LC comments on draft-cotton-rfc4020bis-01.txt
I think the procedures proposed in this draft could be simplified a bit if we'd just focus on the actual issue, viz., that the implementation and deployment of a new specification is completely decoupled from the progress of that specification through the standards process. Once any significant deployment has begun, the codepoints used by that deployment become de facto unavailable for future use. The only real issue is whether we want that codepoint usage to be recorded in a public place, or whether we want to pretend officially that the codepoints aren't used, and just let the conflicts arise in the field. >From this perspective, I think the following procedures are problematic: - Section 2, point d: early allocation is conditioned upon whether there is sufficient interest in early (pre-RFC) implementation and deployment in the community as judged by working group chairs or ADs. What the WG chairs and ADs really have to judge is whether failure to issue an early allocation is likely to result in de facto allocation of a codepoint, with consequent conflicts in the future. Of course, there have also been many cases where the codepoint is already in significant deployment before any "official" request for it has been made; WG chairs and ADs should take notice of this fact. - Section 3.1, step 6: "IANA makes an allocation from the appropriate registry, marking it as 'Temporary', valid for a period of one year from the date of allocation. The date of first allocation the date of expiry should also be recorded in the registry and made visible to the public." What is the point of marking it as "temporary"? Once the codepoint is placed into use, it is not any more or less temporary than any other codepoint; the codepoint is unavailable for future reuse as long as the deployments. Any codepoint that is no longer in use can of course be reused, even if its allocation is not marked "temporary". I do not understand the idea of making the allocation "expire" after just one year. Are the deployments going to disappear after one year? If not, then the codepoint allocation will not expire after one year. - Section 3.3: "If early allocations expire before the document progresses to the point where IANA normally makes allocations, the authors and WG chairs may repeat the process in section 3.1 to request renewal of the code points. At most, one renewal request may be made; thus, authors should choose carefully when the original request is to be made." First, it is not up to the authors to "choose carefully" when the original request is to be made. At a certain point in the implementation process, the codepoint is needed. Failure to get the codepoint soon enough will just cause the implementors to make up their own codepoint, which will invariably leak out into deployment. Second, there is no reason whatsoever to put a two-year limit on the early allocation, unless one expects that the deployments using it will magically disappear after two years. I've seen a number of IETF standardization efforts lag six or seven years beyond the actual deployment of the standard. It might be worthwhile for the WG chairs to ask every few years whether the proposed use of a codepoint has been abandoned, but without some assurance that the codepoint will never be used as specified, there is no sense declaring it to be expired. - Section 3.1: "Note that Internet-Drafts should not include a specific value of a code point until this value has been formally allocated by IANA." To me, this seems entirely counter-productive. Failure to publicize the codepoint you're deploying doesn't solve or prevent any problems. To the contrary, including the codepoint values you're using helps to find conflicts. - Section 5: "There is a significant concern that the procedures in this document could be used as an end-run on the IETF process to achieve code point allocation when an RFC will not be published." This concern only makes sense when the codepoint is being requested by another SDO. If it's being requested by a vendor (or set of vendors), well, no vendor will tell it's customers "sorry, you can't have this feature because IETF hasn't allocated a codepoint". The real concern is the opposite -- folks will try to delay the allocation of a codepoint, in order to prevent their quicker competitors from gaining an advantage. But this only leads to the use of codepoints that are not officially allocated. My concrete suggestions for improving the draft would be: - Emphasize that WG chairs and ADs should base their approval of a request on the likelihood that an unofficial codepoint will otherwise get allocated (or the fact that it is already in use). - Eliminate the "temporary" and "automatic expiry" features, aldn eliminate the need to refresh a request after one year. - Eliminate the requirement that an i-d not
Re: RFC 2119 terms, ALL CAPS vs lower case
> So, I recommend an errata to RFC 2119: "These words MUST NOT appear in a > document in lower case." I'm glad you said "I recommend" instead of "I have recommended", as the latter would violate the recommended (oh dear) rule. This RECOMMENDED rule would also imply that documents can no longer be published during the month of May, as otherwise the date line would put the document out of compliance with this erratum. Also, it would no longer be possible for the references to list any document that was published in the month of May. I suppose we could make it a rule that the month of May be referred to as "the month between April and June". Of course, you didn't say that the reserved words muSt nOt appear in mixed case, so mAybe that will become a workaround. > Seems to me that precision of meaning overrides graceful use of the > language. Many IETF specs lack the desirable degree of precision, but the use of the "reserved words" in capitalized or uncapitalized form just is not that big a contributor to the imprecision. Failure to properly specify all the actions in all the possible state/event combinations is a much bigger source of confusion than failure to capitalize "must" (I mean failure to capitalize "MUST"). I don't know why so much attention is being given to something that is NOT going to improve the quality of the specs by any appreciable amount.
Re: Last Call: (IPv6Support Required for all IP-capable nodes) to Proposed Standard
This document really wants to be a BCP that makes deployment and strategy recommendations. But for some reason it has been disguised as a standards track document (i.e., as a protocol specification), and the result is that RFC 2119 language is being used in a very peculiar way. I think this is what is responsible for the impression that the document is "bizarre". Some examples: a best effort SHOULD be made to update existing hardware and software to enable IPv6 support. RFC 2119 language is used to distinguish optional from mandatory features in an implementation. But "a best effort ... to update existing hardware and software" does not seem to be a feature of an implementation. I just don't understand what this statement requires, or how one would tell if a given implementation is compliant with it or not. Current IP implementations SHOULD support IPv6. Presumably, "current" implementations support whatever they support, I don't really understand what is being required here. New IP implementations MUST support IPv6. Is there some objective difference between a "new" implementation and a "current" implementation? How many lines of code have to change before a "current" implementation becomes a "new" one? But I don't really see why "new" vs. "current" is even relevant. If there were a lot of folks writing IP implementations from scratch, using only the RFCs for guidance, it would be really important to make sure they know about IPv6. Does anyone think that that's a real problem? It may be a problem that new products continue to come out without IPv6 support, but in general, those new products use current implementations. So again, there doesn't actually seem to be a useful requirement expressed. IPv6 support MUST be equivalent or better in quality and functionality when compared to IPv4 support in an IP implementation. So if the v6 support has more bugs than the v4 support (and hence lesser quality), the implementation would be out of compliance with standards? I don't think IETF standards set quality metrics on implementations. As for functionality, consider the following bit of functionality: the ability to communicate with an IPv4-only host. This is a piece of functionality that v4 support has but v6 support doesn't. So I guess no implementation could ever meet the "requirements" of this document. Finally: MUST NOT require IPv4 for proper and complete function. Suppose the "proper and complete" function of a box requires the downloading of updates from a server the box cannot necessarily reach using IPv6. Then IPv4 would be required for proper and complete function of the box. Or perhaps the box is a network monitoring appliance of some sort, and has to use both IPv4 and IPv6 to do its job. In this case too, v4 is required for "proper and complete" function, but that shouldn't lead anyone to say that the box is non-compliant with IETF standards. This document just does not say what it means. The RFC 2119 language is not being used in the usual manner, it's really being used for its rhetorical value. This is not appropriate. ___ Ietf mailing list Ietf@ietf.org https://www.ietf.org/mailman/listinfo/ietf
Re: Use of "unassigned" in IANA registries
Phillip> But I rather suspect that the reason that this is happening is that Phillip> people know full well that there is a process and choose to ignore Phillip> it because they either can't be bothered to put up with the hassle Phillip> or don't think that the application will be accepted. Lars> Suspect all you want, but it doesn't match my experience. Phillip's suspicion certainly matches my experience over the past 15 years, and I've even done my own share of codepoint squatting. He is also correct to state that many folks try to use control over the codepoint space as part of their competitive marketing strategy. The only way to avoid collisions due to "squatting" is to adopt a policy that all codepoint fields be large enough so that a significant number of codepoints are available for FCFS allocations. ___ Ietf mailing list Ietf@ietf.org https://www.ietf.org/mailman/listinfo/ietf
Re: Old transport-layer protocols to Historic?
