[lisp] Re: Gunter Van de Velde's No Objection on draft-ietf-lisp-name-encoding-10: (with COMMENT)

[Dino Farinacci <[e-mail]>]

> ----------------------------------------------------------------------
> COMMENT:
> ----------------------------------------------------------------------
> 
> # Gunter Van de Velde, RTG AD, comments for draft-ietf-lisp-name-encoding-10

Thanks for your review Gunter. See responses inline.

> #DETAILED COMMENTS
> #=================
> ##classified as [minor] and [major]
> 
> 10      Abstract
> 11
> 12         This draft defines how to use the AFI=17 Distinguished Names in LISP.
> 
> [minor]
> This abstract is rather brief and could use some more meat to the bone to
> sumamrize the content of the document. What about the following proposed
> textblob:
> 
> "
> This document specifies an encoding format for names in LISP. The proposed
> encoding supports various naming schemes, including DNS names, distinguished
> names, and user-defined names, facilitating the integration of LISP with
> diverse applications and services. The encoding ensures efficient and scalable
> name resolution within the LISP mapping system. Additionally, the document
> addresses interoperability considerations and provides guidelines for
> implementation. This work aims to enhance the flexibility and applicability of
> LISP in modern network environments. "

We describe in more detail in the introduction. The abstract is purposely short.

> 
> 116        17.  This draft defines a termination character, an 8-bit value of 0
> 117        to be used as a string terminator so the length can be determined.
> 
> [minor]
> RFC0020 seems to name the 0000 000 ascii characted 'NUL'. WOuld that make sense
> to mention or name the character like that in this document?
> 
> 139          0                   1                   2                   3
> 140          0 1 2 3 4 5 6 7 8 9 0 1 2 3 4 5 6 7 8 9 0 1 2 3 4 5 6 7 8 9 0 1
> 141         +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+
> 142         |            AFI = 17           |       ASCII String ...        |
> 143         +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+
> 144         |               ...  ASCII String             |       0         |
> 145         +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+

This spec wants to use a one-byte value of 0 to teriminate the string. Current implementations have zero problem with this.

> [minor]
> Clarification. Is the '0' termination character assumed to be at a 32bit
> boundary? or can it be somewhere else? Maybe worthwhile to explicit document
> the expectation. RFC states that an ASCII character is represented using 7
> bits. However, in practice, it is often stored in an 8-bit byte, with the extra
> bit typically set to zero.

No because the string is variable length and can be in the middle of any short-word (16-bit) or long-word boundary (32-bit).

> 218     9.  Sample LISP Distinguished Name (DN) Deployment Experience
> 
> 220        Practical implementations of the LISP Distinguished Name
> 221        specification have been running in production networks for some time.
> 222        The following sections provide some examples of its usage and lessons
> 223        gathered out of this experience.
> 
> [minor]
> I believe that this complete section is informational and belongs more in an
> appendix to make it explicit that its not part of the formal procedure outlined
> in this document and are examples.

I don't think we need to split hairs. Going to leave it where it is since your comment is minor.

Dino