[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

RE: [PWE3] Oustanding Discusses and comments on draft-ietf-pwe3-i ana-12.txt ( BCP ) -Brian Carpenter



Luca,

	You need to include RFC 2119 and BCP 79 in your reference section.

	Easy to forget, being part of the template, rather than part of your
own text...

--
Eric

--> -----Original Message-----
--> From: pwe3-bounces at ietf.org 
--> [mailto:pwe3-bounces at ietf.org]On Behalf Of
--> Luca Martini
--> Sent: Friday, October 21, 2005 12:09 AM
--> To: brc at zurich.ibm.com
--> Cc: Danny McPherson; pwe3; The IESG; Stewart Bryant
--> Subject: [PWE3] Oustanding Discusses and comments on
--> draft-ietf-pwe3-iana-12.txt ( BCP ) -Brian Carpenter
--> 
--> 
--> Brian,
--> 
--> Thanks you for your very good comments. Please find the 
--> replies in line below.
--> 
--> Luca
--> 
--> 
--> DISCUSSES AND COMMENTS:
--> ======================
--> Brian Carpenter:
--> 
--> Discuss [2005-10-11]:
--> Clarity and imprecision issues
--> (Some from me and some from a Gen-ART review by
--> Joel Halpern.)
--> 
--> LM: Note: I never saw a Gen-ART review. Did an e-mail get lost ?
--> 
--> > 2.1. Pseudowire Type
--> ...
--> >   ...PW Type values 31 through 1024 are to be assigned by IANA,
--> >   using the "Expert Review" policy defined in [RFC2434].
--> 
--> There are multiple calls for Expert Review in this draft
--> and no guidance as to what the Expert will be looking for. This
--> makes it a guessing game - can we have a few guidelines for each 
--> case of Expert Review?
--> 
--> LM: Yes. I did not want to be too restrictive. However I 
--> will add the following:
--> 
--> -----------------------------
--> .Pb "Expert Review Directives"
--> Throughout this document allocation procedures for several 
--> registries call for
--> an expert review process according to [RFC2434]. The expert 
--> should consider
--> the following points:
--> 
--> .Pz *
--> Avoid Duplication of code point allocations.
--> .Pz *
--> A brief clear desctiption of the code point allocation requested.
--> .Pz *
--> Whether the type allocaton requested is appropriate for the 
--> particular
--> requested value range in the registry.
--> .En
--> 
--> The Expert reviewing the request MUST provide an answer, 
--> approving, or
--> disapproving the request within 10 business days from when 
--> the he or she
--> received the expert review request.
--> 
--> -------------------------------------
--> 
--> 
--> >   A Pseudowire Type description of up to 65 characters is 
--> required for
--> >   any assignment from this registry.
--> 
--> I don't believe this length restriction is reasonable. I 
--> believe it is 
--> impossible to give an adequate summary in 65 characters in 
--> every case.
--> On the contrary, I believe there should be a minimum length, no
--> particular maximum length, a requirement to cite a reference, and
--> a requirement to cite a person or company. For the other sections
--> of the document, the length restriction is probably OK, but
--> the citations are needed.
--> 
--> LM: Citing a person/company is no problem. I will also 
--> remove the length restriction for 
--> the pw type allocation, as you requested. 
--> Citing a reference is a little more complicated. However I 
--> will add text to require a reference to a document.
--> 
--> 
--> >   Initial Pseudowire type value allocations are specified below:
--> >
--> >      PW type Description
--> >
--> >      0x0001  Frame Relay DLCI ( Martini Mode )
--> >      0x0002  ATM AAL5 SDU VCC transport
--> ...
--> 
--> These are not self-explanatory. I believe they should all 
--> include citations to wherever they are actually described.
--> 
--> LM: I think that this document is not supposed to be a 
--> glossary for PW types, so it's perfectly fine to have to go lookup
--> other documents that have the same text in the title. For 
--> example, presumably one would first read a PW encapsulation 
--> document, then look up this registry on the IANA web page. 
--> On the other hand , if ones searched for "Frame Relay Pseudowire" 
--> the future RFC describing in great details the frame relay 
--> encapsulation should be easily found.
--> However a reference to the relative document should be included.
--> 
--> 
--> > 2.2. Interface Parameters Sub-TLV type
--> ...
--> >   Any assignments requested from this registry require a 
--> description up
--> >   to 54 characters. Furthermore a length specified as follows:
--> >     - A decimal integer
--> >     - Text as follows:"up to X", where X is a decimal integer
--> >     - Up to 3 different decimal integers.
--> 
--> Does this mean
--> 
-->    The length is specified in one of the following formats:
--> 
--> or something else?
--> 
--> Why is "A decimal integer" listed? It is a subset of "Up to 
--> 3 different decimal
--> integers".
--> Does "up to X" mean "up to and including X"?
--> 
--> Same comments apply to section 2.3.
--> 
--> LM: it means " The length is specified in one of the 
--> following formats "
--> I will fix the text in the hole document.
--> 
-->  
--> Big nit: wherever RFCs are cited they should be proper references,
--> either normative or informative.
--> 
--> LM: I do not understand this comment. The only RFC cited is 
--> Normative. This is clearly indicated in the
--> "normative reference" section.
--> IANA registries do not have a reference section, so I do 
--> know how to put a reference there.
--> 
--> 
--> 
--> _______________________________________________
--> pwe3 mailing list
--> pwe3 at ietf.org
--> https://www1.ietf.org/mailman/listinfo/pwe3
--> 

_______________________________________________
pwe3 mailing list
pwe3 at ietf.org
https://www1.ietf.org/mailman/listinfo/pwe3