> RDP is still in use > was the initial transport used for gateway control (such as SGCP) until > SCTP was developed. > Many commercial gateways still support these older pre-standard (to > MEGACO) control protocols. Some older devices still provisioned in the > network only support the older protocols. > but we had nothing before SCTP that would fit the bill. Its limitations > was one of the driving forces behind developing SCTP. So RDP is a useful and deployed pre-standards protocol that was one of the driving forces behind a successful standardization effort. But newer applications that need that kind of transport instead use the standard, SCTP. I don't know how one could possibly make a stronger case for classifying the RDP spec as Historic! However, I do agree with John Klensin's remark that reclassification is not worth the energy it takes. ___ Ietf mailing list Ietf@ietf.org https://www.ietf.org/mailman/listinfo/ietf
Re: Gen-ART review of draft-ietf-l3vpn-mvpn-spmsi-joins-01
James> perhaps this needs to be stated (that the Type 4 is created by this James> doc for your purpose)? I think the doc already makes this clear, maybe I'm not sure what you are asking. James> You can probably imagine how many SIP and RSVP protocol extensions James> there are (70+ and about 20 respectively off the top of my head), and James> yet every one of them have to state "Session Initiation Protocol James> (SIP)" and "ReSource ReserVation Protocol (version-1) (RSVPv1)" the James> first time they appear, no matter how big the community of interest James> is. And this makes sense to you? Okay, I will expand the occurrence of "S-PMSI" in the abstract. On the issue of the maximum UDP packet size, I think that is an implementation issue and I don't think it is appropriate to raise it in this document. ___ Ietf mailing list Ietf@ietf.org https://www.ietf.org/mailman/listinfo/ietf
Re: Gen-ART review of draft-ietf-l3vpn-mvpn-spmsi-joins-01
> Minor issues: > - Section 3, 2nd para, second sentence is: > > "A Type 4 S-PMSI Join may be used to assign a customer >IPv6 (C-S,C-G) flow to a P-tunnel that is created by >PIM/IPv4. " > I'm curious how else might a Type 4 S-PMSI be used? This sentence > makes it seem unclear as to whether there are any uses a Type 4. While there is no other use of the Type 4 S-PMSI Join, there are other mechanisms that may be used for the same purpose. If the text read "is used", someone might interpret that as meaning that there is no other mechanism for assigning a v6 flow to a v4 P-tunnel. So I think the wording should remain as is. > - Section 3.2, 2nd para, 1st sentence is: > > "A single UDP datagram MAY carry multiple S-PMSI Join Messages, >as many as can fit entirely within it." > > What's the MTU of a UDP packet (or frame)? > Might there be a problem if the MTU of UDP doesn't match the MTU of > the link on the PE? This doesn't appear to be covered by the sentence > above, but should be, IMO (i.e., state that the link MTU is the max > that can fit new S-PMSI Join Messages, or the max bytes UDP allows > should be stated here explicitly). I think it would be an inappropriate layering violation to state the maximum UDP size in this specification. If someone were to create a UDP packet that exceeded the link MTU, presumably it would get fragmented and reassembled. This might not be a wise implementation, but it would still work, and I don't see a reason to prohibit it. > - as someone not familiar with Multicast VPNs, having the acronym > "S-PMSI Join" messages *not* exploded in the abstract is a bit > confusing. I did think about this, but I thought that using the term "Selective Provider Multicast Service Interface Join message" in the abstract would not be any less confusing to someone who is not familiar with the MVPN work, and it would be more confusing to someone who is familiar with the MVPN work. > I had to look into the first and second paragraphs of the > intro to determine for myself that it means "Selective Provider > Multicast Service Interface", That's a technical term defined in draft-ietf-l3vpn-2547bis-mcast. I'll bet you didn't look in that draft to see what this really means! This is one of those cases where anyone who actually has to use the document will know what an "S-PMSI Join" message is, even if they don't know what the acronym expands to. I realize that the RFC Editor will probably expand it anyway ;-) > - Intro, 3rd para, 1st sentence > s/specifications/capability (or capabilities) I think s/specifications/specification would be better. MVPN implementers are likely to also be BGP and/or LDP implementers, and in those protocols the term "capability" is used to refer to something that requires explcit advertisement. You had a number of valid complaints about the writing of the first two paragraphs of the introduction. What do you think of the following proposed rewrite: The Multicast Virtual Private Networks (MVPN) specification [MVPN] defines the notion of a "PMSI" (Provider Multicast Service Interface), and specifies how a PMSI can be instantiated by various kinds of tunnels through a service provider's network ("P-tunnels"). It also specifies the procedures for using PIM ("Protocol Independent Multicast", [RFC4601]) as the control protocol between Provider Edge (PE) routers. When PIM is used as the control protocol, PIM messages are sent through a P-tunnel from one PE in an MVPN to others in the same MVPN. These PIM messages carry customer multicast routing information. However, [MVPN] does not cover the case where the customer is using IPv6, but the service provider is using P-tunnels created by PIM over an IPv4 infrastructure. The MVPN specification also specifies "S-PMSI (Selective PMSI) Join" messages, which are optionally used to bind particular customer multicast flows to particular P-tunnels. However, the specification does not cover the case where the customer flows are IPv6 flows. It's true that both paragraphs end with the same "However, ...", but the first paragraph is about sending customer PIM/IPv6 messages though a P-tunnel in a v4 infrastructure, while the second paragraph is about using the S-PMSI Join message to assign specific customer multicast flows to a particular P-tunnel. ___ Ietf mailing list Ietf@ietf.org https://www.ietf.org/mailman/listinfo/ietf
Re: [mpls] Last Call: draft-ietf-mpls-ldp-upstream (MPLS Upstream Label Assignment for LDP) to Proposed Standard
With regard to this draft, I need to reiterate a comment I made during WG last call, as I think there is a procedural issue that needs to be brought to the IESG's attention. The draft has a normative reference to RFC 3472 "Generalized Multi-Protocol Label Switching (GMPLS) Signaling Constraint-based Routed Label Distribution Protocol (CR-LDP) Extensions", despite the fact that CR-LDP has been deprecated by RFC 3468 ("The MPLS Working Group decision on MPLS signaling protocols", February 2003). I don't think this is allowable; I interpret RFC 3468 as requiring that we do not take any action that prevents the deprecated CR-LDP documents from being reclassified as historic. The only reason the CR-LDP documents were not classified as historic seven years ago is that there are other standards organizations that have produced specs with normative references to CR-LDP. Section 4.3 of RFC 3468 says: standards organizations which reference the document [the CR-LDP specs], need to be notified of our decision so that they (at their own pace) can change their references to more appropriate documents. It is also expected that they will notify us when they no longer have a need to normative reference to CR-LDP. I think the clear implication of this is that neither the IETF nor any other SDO should be creating new normative references to CR-LDP documents. Note that RFC 3468 explicitly calls out RFC 3472 as one of the deprecated documents. During WG LC, one of the authors stated that he disagreed with my interpretation, but WG consensus on this issue was never obtained. At this point, I believe the only change that is needed to draft-ietf-mpls- ldp-upstream is to move the reference to RFC 3472 into the "Informational References" section. That is, I think that recent revisions to the draft have made the normative reference gratuitous, as one does not need to read RFC3472 in order to implement any part of this draft. If the draft is published with this normative reference, it is almost inevitable that someone will someday say "we don't want to use LDP upstream-assigned labels, because they depend on CR-LDP and CR-LDP has been deprecated". ___ Ietf mailing list Ietf@ietf.org https://www.ietf.org/mailman/listinfo/ietf
Re: draft-housley-two-maturity-levels
I don't think folks have appreciated how truly insidious Russ' document is. First, he proposes to eliminate a set of processes that are frequently used to portray the IETF in a negative light: - The use of the label "Standard" for the never-used third level of standardization - The never-used two year limit on "Proposed Standards" - The process by which "STD numbers" are assigned to obsolete documents, but not to the documents that obsolete them. This blatant attempt to improve the public image of the IETF is an obvious conflict of interest for the IETF chair or any other IESG member! To the rest of us, the ability to portray the IETF as a laughing stock is of great value, especially when we disagree with its results. But then comes the most insidious proposal of all: the elimination of the prohibition against "downward references". This clever attempt to remove one of the disincentives to advancing documents along the standards track may at first glance seem innocent enough. However, if the community were to go along with this attempt to remove disincentives, we might actually end up with a two level standards process. Those of us who are happy with the de facto single level standards process should oppose this change at all costs. (Fortunately, this tricky proposal is unlikely to succeed in its goal, as there are so many other disincentives that the document fails to address.) This document should be sent to a WG where it can be extensively discussed and analyzed by those members of the community who have the most experience in failing to achieve consensus on process change. (Of course, first the broader community must spend a year or two agreeing on the charter of the WG.) The document should then be advanced on the Standards Track, under the current standards process. That means that two different Standards Development Organizations must be created independently from the process specification in the document, and we must see an implementation report proving that those two organizations can interoperate. Only then will it be appropriate to decide whether the document should become an IETF standard. Just in case this is an insufficient method of ensuring that there is no progress, I strongly suggest that the issue of specifying the standards process be delayed until all issues of document input and output formats, including internationalization, and representation of diagrams, have been fully discussed and decided upon. ___ Ietf mailing list Ietf@ietf.org https://www.ietf.org/mailman/listinfo/ietf
