Dear Elwyn, The just submitted draft-ietf-v6ops-addcon-08 has the following changes injected based on your review. Please let me know if you see further open items.
******** I have been selected as the General Area Review Team (Gen-ART) reviewer for this draft (for background on Gen-ART, please see http://www.alvestrand.no/ietf/gen/art/gen-art-FAQ.html). Please wait for direction from your document shepherd or AD before posting a new version of the draft. Document: draft-ietf-v6ops-addcon-07.txt Reviewer: Elwyn Davies Review Date: 6 May 2008 IESG Telechat date: 08 May 2008 Summary: This document looks to be in fairly good shape but there are a number of areas where it really needs attention from an editor with English mother tongue - IMO this ought to be done before it goes to the RFC Editor as in some cases they make it technically unclear. There are also a couple of places (notably around ULAs) where there is some FUD in the document that is (apparently) not tied to specific issues. This is undesirable as a prospective user has no idea whether these alleged issues affect his/her network. Comments: s1: Needs a little additional explanation of 'interface-id'. GV> WAS: GV> and the 'interface-id'. GV> CHANGED TO: GV> and the 'interface-id' (the identification of the interface within a subnet). s1, bullet 3 (near end): s/traffic storm and level/potential traffic storms and the level/ No further mention of traffic storms is made in the document... should it be to give some hints on subnet sizing? GV> WAS: GV> Even though there is no broadcast for the IPv6 protocol, there is still need to GV> consider the number of devices in a given subnet due to traffic storm and level of GV> traffic generated by hosts. GV> CHANGED TO: GV> REMOVED this bullet s2.1, para 2: s/may thus have two addresses/may thus have two or more addresses/ GV> WAS: GV> may thus have two addresses GV> CHANGED TO: GV> may thus have two or more addresses s2.2, para3: 'Using a random chosen ULA address will be conform in case (a) provide suboptimal aggregation capability, while in case (c) there will be overconsumption of address space.' This sentence is not good English and I don't understand what it is trying to say regarding case (a). GV> WAS: GV> Using a random chosen ULA address will be conform in case (a) provide suboptimal GV> aggregation capability, while in case (c) there will be overconsumption of address GV> space. GV> CHANGED TO: GV> Using random chosen ULA addresses will provide in case (a) suboptimal aggregation GV> capabilities, while in case (c) a /48 ULA address is larger then the less-than-/48 GV> prefix and will hence result in address space overconsumption. s2.2, last para: ' The usage of ULAs should be carefully considered even when not attached to the IPv6 Internet as some IPv6 specifications were created before the existence of ULA addresses.' This piece of FUD is not terribly helpful - do we have any idea what pieces of specification might conflict or (at least) work differently when using ULAs - or are you effectively just repeating what has been said already? GV> Note from Gunter Van de Velde: GV> Source address selection with multicast addresses might for example go wrong as GV> the ULA is selected due to longest match. If this stream would hit any other networks GV> in anyway, then things might break. For clarity sake and to avoid FUD perception the GV> para has been removed GV> WAS: GV> The usage of ULAs should be carefully considered even when not attached to the IPv6 GV> Internet as some IPv6 specifications were created before the existence of ULA GV> addresses. GV> CHANGED TO: GV> REMOVE para s3.3, para 3: Ir would be useful to explain why the u and g bits might come back to bite us in future or provide a reference which discusses why they are relevant at all. GV> WAS: GV> When assigning a subnet prefix of more then 70 bits, according to RFC4291 [26] "u" and GV> "g" bits (respectively the 71st and 72nd bit) need to be taken into consideration and GV> should be set correctly. In currently implemented IPv6 protocol stacks, the relevance GV> of the "u" (universal/local) bit and "g" (the individual/group) bit are marginal and GV> typically will not show an issue when configured wrongly, however future GV> implementations may turn out differently. GV> CHANGED TO: GV> When assigning a subnet prefix of more then 70 bits, according to RFC4291 [26] "u" and GV> "g" bits (respectively the 71st and 72nd bit) need to be taken into consideration and GV> should be set correct. GV> The "u" (universal/local) bit is the 71st bit of IPv6 address and is used to determine GV> whether the address is universally or locally administered. If 0, the IEEE, through the GV> designation of a unique company ID, has administered the address. If 1, the address is GV> locally administered. The network administrator has overridden the manufactured address GV> and specified a different address. GV> The "g" (the individual/group) bit is the 72st bit and is used to determine whether the GV> address is an individual address (unicast) or a group address (multicast). If "0", the GV> address is a unicast address. If "1", the address is a multicast address. GV> In current IPv6 protocol stacks, the relevance of the "u" and "g" bit is marginal and GV> typically will not show an issue when configured wrongly, however future GV> implementations may turn out differently if they would be processing the "u" and "g" GV> bit in IEEE like behavior. s3.3.1.1, last para: 'No additional dependencies for the subnet prefix while the EUI-64 and IID dependencies will be discussed later in this document.' I cannot parse this sentence. It probably needs 'There are..' added to the front and some additional wordsmithing (maybe like the last sentence of 3.3.1.2). GV> WAS: GV> No additional dependencies for the subnet prefix while the EUI-64 and IID dependencies GV> will