%, on a Kent: guess) most of the signal will go away in search of a more Kent: forgiving communications channel. Kent: I'd guess the posted rules were not written by people who are Kent: daily _contributing_ participants in any widely distributed Kent: net, or they'd have better sense. -- Again, my opinions, not the account furnishers'. xanthian@well.sf.ca.us (Kent Paul Dolan) xanthian@ads.com - expiring soon; please use Well address for replies. Kent, the (bionic) man from xanth, now available as a build-a-xanthian kit at better toy stores near you. Warning - some parts proven fragile. Just another pair of enemployed graphics programmer hands doing the Devil's work in frustration from enforced idleness. -> METAFONT, TeX, graphics programming done on spec -- (415) 964-4486 <-
csg@pyramid.pyramid.com (Carl S. Gutekunst) (02/13/90)
There is nothing new here. It has long been known by Usenet/Internet old hands that USENET crosses over the line of what types of traffic are supposed to be passed over the research Internets. The same was and is true of many of the Internet mailing lists, many of which now use Netnews and NNTP as a transport protocol. Back when NNTP was brand new (three, four years ago?), there was a lot of concern about this. It was generally accepted that some day, in the indefinite future, USENET would lose its privilege of running across the research Inter- nets, perhaps very soon. Not only hasn't that happened, but if anything USENET is more accepted now than then. There was a lot of concern that DARPA, CSNet, and other "officials" allowed USENET traffic simply because they didn't know what it was. When we were considering joining CSNet, I asked them about USENET traffic. They understood exactly what I meant, and had no objections at all. For the paranoid types, the solution is commercial internets, instead of re- search. Nysernet is offering a commercial IP network around New York, and the UUNet AlterNet service is now coming on line. Expect others in the future. [Followups directed to news.admin, where they belong. News.groups is not read by news administrators.] <csg>