Re: Last Call: draft-ietf-l3vpn-2547bis-mcast (Multicast in MPLS/BGP IP VPNs) to Proposed Standard
Thank you for your review. I will try to address your technical comments. Some of your comments are not really technical objections, but criticisms of the way the document is constructed, or of the scope of the document, or of the approach adopted by the WG. As the construction of the document is the result of years of difficult negotiation and careful compromise, I will not address those issues. Also, I will not attempt to address the "philosophical issues" having to do with what is and is not a proper use of BGP. Pekka> 1) IPv6 support. The spec apparently aims to support both IPv4 and Pekka>IPv6 because it refers to both in a couple of places. Yet, there Pekka>is at least one explicit place in the spec (S 7.4.2.2) that's not Pekka>compatible. Pekka>I suspect many of the BGP attributes used, possibly Pekka>also the MCAST-VPN BGP SAFI and others are not IPv6 compatible. MCAST-VPN BGP SAFI is certainly IPv6 compatible; the details may be found in draft-ietf-l3vpn-2547bis-mcast-bgp-08.text. Section 4 of that draft discusses the use of AFI 2 (IPv6) with the MCAST-VPN SAFI, and all talk of "VPN-IP" routes in that document is meant to indicate either VPN-IPv4 or VPN-IPv6. Pekka>At the minimum, the status (intent) of the spec should be Pekka>clarified. Even better would be to improve and include the support Pekka>here. In general, the procedures specified in the document will enable an IPv4 SP backbone to support customer use of IPv6 multicast. You are correct that section 7.4.2.2 is incomplete in this respect. Pekka> 2) RP configuration in SP network. It's not clear if SP network Pekka>needs to know how customer sites have configured their RPs (when Pekka>the customer provides the RP). At least traditional PIM Pekka>signalling would require SP to know this. But if auto-rp or BSR Pekka>is not used by the customer, how is this information learned and Pekka>maintained? Would it require manual configuration? A PE router does function as a PIM peer of a CE router, and hence needs some way to get the group-to-RP mappings of the customer. How this is done is for the SP and the customer to determine. Pekka> 5) Active source BGP messages. This is a duplication of a similar Pekka>mechanism in MSDP (RFC3618) which has caused much grief in Pekka>Internet. Does this meant that when a host does 'nmap -sU Pekka>224.0.0.0/4' at a VPN site, this will result in about 268 million Pekka>BGP active source updates being sent (2^28) in the SP backbone? There are two uses of Source Active routes, described respectively in sections 9 and 10. As used in section 9, this problem cannot arise because the SA routes are not generated as the result of receiving data traffic. The use described in section 10 does however need to be discussed in the security considerations. Pekka>This problem is not described in security considerations. >From the Security Considerations section: "an implementation SHOULD provide mechanisms that allow a SP to place limitations on the following: - total number of (C-*,C-G) and/or (C-S,C-G) states per VRF" Since SA AD routes are generated only as a result of creating the corresponding multicast states, limiting the number of multicast states per VRF results in limiting the number of Source Active routes. A more specific discussion can be found in the Security Considerations section of draft-ietf-l3vpn-mcast-bgp: "In conjunction with the procedures specified in Section "Supporting PIM-SM without Inter-Site Shared C-trees" an implementation SHOULD provide capabilities to impose an upper bound on the number of Source Active A-D routes, as well as on how frequently they may be originated. This SHOULD be provided on a per PE, per MVPN granularity." Pekka> 6) PIM-BIDIR usage. May the SP use PIM-BIDIR internally even if the Pekka>customer interface would use PIM-SM? I'm not sure I understand exactly what you are asking. The technology for building the P-tunnels is completely independent of the multicast technology used by the customer. Pekka> 7) Type 0 Route Distinguisher. The spec mandates using type 0 RD which Pekka>embeds 16-bit AS-number. Good catch. This is actually an error in the spec. The material in section 8.2 reflects some earlier thinking about procedures that were revised and worked out in more detail in draft-ietf-l3vpn-2547bis-mcast-bgp. What we will do is eliminate the text in section 8.2 and replace it with a reference to the mcast-bgp draft. > 3.2. P-Multicast Service Interfaces (PMSIs) > Multicast data packets received by a PE over a PE-CE interface must > be forwarded to one or more of the other PEs in the same MVPN for > delivery to one or more other CEs. Pekka> .. is this strictly accurate? doesn't this depend on where the RP is Pekka> configured to be? This seems to assume that the RP configuration is Pekka> always provided by th
Re: RFC archival format, was: Re: More liberal draft formatting standards required
> huge number of mobile devices that handle HTML effortlessly and IETF > legacy ASCII not at all HTML is a good presentation format, but as an archival format it seems to leave a lot to be desired, as the included links always seem to go stale. Also, I don't think that the notions of "page numbers" and "table of contents" have quite reached the status of "quaint and archaic". > large number of standard office printers that print HTML instantly and > correctly My experiences printing HTML docs are a bit different, I guess. Is there no tool that will html-ize an RFC or i-d for presentation? Folks who want to read RFCs on their cell phones should be able to do it, but I really don't see what that has to do with archival formats. ___ Ietf mailing list Ietf@ietf.org https://www.ietf.org/mailman/listinfo/ietf
Re: xml2rfc is OK ( was: Re: XML2RFC must die, was: Re: Two different threads - IETF Document Format)
Lars> since you asked: I have absolutely no problems with xml2rfc. I find that xml2rfc takes too much control over the boilerplate and the references to be a really useful tool. I dropped it after one attempt. However, many of my colleagues use it, and as a result I've gotten many questions of the form "what do I have to do to make xml2rfc produce output that will pass idnits?". I can't tell them "just put in the following boilerplate", instead I've had to figure out the right value of the "ipr" variable. (BTW, no one ever cares what the boilerplate actually says, just whether it will pass idnits; xml2rfc really encourages folks to ignore the semantics of the boilerplate.) Joel> One large draft I was working on was originally written using WORD. I Joel> found it extremely difficult to work with (although I have a current Joel> version of Word available at all times.) Instead, working with Joel> another author we converted the document to XML for XML2RFC. Hey, I've converted from both Word and XML to nroff; that way I don't get any surprises ;-) OTOH, I have to admit that nroff was a bit challenging when I moved from Solaris to Linux. Joel> I have seen some folks arguing that we should make XML2RFC normative Joel> and mandatory. Of course, in the IETF it is very common for folks to think that their personal preferences are objectively superior. ___ Ietf mailing list Ietf@ietf.org https://www.ietf.org/mailman/listinfo/ietf
Re: LC summary for draft-ietf-opsawg-operations-and-management
Looking through your summary of the LC comments, it appears that there is considerable sentiment to publish this document as Informational rather than as a BCP. Yet the new revision still says "Intended Status: BCP". > [dbh: this document went to great lengths to say that it was NOT > prescribing a Management Considerations requirement. sigh] If this document were to proceed as a BCP, the following text could be interpreted by an AD as license to require a "Management Considerations" section: "Any decision to make a Management Considerations section a mandatory publication requirement for IETF documents is the responsibility of the IESG, or specific area directors, or working groups, and this document avoids recommending any mandatory publication requirements." Since assigning responsibilities to the IESG is presumably out of scope of this document, why not shorten this sentence to: "This document avoids recommending any mandatory publication requirements" ___ Ietf mailing list Ietf@ietf.org https://www.ietf.org/mailman/listinfo/ietf
Re: Last Call: draft-ietf-opsawg-operations-and-management(Guidelines for Considering Operations and Management of NewProtocols and Protocol Extensions) to BCP
> This does not mean we have to simply accept what they (OPS) say. But it > does mean we should give it a fair review, looking at the details, > rather than objecting on principle. This is absolute nonsense. Most of the people actually doing work in the various areas do not have the time, interest, or expertise to do a detailed review of an OPS document. However, these are the people who are in the best position to determine whether "OAM Considerations" would help or hinder the work that they do. If we are going to talk about adding new hoops for folks to jump through, we should first discuss whether any such hoops are necessary. We should not start the discussion by looking at the details of the particular proposed hoops. > the OPS area has as much right to propose their requirements as any other > area (Transport Congestion, Security, ...) has. And generally, the > community has listened to such requests and gone along with them. Generally, the community (i.e., the folks doing the work in the various areas) has never even heard about these proposed requirements until after a BCP appears, at which time they are told that the BCP "has community consensus". Perhaps you're familiar with Douglas Adams' "The Hitchhiker's Guide to the Galaxy". To paraphrase, "but the plan for the destruction of earth was clearly posted in the planning department on alpha centauri, it's not our fault you didn't see it". ___ Ietf mailing list Ietf@ietf.org https://www.ietf.org/mailman/listinfo/ietf