be discussed later in this document. GV> CHANGED TO: GV> There is no additional dependency for the subnet prefix with the exception of the GV> EUI-64 and an IID dependency. These will be discussed later in this document. s4.2: RFC 3041 has been obsoleted by RFC 4941. I believe this covers the same ground needed here. GV> WAS: GV> Existing references to RFC3041 GV> CHANGED TO: GV> New References to RFC4941 sA.2.1.1: References to LIR/RIR allocation policies would be useful. GV> Done Editorial: General: idnits reports several out of date references. GV> Done General: s/Global Unique Address(es)/Globally Unique Address(es)/ GV> Done General: Need to use (modified) title case for section headings GV> Done Reference format: The RFC Editor now insists on symbolic references rather than numeric. You may wish to choose the symbols and modify refs appropriately. GV> Added in the document: GV> <?rfc symrefs="yes" ?> s1, last para: s/is provider/is either provider/ GV> Done s2.1, para 2: s/operative/operational/ GV> Done s2.2, para 6: Expand RPF acronym. GV> Done s2.3: Expand RIR acronym. GV> Done s2.4.2, para 2: Expand HD acronym. GV> Done s3.2, last para: s/for different then/with other than/ GV> Done s3.3, para 1: s/futile benefit/little benefit/ GV> Done s3.3.1.1, last para: expand EUI-64 and IID. (IID is expanded later in s4) GV> Done s3.3.2, last para: s/eventhough/even though/ GV> Done s3.3.3, last para: Not good english: replace with 'Note that the definition of ISATAP does not support multicast.' GV> Done s3.3.4: s/The 126/126/ GV> Done sA.1.1, last para: expand GLOP, ASN, BGP GV> GLOP is not an acronym, but I added reference to RFC where GLOP is defined GV> Expanded on both BGP and ASN sA.2.1.1: Expand LIR GV> Done sA.2.1.4: s/thumb rules/rules of thumb/ GV> Done -----Original Message----- From: Elwyn Davies [mailto:[EMAIL PROTECTED] Sent: Tuesday, May 06, 2008 7:52 PM To: General Area Review Team Cc: Mary Barnes; gunter (mailer list); Chip Popoviciu (cpopovic); [EMAIL PROTECTED]; [EMAIL PROTECTED]; [EMAIL PROTECTED]; [EMAIL PROTECTED] Subject: Gen-art review of draft-ietf-v6ops-addcon-07.txt I have been selected as the General Area Review Team (Gen-ART) reviewer for this draft (for background on Gen-ART, please see http://www.alvestrand.no/ietf/gen/art/gen-art-FAQ.html). Please wait for direction from your document shepherd or AD before posting a new version of the draft. Document: draft-ietf-v6ops-addcon-07.txt Reviewer: Elwyn Davies Review Date: 6 May 2008 IESG Telechat date: 08 May 2008 Summary: This document looks to be in fairly good shape but there are a number of areas where it really needs attention from an editor with English mother tongue - IMO this ought to be done before it goes to the RFC Editor as in some cases they make it technically unclear. There are also a couple of places (notably around ULAs) where there is some FUD in the document that is (apparently) not tied to specific issues. This is undesirable as a prospective user has no idea whether these alleged issues affect his/her network. Comments: s1: Needs a little additional explanation of 'interface-id'. s1, bullet 3 (near end): s/traffic storm and level/potential traffic storms and the level/ No further mention of traffic storms is made in the document... should it be to give some hints on subnet sizing? s2.1, para 2: s/may thus have two addresses/may thus have two or more addresses/ s2.2, para3: 'Using a random chosen ULA address will be conform in case (a) provide suboptimal aggregation capability, while in case (c) there will be overconsumption of address space.' This sentence is not good English and I don't understand what it is trying to say regarding case (a). s2.2, last para: ' The usage of ULAs should be carefully considered even when not attached to the IPv6 Internet as some IPv6 specifications were created before the existence of ULA addresses.' This piece of FUD is not terribly helpful - do we have any idea what pieces of specification might conflict or (at least) work differently when using ULAs - or are you effectively just repeating what has been said already? s3.3, para 3: Ir would be useful to explain why the u and g bits might come back to bite us in future or provide a reference which discusses why they are relevant at all. s3.3.1.1, last para: 'No additional dependencies for the subnet prefix while the EUI-64 and IID dependencies will be discussed later in this document.' I cannot parse this sentence. It probably needs 'There are..' added to the front and some additional wordsmithing (maybe like the last sentence of 3.3.1.2). s4.2: RFC 3041 has been obsoleted by RFC 4941. I believe this covers the same ground needed here. sA.2.1.1: References to LIR/RIR allocation policies would be useful. Editorial: General: idnits reports several out of date references. General: s/Global Unique Address(es)/Globally Unique Address(es)/ General: Need to use (modified) title case for section headings Reference format: The RFC Editor now insists on symbolic references rather than numeric. You may wish to choose the symbols and modify refs appropriately. s1, last para: s/is provider/is either provider/ s2.1, para 2: s/operative/operational/ s2.2, para 6: Expand RPF acronym. s2.3: Expand RIR acronym. s2.4.2, para 2: Expand HD acronym. s3.2, last para: s/for different then/with other than/ s3.3, para 1: s/futile benefit/little benefit/ s3.3.1.1, last para: expand EUI-64 and IID. (IID is expanded later in s4) s3.3.2, last para: s/eventhough/even though/ s3.3.3, last para: Not good english: replace with 'Note that the definition of ISATAP does not support multicast.' s3.3.4: s/The 126/126/ sA.1.1, last para: expand GLOP, ASN, BGP sA.2.1.1: Expand LIR sA.2.1.4: s/thumb rules/rules of thumb/ _ _______________________________________________ Gen-art mailing list [email protected] https://www.ietf.org/mailman/listinfo/gen-art