Re: Last Call: draft-ietf-opsawg-operations-and-management(Guidelines for Considering Operations and Management of NewProtocols and Protocol Extensions) to BCP
Adopting this document as a BCP would be a serious mistake, and I hope it will be strongly opposed. There is absolutely no evidence that following the dictates of this document will improve the quality of the IETF's work, and we certainly know it won't improve the timeliness. There is no evidence that ignoring it will harm the Internet. I don't see that OPSAWG has any business imposing requirements on work done by other WGs in other Areas. There are already an adequate number of obstacles to getting work done. The last thing we need is another required "considerations" section, or another set of excuses for ADs from one Area to block documents from other Areas. ___ Ietf mailing list Ietf@ietf.org https://www.ietf.org/mailman/listinfo/ietf
Re: Proposed Experiment: More Meeting Time on Friday for IETF 73
I oppose this experiment. A better experiment would be to eliminate the Friday morning sessions. ___ Ietf mailing list Ietf@ietf.org https://www.ietf.org/mailman/listinfo/ietf
Re: Nomcom process realities of "confidentiality"
> The random selection process coupled with the relatively small number > of volunteers is an interesting factor in all of this Nomcom is a group of people randomly selected from among a set of folks whose only qualifications are that they want to be on nomcom and they like traveling to meetings. I've never understood the logic behind the theory that this is supposed to lead to a good slate of candidates. > The Giant Reset of 2006 Oh I remember that. For fairness, we replaced one randomly selected group of folks with another randomly selected group of folks from the same volunteer pool. We wouldn't want to end up with the wrong randomly selected group, would we? That's another bit of logic I never understood. Given that the whole process is whacky, fine tuning of various little details probably won't make much of a difference. ___ IETF mailing list IETF@ietf.org https://www.ietf.org/mailman/listinfo/ietf
Re: Experimental makes sense for tls-authz
As a personal political view, I happen to be opposed to the notion of software patents. But I still think that the document in question should be published as Experimental: - It's quite plain that this political view has never been adopted by IETF consensus. (I also think it plain that it has no chance of being adopted by IETF consensus.) - I don't think the IETF considers "this document offends my political point of view" to be a legitimate reason for opposing the document. The degree of passion and/or repetition with which the political view is expressed is irrelevant. (The suppression of a document for political reasons is frequently called "censorship", even if other avenues of publication still exist.) - It's really within the province of each WG to determine whether its standards are implementable by whoever needs to implement them in order for the standard to be successful. This may or may not include open source implementations. - If a particular proposal is technically sound, but not adopted because the WG thinks that its patent encumbrances are a bar to implementation, then it is perfectly valid to publish the proposal as a non-standard track RFC. The only real criterion is that the technical content be interesting or otherwise worth preserving. With regard to the coordinated letter writing attack being waged on this list, well, we're all familiar with the situation in which folks try to get their way by getting lots of non-participants to send scripted messages. Often you can tell that the message writers don't even know what the issues are, but at least most of the letter writing campaigns pretend to be about technical issues; the current campaign doesn't even bother with the pretense! ___ Ietf mailing list Ietf@ietf.org https://www1.ietf.org/mailman/listinfo/ietf
Re: Charging I-Ds
Eric Gray> The discussion is essentially inane I think this is an excellent observation. It suggests to me though that perhaps the best way to get more funding for the IETF is to impose a surcharge on inane messages to the ietf mailing list. The surcharge can be based on the degree of inanity of the message. I suggest the following schedule of charges: - $10 for a generic message whining about US customs/immigration processes - $10 for a clueless message suggesting a reorganization of the IETF or a change of the fee structure (fortunately not to be imposed retroactively) - $10 for a message about the value or lack thereof of Ascii art - $10 for a message about the format of RFCs - $15 for a message whining about US customs/immigration processes, if the whine is backed up only by anecdotes - $100 for a message suggesting that US customs/immigration processes are unfair to white men from western europe. I'd raise the fee to $500 if sent by someone with an obvious chip on his shoulder. - $100 for a message suggesting that IETF meetings be held in peculiar locations - $100 for a message suggesting that the cookies at IETF meetings should be rationed - $100 for a message stating that the list is full of inane messages (not to be imposed retroactively) - $200 for a message saying that NAT is evil - $200 for a message whining about the IETF's lack of sufficient emphasis on IPv6 - $500 for a message whining about the fact that IETF meetings do not routinely occur in one's home town. I would raise the fee to $1,000 if Barcelona is mentioned. - $500 for a message saying that the job of the IETF is to prevent the marketplace from making technology choices - $1000 for a message stating that the poster knows how to solve the spam problem once and for all. Of course, this is not an exhaustive list of inane categories of message, it's just a start. If, during the course of a week, a single poster sends multiple inane messages which say exactly the same thing, I would double the fee for each subsequent message. Putting such a schedule of charges into place would either eliminate the IETF's budget problems or else make its mailing list a lot more useful. ___ Ietf mailing list Ietf@ietf.org https://www1.ietf.org/mailman/listinfo/ietf
Re: Last Call: draft-klensin-norm-ref (Handling Normative References for Standards Track Documents) to BCP
This document has a reasonable goal, but the implementation is objectionable. The reasonable goal (from the introduction): This document replaces the "hold on normative reference" rule with a "note downward normative reference and move on" approach for normative references to standards-track documents and BCPs. The objectionable implementation: A note is included in the reference text that indicates that the reference is to a target document of a lower maturity level, that some caution should be used since it may be less stable than the document from which it is being referenced In many cases (probably the vast majority) where a document is advancing despite a "downward normative reference", the referenced document (and the technology described therein) is no less stable than the referencing document, and no "caution" is required. It's great to be able to "make a downward reference and move on", but we should not be required to tell lies and spread FUD. An acceptable implementation would be: A note is included in the reference text that indicates that the reference is to a target document which is at a prior step in the official IETF standardization process. That statement adequately captures the facts, and does not require the WG to make dishonest statements or to spread FUD. Another alternative, probably a better one, is simply to annotate each reference with its standards status, and make no editorial comment about it at all. ___ Ietf mailing list Ietf@ietf.org https://www1.ietf.org/mailman/listinfo/ietf
Re: As Promised, an attempt at 2026bis
On the issue of whether we have a de facto one-step process, the real question is not whether subsequent steps are ever invoked, but whether the subsequent steps actually have any practical impact on the Internet. One can certainly point to a handful of cases where the subsequent steps are invoked, but the point is that it makes no difference to the Internet whether the subsequent steps are invoked or not. So I think it is quite accurate to say that we have a de facto one-step process. It is thus logical for advocates of the one-step process to argue that we in fact have more or less what we need, and to be skeptical of anything that might result in giving more credence to (or even calling more attention to) the subsequent steps. The real problem with the process is that a protocol can be widely deployed in multiple interoperable implementations for six or seven years before its specification even achieves this one step. This can happen because the WG gets inundated with idiots, and/or because companies are using the IETF as a marketing battleground, and/or because the IESG deliberately tries to obstruct progress, and/or because the security ADs require you to figure out the insanely complicated endsystem-oriented security architecture so you can explain why you don't need to adhere to it. I'm pretty sure there is no IESG-wide consensus on how to address these issues, but if one has suffered through any of these multiple year delays, one is likely to oppose anything that reeks of "more process". This can lead one to be suspicious even of writeups that claim to be "descriptive", as the writeups may (whether intended to do so or not) serve to extend the life of various problematical processes that might wither away faster if they were never written down. Sometimes it's just better to leave well enough alone. ___ Ietf mailing list Ietf@ietf.org https://www1.ietf.org/mailman/listinfo/ietf
Re: My notes on draft-carpenter-newtrk-questions-00.txt
The focus on document relationships rather than on simplifying the standards track is what (well, is one of the things) that sent newtrk off into the weeds. Frankly, I don't care if someone on a desert island cannot figure out from the RFCs alone how to implement some protocol. I just don't see that as a problem we have to solve. Anyway, implementing a protocol requires so much more knowledge than can be obtained from the RFCs that no amount of messing with the document strategy is going to have any impact on the ability of our castaway to become a successful implementer. > it's worth revisiting the whole ISD/SRD thing with one new ground rule - > no more work for IESG in steady state Let me suggest a second new ground rule: no more work for the document authors, document editors, or the WG chairs. ___ Ietf mailing list Ietf@ietf.org https://www1.ietf.org/mailman/listinfo/ietf
Re: Last Call: 'A Process Experiment in Normative Reference Handling' to Experimental RFC (draft-klensin-norm-ref)
> that text is not derogatory, but a simply statement of fact. Sorry, but however you may try to talk your way out of it, a statement like "that technology may be unstable" is derogatory. > Until and unless the definitions of maturity levels are changed, that text > is not derogatory, but a simply statement of fact. I'm afraid that the facts as to whether a technology is stable are in no way dependent on the IETF's definitions of maturity levels. > If a WG agrees with you about a particular piece of technology, > they have three choices: Well, 4: they can issue the doc as a PS obsoleting the old PS. > If I writing a document that needed to reference a specification > that was as well-defined, mature, and stable as you posit, I'd > first try to get that specification advanced to the right > maturity level That's an interesting fact about yourself, but personally I'd prefer to spend my time doing something useful. > But the assertion you are making about a (e.g.) Proposed > Standard specification being stable, mature, well-defined, > widely-deployed, etc., is one that presumably should get some > community review Sure. The WG should not advance a doc to DS if it really depends on something which isn't stable. The WG needs to be aware of the facts, and should not be compelled to insert statements which they know to be false. > you should support this on your theory that it will create more arguments > and bog things down further. No, I don't think there's any need to do anything that creates more arguments and bogs things down further. I understand that there's no consensus on how to avoid the iceberg, but that doesn't mean I want to take the time to run experiments on more complicated ways to arrange the deck chairs. ___ Ietf mailing list Ietf@ietf.org https://www1.ietf.org/mailman/listinfo/ietf
Re: Last Call: 'A Process Experiment in Normative Reference Handling' to Experimental RFC (draft-klensin-norm-ref)
> The IESG plans to make a decision in the next few weeks, and solicits > final comments on this action. If the individual submission is approved as an Experimental RFC, does that mean that the IETF will adopt the proposed "experiment"? If so, I don't think this draft should be approved. (Actually, I suspect the fix is in, but for the record ...) The proposal seems primarily intended to deal with the following "problem". Sometimes there are cases in which a doc is ready to become a DS, but cannot, because of the infamous "downref rule", which states that no DS can normatively reference a PS. The proposal leaves the downref rule in place, but allows it to be waived if the WG is willing to approve derogatory text about the referenced technology: "A note is included in the reference text that indicates that the reference is to a document of a lower maturity level, that some caution should be used since it may be less stable than the document from which it is being referenced," Frankly, I think this wavier procedure is outrageous, and entirely unacceptable. The fact The fact that the referenced document has not gone through some bureaucratic process does not mean that it is any less stable, or that any more caution is required in its use. Inserting this derogatory language about technology which may be well-proven and widely deployed will be extremely misleading to the industry. I think that any rule which requires us to insert false and misleading statements in the documents should be strongly opposed. Even worse: "The IESG may, at its discretion, specify the exact text to be used" Great, not only is the WG required to denigrate its own technology, but the IESG is given free rein to insert whatever derogatory remarks they feel like putting in. Of course, we'll be told not to worry, since: "If members of the community consider either the downward reference or the annotation text to be inappropriate, those issues can be raised at any time in the document life cycle, just as with any other text in the document." Great. Another useless thing to argue about in the WG, and another useless thing to argue about with the IESG. There are also other reasons why I find this proposed experiment disheartening. For one thing, it really misses the point. We need to simplify our processes, not make them more complicated. Either we need the downref rule or we don't. If we want to experiment, let's experiment with eliminating the rule entirely, not with fine tuning it. The real underlying problem of course is the the multi-stage standards process is just a relic from another time, and makes no sense at all in the current environment. Experiments in fine tuning the process are nothing but a distraction. ___ Ietf mailing list Ietf@ietf.org https://www1.ietf.org/mailman/listinfo/ietf
Re: IETF Meeting Venue Selection Criteria
> There is no objective way to identify 'primary contributors' other than by > assuming the regular attendees are also contributors. This is simply silly. It's not much of a secret, in any WG, who does the work and who comes to listen. > We've tried looking at how many local first-time attendees from (say) > Korea later became regular attendees but the data are hard to state in any > meaningful way and the time constants are long (years). This is a somewhat round-about way of saying that you have no data to support your position. > We certainly know that going a long way from most places, > as we did in Adelaide, impacts attendance significantly - > but my recollection is that Adelaide was a very successful > meeting in terms of WGs making progress. Obviously recollections differ. By scattering meetings all over the world, with no consideration of the average travel time, you encourage the creation of a class of professional standards-meeting-attenders, which is just the opposite of what is wanted. > income [from local participants] that we badly need. Well, this is the first I've heard that we want to maximize the number of people who come to listen rather than to work. Everything I've ever heard in the past suggested the opposite. If we now want to maximize the number of passive attendees, I'm sure we can find a way to do it without scattering the meetings all around the world. ___ Ietf mailing list Ietf@ietf.org https://www1.ietf.org/mailman/listinfo/ietf
Re: IETF Meeting Venue Selection Criteria
I find this whole discussion to be nothing but an attempt by some to push their personal political views and preferences down the throats of everyone else. I don't think there should be any political preconditions on the IETF venue. If we are going to start requiring that the host country be governed in accordance to our own political views, I have quite a few preconditions I could offer, but they're quite different from the ones Avri offered. Further, if we're going to select host countries based on how convenient or inconvenient it is to enter that country, the preference should be weighted by the number of attendees (or even better, by the number of active contributors), so that we don't eliminate venues which are convenient to the many but inconvenient to the few. ___ Ietf mailing list Ietf@ietf.org https://www1.ietf.org/mailman/listinfo/ietf
Re: Let's make the benches longer.... (Re: draft-klensin-nomcom-term-00.txt)
> the normal process for AD replacement involved choosing which of the > people who had worked with the AD for a long time could do the job this > time, In American vernacular, this procedure is known as "cronyism". Generally, one doesn't expect to see this advocated in a public forum ;-) ___ Ietf mailing list Ietf@ietf.org https://www1.ietf.org/mailman/listinfo/ietf
Re: [newtrk] Re: List of Old Standards to be retired
Eliot> Even if someone *has* implemented the telnet TACACS user option, Eliot> would a user really want to use it? Eric> I don't know. Do you? Eliot> Yes, I do. Many of us do. And that's the point. I'm sure you think you know, but I don't know that you know, which means that a lot of people have to waste a lot of time preventing you from doing damage. Are you also the one who "knows" that OSPF demand circuits are never used? ___ Ietf mailing list [EMAIL PROTECTED] https://www1.ietf.org/mailman/listinfo/ietf
Re: [newtrk] Re: List of Old Standards to be retired
> This is a simple way to simply do what we said we were going to do. The worries about this whole anti-cruft business have always been (a) the likelihood that it would do more harm than good, and (b) the likelihood that it would waste enormous amounts of time on issues of no importance. The initial foray into this area doesn't do much to relieve these worries. > Even if someone *has* implemented the telnet TACACS user option, would a > user really want to use it? I don't know. Do you? How do we go about answering a question like that? How much time should the IETF spend determining the answer to this question? It's just better to leave well enough alone. ___ Ietf mailing list [EMAIL PROTECTED] https://www1.ietf.org/mailman/listinfo/ietf
Re: List of Old Standards to be retired
I see this exercise has already reached the point of absurdity. How can it possibly be worth anyone's time to look at each telnet option and determine whether it is deployed? What possible purpose could be achieved by changing the standards status of some telnet option? Is there some chance that someone is going to implement one of these by mistake somehow? A similar comment applies to the FDDI MIB. Are we trying to make sure that no one implements that MIB by mistake? Or that no one implements FDDI by mistake, just because he thinks there's an IETF standard about it? Let me echo Bob Braden's "if it's not broken, why break it?" query. ___ Ietf mailing list [EMAIL PROTECTED] https://www1.ietf.org/mailman/listinfo/ietf
Re: archives (was The other parts of the report....
I've never thought that the IETF was OBLIGATED to "hide" old I-Ds; that seems a rather far-fetched interpretation of anything in RFC 2026. However, I think there is a real practical problem in making the old i-d's be too readily available. I frequently get messages asking me questions like "where is draft-rosen-something-or-other-04.txt, I can't find it" to which the answer is one of the following: a. you want draft-rosen-something-or-other-23.txt, or b. you want draft-ietf-somewg-something-or-other-05.txt, or c. you want RFC 12345. What's happened is that they have found some email which references a long outdated draft, and have no clue how to get to the most up-to-date version, which is what they really want to see. If we make it too easy to access the old drafts, a lot of people will just get the old drafts instead of being forced to look for the more recent work. Sure, people who really want to see the old drafts should be able to get them, but people who really want to see the most up-to-date versions shouldn't get the old drafts just because they only know an old draft name. In a perfect system, someone would go to the IETF's official I-D page, enter a draft name, and get a prominent pointer to the most recent version (even if it is now an RFC or a draft with a different name), along with a less prominent pointer to the thing they actually asked for. If that can't be done, it might be better to keep the expired drafts "officially hidden". Not for the reasons being given by our more academically inclined colleagues, but for the practical reasons described above. Sure, the expired drafts might be obtainable via Google, but getting something from Google is a bit different than getting it via the IETF's official web page. ___ Ietf mailing list [EMAIL PROTECTED] https://www1.ietf.org/mailman/listinfo/ietf
Re: IETF mission boundaries (Re: IESG proposed statement on the IETF mission )
> The gist of this comment is that someone developing a network > application protocol ought to somehow get a blessing from the IETF. > Reality check. Who got the IETF approval to deploy ICQ, Kazaa, or for > that matter HTTP? The fact that someone did something without the IETF's approval does not imply that what they did is outside the scope of the IETF, or that it is beyond the IETF's mission.
Re: IETF mission boundaries (Re: IESG proposed statement on the IETF mission )
> Sheesh!--next you'll be telling us that you never heard the phrase > "out of scope" before last week. Sure I have. There's hardly a piece of work done by the IETF that someone hasn't claimed to be out of scope. It's just that the phrase is not used consistently. If we look at the historical facts about the work that the IETF has traditionally taken on, it's hard to draw any conclusion other than that anything is in scope which promotes and facilitates the use of the Internet and of IP infrastructure. And I think that's exactly what the IETF should be doing. > The example I'm thinking about involved predecessors to OpenGL. As this example doesn't even involve communication over a network, I would agree that it is out of scope. But that's a rather extreme case, most of the contentious areas do involve communications over an IP infrastructure. > Often the brutal WG chairs say they don't think the WG knows enough, but > it's the scope arguments that carry the day. I've never had much luck myself with scope arguments, unless they could be backed up with an argument either that the center of expertise is elsewhere, or that the topic has no bearing on IP. Of course, people will sometimes be willing to agree that the center of expertise is elsewhere without necessarily agreeing that they themselves aren't experts ;-) Sometimes scope arguments are merely face-saving ways of saying "we don't know what we are doing". Other times, scope arguments are merely "polite" ways of saying "we don't think you know what you are doing". You almost never hear someone saying "that sounds like a really good idea, but unfortunately it is out of scope".
Re: IETF mission boundaries (Re: IESG proposed statement on the IETF mission )
> That is wrong or at least a gross overstatement. If that's what you think, I invite you to make a list of all the IETF-standardized protocols and explain how they are all (or even more than 50% of them) needed to make the Internet work. > There have been many things that the IETF has chosen to step away from but > that ran and run over the Internet. Some graphics standards come > immediately to my mind ... Those graphics standards were kept out of the > IETF not because the working groups involved thought they didn't think > they were experts, but because the subject was out of scope for the IETF." I'm not familiar with this particular case, but I don't see why protocols for distributing graphics would be thought to fall outside the scope of the IETF, any more than protocols for distributing voice or video. Of course, graphics standards that have nothing do with distribution of the graphics over IP would be out of scope. > No committee is ever able to limit itself on grounds of insufficient > expertise. Now, there is a gross overstatement! For everyone who proclaims himself (rightly or wrongly) to be an expert on some topic, there are always two other people who claim that he is clueless. It's not uncommon for a WG to refuse to pick up a topic because the consensus is that the topic's proponents are clueless.
Re: IETF mission boundaries (Re: IESG proposed statement on the IETF mission )
> - "For the Internet" - only the stuff that is directly involved in making > the Internet work is included in the IETF's scope. In other words, routing, DNS, and Internet operations/management. Adopting this as the IETF's mission would be a very radical change indeed! While this particular mission statement does seem to reflect the interests of a certain notorious IESG member, let's not pretend that this has ever been the limit of the IETF's mission. The IETF has always been concerned with things that make the Internet more useful, and with things that expand the utility of the IP protocol suite. There's never been a time when "for the Internet" was an accurate representation of the IETF's concerns. You are of course welcome to propose such a radical change to the IETF's mission. But if you are going to circulate a document under the subject line "IESG proposed statement on the IETF mission", you should make it clear that the IESG is proposing to make a complete change in the IETF mission. Instead, you give the impression that the IESG thinks that "for the Internet" is and has always been the IETF's mission. The formulation I like is "Everything that needs open, documented interoperability and runs over the Internet is appropriate for IETF standardization". This is much truer to the IETF's current and historical practice. That doesn't necessarily mean that the IETF has to standardize everything that falls within its mission. For instance, a particular area might fall within the mission, but the IETF might not have the expertise to tackle it. A WG in that area could then be rejected on the grounds of "insufficient expertise". Such decisions would have to be made on a case-by-case basis. Again, this is the way such decisions have always been made in the IETF.
Re: IESG proposed statement on the IETF mission
> "The purpose of the IETF is to create high quality, relevant, and timely > standards for the Internet." > It is important that this is "For the Internet," and does not include > everything that happens to use IP. IP is being used in a myriad of > real-world applications, such as controlling street lights, but the > IETF does not standardize those applications. Well, let's test this assertion. Suppose a consortium of electric companies develops a UDP-based protocol for monitoring and controlling street lights. It turns out that this protocol generates an unbounded amount of traffic (say, proportional to the square of the number of street lights in the world), has no congestion control, and no security, but is expected to run over the Internet. According to you, this has nothing to do with the IETF. It might result in the congestive collapse of the Internet, but who cares, the IETF doesn't do street lights. I would like to see the criteria which determine that telephones belong on the Internet but street lights don't! Another problem with your formulation is that the Internet is a growing, changing, entity, so "for the Internet" often means "for what I think the Internet should be in a few years", and this is then a completely unobjective criterion. One would hope instead that the IETF would want to encourage competition between different views of Internet evolution, as the competition of ideas is the way to make progress. I also do not understand whether "for the Internet" means something different than "for IP networking" or not. I think it should also be part of the mission to produce standards that facilitate the migration to IP of applications and infrastructures that use legacy networking technologies. Such migration seems to be good for the Internet, but I don't know if it is "for the Internet" or not.
Re: The requirements cycle
Eric> Not sure what you mean, it always takes time to produce a document, Eric> even if the document is just a "rock fetch". Harald> sorry; "rock fetch" is beyond my scope of American idiom. "Rock fetch": when the "boss" sends the workers out on useless but time-consuming tasks. Harald> But version -01 of the framework document is dated July 19, 2001, Harald> and the first version submitted to the IESG is dated February 15, Harald> 2002. About six months to get the WG to agree on the framework, that doesn't seem excessive for a document. It's a rock fetch though because there is no real need for a framework document. Harald> I took that as a hint that there might have been controversy in the Harald> working group about it. It was never a very controversial document. My recollection is that framework and requirements were ready about January of 2002, which is why I said that they were ready about 18 months ago. So were the protocol specs, applicability statements, etc. Eric> Well, each objection from the IESG needs to be discussed and a Eric> response crafted. Harald> which should take approximately 3 days of work, IMHO. Comments that Harald> translate to "you are referencing an obsolete version of LDAP" Harald> should take approximately 2 minutes to fix. Comments which were received last fall (I first saw them a few weeks prior to the Atlanta IETF meeting) required a considerable reworking of the document. (Sisyphus comes to mind here ;-)) Harald> Did the WG declare consensus on all those documents 18 months ago Harald> (January 2002)? The WG was told by the WG chairs that the IESG would not allow the WG to even consider the solutions documents until the framework and requirements documents were approved by the IESG. Something is very wrong with the process here. The L3VPN protocol specs themselves haven't changed in years, which is a good thing, given the large amount of interoperable deployment by multiple vendors!
Re: The requirements cycle (Re: WG review: Layer 2 Virtual....)
Harald> did any of the technologies change because of issues that were Harald> discovered in the discussions that were needed to clarify the Harald> requirements and framework? No. Harald> If no - why did it take any time at all to produce them? Not sure what you mean, it always takes time to produce a document, even if the document is just a "rock fetch". Harald> there is little that the IESG can do when the WG knows what the Harald> comments are and chooses not to act upon them for 2-5 months. This reminds me of Dilbert's pointy-haired boss, who says "your project is late, so I want you to give me hourly status reports." When we have documents which aren't really necessary in the first place, which ultimately will not have any impact on the technology, but which need to be massaged and remassaged so as to get them past the IESG, I think it's quite clear where the responsibility for the delay is coming from. Harald> And I don't understand why WG updates to fix problems take 2-3 Harald> months per cycle when the WG thinks that it's important to be Harald> finished with the docs. Well, each objection from the IESG needs to be discussed and a response crafted. Harald> is the IESG supposed to care about inconsistencies between the Harald> requirements (which are what the *WG* thinks should be satisfied) Harald> and the technologies that will be proposed for standardization? Sure; but the reqs, framework, protocol specs, and applicability statements were all ready 18 months ago. They could have been submitted as a group. But we were told, "first you need to submit the first document, then a year or so later you can submit the second". This is a very peculiar way to encourage progress ;-) From the WG perspective, the specs have been ready for review forever, but the IESG has refused to look at them because of bogus process issues. And then they turn around and accuse the WG of making slow progress!
Re: WG review: Layer 2 Virtual Private Networks (l2vpn)
Harald> It might have something to do with the fact that the WG has not Harald> requested that the IESG process these drafts if the WG has not Harald> come to consensus on asking for the drafts to be published, I'm Harald> afraid the IESG cannot do anything. I consider this answer to be rather disingenuous. The WG has not requested that the IESG process these drafts because the WG chairs have told the WG that the ADs have told them that the drafts in question cannot be submitted to the IESG until numerous other drafts that no one will ever read (requirements, framework, architecture) have been approved by the IESG. Of course, most of those numerous other drafts were completed about 18 months ago, though a few of them have now come out of the seemingly endless "IESG reviews, WG makes minor change, IESG reviews, WG change" cycle. So you can't honestly answer Yakov by saying "the WG hasn't asked us to process these drafts"; the answer to Yakov's question would be an explanation of (a) why all these prior drafts are really necessary, (b) why it is reasonable for such a long review cycle, and (c) why it is reasonable to delay starting to process the protocol specs until the prior specs are already on the RFC Editor's queue.
Re: WG review: Layer 2 Virtual Private Networks (l2vpn)
People need to understand that the purpose of the Pseudowire stuff (PWE3) is to enable service providers to offer existing services over IP networks, so that they can convert their backbones to IP without first requiring that all their customers change their access equipment. Producing the protocols needed to enable migration from legacy networks to IP networks seems to me to be quite in the mainstream of IETF. The technical issues, involving creating tunnels, multiplexing sessions through tunnels, performing service emulation at the session endpoints, are all issues that the IETF has taken up in the past, there is nothing radically different going on here. (To those who think that other standards organizations can do this better, well, representatives from those other organizations feel free to drop in on the WGs in question, so we are familiar with their level of expertise on IP. Let's just say that if we want to aid in the migration of legacy networks to IP, these other organizations are not what we would want to rely on.) One can think of the VPWS work in L2VPN as taking the PWE3 stuff and adding some IP-based auto-discovery mechanisms to facilitate provisioning. Again, this isn't out of line with what the IETF typically does. The VPLS work is more difficult to position within the IETF, as it is hard to avoid a lot of stuff that overlaps with IEEE (a standards org which really is worthy of respect, unlike some others), and extending ethernet over an IP network is arguably a bad idea. On the other hand, the purpose is the same as indicated above; service providers can migrate from their Q-in-Q ethernet networks to IP networks, without first requiring their customers to switch from an ethernet service to an IP service.
Re: v6 support (was Re: Thinking differently about the site local problem (was: RE: site local addresses (was Re: Fw: Welcome to the InterNAT...)))
Steve> I can't get upset about Microsoft declining to ship poorly-tested Steve> code. Given how many security holes are due to buggy, poorly-tested Steve> programs, I applaud anyone who takes that seriously. Well, suppose they were to ship IPv6 without IPsec, on the grounds that they didn't have the testing resources for IPsec. Would you still be applauding them? Or would you be questioning whether they have their priorities right? Features always fall off due to the inability to allocate sufficient testing resources, but a vendor does have some choice over which features those are.
Re: IETF Sub-IP area: request for input
Lars> An example is PPVPN, which is chartered to work on specification of Lars> requirements, with new protocol work being explicitly out-of-scope. Lars> However, some current PPVPN IDs (and several more targetted at it) Lars> read more like solution documents >From the PPVPN charter: This working group is responsible for defining and specifying a limited number of sets of solutions for supporting provider-provisioned virtual private networks (PPVPNs). It is somewhat difficult to define and specify a solution without writing something that "reads like a solution document". Lars> for various existing vendor schemes, >From the PPVPN charter: The working group is expected to consider at least three specific approaches Various "existing vendor schemes" are then explicitly mentioned. Lars> new protocol work being explicitly out-of-scope [but PPVPN documents Lars> are] specifying packet headers and MIBs In some cases the PPVPN docs do have protocol work in them which needs to be moved to another working group. But I don't think this is a case of the group going beyond its charter, it's just a matter of the individual contributors getting the time to divide up the documents properly. In other cases, the PPVPN docs just specify how existing protocols are to be used to achieve various functions. I don't think defining a MIB counts as "new protocol work".
Re: IETF Sub-IP area: request for input (fwd)
I might as well chime in on the actual question that was asked. I guess I disagree with the majority of folks working in the sub-IP area. I never thought it made any sense to move all those working groups out of their original areas into a "sub-IP" area, and I never understood the "sub-IP area hourglass architecture" that was foisted on us by the IESG. So I've never thought that it makes much sense to have a separate "sub-IP" area, and I don't think it makes sense to keep it as a separate area in the future. The advantage of maintaining the status quo is that everyone has gotten used to the current ADs, and most people figure that any change will make things worse. When new ADs get involved, they tend to "reinterpret" the charters and disrupt the work. But if one ignores the possibility of short-term personality issues, I think it would be better to choose established areas for the MPLS, CCAMP, and PPVPN WGs.
Re: a personal opinion on what to do about the sub-ip area
Keith> In my experience, IESG has tremendous breadth - considerably Keith> exceeding that of any single WG. You must be joking. Or perhaps you just mean that you tend to agree with the IESG's program of trying to preserve the academic, ivory tower vision of the Internet against the advance of the barbarians (i.e., marketplace forces). Naturally every special interest group claims to be the defender of the values of the larger community. Since there is no way to determine objectively what is or is not in the "larger community's" interest, a properly functioning IESG would not try to impose a particular vision, but would just work to ensure that the output of the WGs is of suitable technical quality. (Of course, every attempt to impose by fiat a particular vision of the future is portrayed as an attempt to ensure technical quality.) Keith> I've certainly seen ... the contributions of "outside" participants Keith> dismissed as irrelevant, by even working group chairs. Gee, we often hear from the in-crowd that one of the problems with the IETF is that the WG chairs aren't forceful enough in dismissing irrelevant input. Perhaps "irrelevant" is in the eye of the beholder. I tend to think that irrelevant input should be dismissed more often by the chairs, as long as "irrelevant" doesn't become a smokescreen for "doesn't accord with my personal vision of the future." Keith> I've also seen working groups drastically exceed, and in some cases Keith> ignore, charters which were designed to limit the harm they could Keith> do. In cases like this, the charters are often dictated by the IESG, do not necessarily reflect a good understanding of the WG's subject matter, and may place arbitrary "prior restraint" restrictions on the solution. Sometimes it's difficult to do a good job while trying to strictly follow the charter.
Re: a personal opinion on what to do about the sub-ip area
> The workings of special interest groups can and often do have a > significant effect on the general population, but nobody can afford the > time and energy it takes to keep track of every special interest group > that might affect him. Often it seems as though the WGs reflect the broad consensus of the community, and the IESG is the special interest group.
Re: IETF Sub-IP area: request for input (fwd)
Joe> Many of these discussions (layer 2 VPNs, in particular) would be better Joe> served by occuring within the context of their original host Joe> organization (i.e., IEEE for ethernet over IP), since it was those Joe> organizations that defined those LANs, and they who would best comment Joe> on the correctness (or lack) of proposed solutions. IEEE is certainly not the right place to determine how to carry ethernet data and control frames over IP networks. Nor is it the right place to design the auto-discovery mechanisms for figuring out which IP systems need to connect to which other IP systems in order to provide layer 2 VPN services. Nor is it the right place to determine how the characteristics of IP networks may be leveraged to optimize for layer 2 VPNs. Nor is it the right place to address the scalability, security, and applicability issues of layer 2 VPNs over IP networks. In fact, the IEEE 802.1 participants who are following the layer 2 VPN work have not even claimed that these issues fall within their scope. While I am not a big fan of emulating ethernet networks over IP networks, this is a pretty clear example of a topic that has both IETF and IEEE components, and which needs attention from BOTH groups.
Re: IETF Sub-IP area: request for input (fwd)
Aaron> I can easily imagine this is so, although, as I say above, I have no Aaron> facts to back this up. Wouldn't it be nice if people based their feedback on facts, rather than on what they imagine! Well, at least you're honest about it ;-) Aaron> If sub-ip represents technologies that don't, and will never, get Aaron> involvement from a "broad spectrum of the IETF community," we Aaron> shouldn't institutionalize it. How would you define "a broad spectrum of the IETF community"?
Re: Why IPv6 is a must?
Sure, in theory one could add zillions of new globally routable addresses without increasing the size of the routing tables in the default-free zone at all. The skepticism is about whether there is (or even could be) a realistic plan to make this happen.
Re: trying to reconcile two threads
Fred> I see a longish thread about the fact that some cable companies Fred> apparently are desperate to charge per IP address (something one can Fred> only do if IP addresses are in fact a scarce resource) I think you miss the point here. The cable companies want to charge per computer, and the only way they can do this is to count the number of IP addresses they see. If you use NAT, they only see one. This has nothing whatsoever to do with scarcity of addresses. Even with IPv6, they would still want to charge by the IP address, and people would still use NAT to save money.
Re: Why IPv6 is a must?
Eric> NAT is what has prevented us from returning to the pre-1978 situation. Keith> this is true only if you believe that [blah blah blah] The situation today with NAT is that hosts in separate realms can only communicate in 99% of the desired applications, though perhaps this falls to 80% if one stubbornly ignores the existence of tunneling and port redirection. Pre-1978, you were either directly attached to the Arpanet or you were pretty much out of luck. You have to be very much in the grip of a theory to regard these situations as comparable. Granted, it's easier to talk about the evils of NAT than to explain how billions of new routable addresses are going to be added to the existing routing system.
Re: Why IPv6 is a must?
Brian> NAT has simply pushed us back to the pre-1978 situation. On the contrary, NAT has allowed us to maintain global connectivity without requiring every system to have a globally unique address. NAT is what has prevented us from returning to the pre-1978 situation. That's not to say it wouldn't be better to have a million more globally unique addresses. Sure it would, unless that would stress out the routing system unduly. If adding a million more globally unique addresses will stress out the routing system, then one might argue that a solution which provides the addresses but doesn't change the routing system isn't really deployable, and hence doesn't really solve the addressing problem. I think this is the point that Noel keeps trying to drive home, and I'm not sure I understand what the answer is supposed to be.
Re: filtering of mailing lists and NATs
Maurizio> but this means Maurizio> - that there is a person who has the right to decide whether the Maurizio> message is spam or not Maurizio> - that this person is willing to bear the burden for the sake of the Maurizio> whole community. Maurizio> I happen to do this for some lists, but it's a nuisance, I may Maurizio> assure you. I do this for the mailing list of the MPLS working group, so I'm aware of what a nuisance it is. But as far as mailing list management goes, it's not nearly as big a nuisance as trying to figure out which of the error messages to owner-mpls are bogus and which are real. (The mailing list has 3000 members and each message to it results in 100 error messages.) It's not hard to decide whether a particular message is unsolicited commercial email or not, that's not something that people disagree about.
Re: filtering of mailing lists and NATs
Christian> I would much rather receive and delete another annoying Christian> proposition to get rich quick or see lurid pictures than tolerate Christian> any form of censorship. As has been pointed out, the non-member messages can be moderated. It takes about one second to look at a message and tell whether it is unsolicited commercial or not. So the downside is that the non-member message may be delayed for a bit until the moderator gets to it. I wouldn't call that censorship. (I think one has to be very privileged indeed to confuse a small inconvenience with censorship.) Unless, of course, you think that people have a RIGHT to send unsolicited commercial email to IETF mailing lists.
Re: filtering of mailing lists and NATs
So, here are the choices: 1. Save thousands of people from having to deal with multiple spams per day, at the cost of presenting a minor inconvenience to a few, or 2. Require thousands of people to receive and deal with spam (or to learn all about mail filtering), in order to avoid inconveniencing a few. Easy decision to make. For every bit of whining by the usual suspects, there are thousands of folks that are very happy to have the spam kept out of their mailbox automatically. (Every mailing list manager knows that the whining by Keith and Lloyd is nothing compared to the whining by the list members as they get spammed multiple times per day.) Indeed, this is a lot like the arguments re NAT. There are the thousands of people it helps, vs. the few who are yelling that the sky will fall if it is not stamped out.
Re: rfc publication suggestions
Fred> the most common hold-up that I see is a wait for normative Fred> references. That held the MPLS documents up for a year. This is not exactly true. In a number of cases, the documents were explicitly crafted to have only non-normative references to internet-drafts, but when the documents reached the RFC Editor's queue, all the references were mysteriously declared to be normative. Questions about this mysterious reinterpretation just went unanswered. My suspicion is that this has more to do with the overload on the IESG members than anything else. Another source of delay was the tendency of IESG comments to dribble in. Just as the documents were about to be sent to the RFC editor, another IESG member would have a comment. By and large, the IESG comments were very reasonable, but for a long time it seemed like the review process would never end. Again, overload on the IESG members is likely responsible for this.
Re: specific questions about RFC publication and I-Ds
Pete> Does this particular entry mean is being Pete> held for normative reference to the I-Ds listed below it, or that all Pete> those I-Ds are being held for normative references? The meaning of that queue entry is that the first document mentioned is being delayed because of normative references to the I-Ds that appear beneath it in the list. Pete> Assuming that the entry above indicates that the first I-D (in this Pete> case, ) is being held until the other Pete> RFCs become RFCs as well, doesn't this mean that it might be years Pete> before an approved I-D would ever be published as an RFC? In fact, this is the actual situation.