Hi, Richard,

Yes, -09 does address all the comments I sent on -07.

Thanks for your quick response!

Spencer

----- Original Message ----- From: "Richard Barnes" <rbar...@bbn.com>
To: "Spencer Dawkins" <spen...@wonderhamster.org>
Cc: <draft-ietf-georpiv-lbyr-requireme...@tools.ietf.org>
Sent: Monday, November 30, 2009 4:39 PM
Subject: Re: Gen-ART review of draft-ietf-geopriv-lbyr-requirements-07


Spencer,

-09 is out now:
<http://tools.ietf.org/rfcdiff?url2=draft-ietf-geopriv-lbyr-requirements-09.txt
>
Does these changes address your concerns?

--Richard



On Oct 21, 2009, at 8:30 PM, Spencer Dawkins wrote:

Hi, Russ,

I'm sorry, but I don't see text changes based on my comments in http://tools.ietf.org/wg/geopriv/draft-ietf-geopriv-lbyr-requirements/draft-ietf-geopriv-lbyr-requirements-08-from-07.diff.html .

Some of them were nits, which aren't actually part of the Gen-ART review, but my concerns about 2119 language and a couple of sentences that didn't parse probably are worth looking at (just search for "pars" in my review).

Thanks,

Spencer


Spencer:

Draft -08 is on the telechat agenda for tomorrow. Does it resolve your concerns?

Russ

At 06:47 PM 6/3/2009, Spencer Dawkins wrote:
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 resolve these comments along with any other Last Call  comments
you may receive.

Document: draft-ietf-geopriv-lbyr-requirements-07
Reviewer: Spencer Dawkins
Review Date: 2009-06-03
IETF LC End Date: 2009-06-09
IESG Telechat date: (not known)

Summary: This document is almost ready for publication as an Informational RFC.

Comments: Most of my notes below involve text clarity. I did question one 2119 keyword use in Section 4.1, as well.


1.  Introduction

 Location determination, different than location configuration or

Spencer (clarity): Is this s/different than/as distinct from/ ? but this sentence didn't parse for me.

 dereferencing, often includes topics related to manual provisioning
 processes, automated location calculations based on a variety of
 measurement techniques, and/or location transformations, (e.g.,  geo-
 coding), and is beyond the scope of this document.

 The issues around location configuration protocols have been
 documented in a location configuration protocol problem statement  and
 requirements document [I-D.ietf-geopriv-l7-lcp-ps].  There are
 currently several examples of a location configuration protocols
 currently proposed, including, DHCP
 [I-D.ietf-geopriv-dhcp-lbyr-uri-option], LLDP-MED, and HELD

Spencer (clarity): got a reference for LLDP-MED here?

 [I-D.ietf-geopriv-http-location-delivery] protocols.


2.  Terminology

 Location Configuration Protocol:  A protocol which is used by a
    client to acquire either location or a location URI from a

Spencer (clarity): I'm guessing here that you mean s/either location/either a location object/

    location configuration server, based on information unique to  the
    client.

3.3.  Location URI Authorization

 Location URIs manifest themselves in a few different forms.  The
 different ways that a location URI can be represented is based on
 local policy, and are depicted in the following four scenarios.

 1. No specific information at all:  As is typical, a location URI  is

Spencer (clarity): could this be something more like "No location information included in the URI:"?

    used to get location information.  However, in this case, the  URI
    representation itself does not need to reveal any specific
    information at all.  Location information is acquired by the
    dereferencing operation a location URI.

 2. No user specific information:  By default, a location URI MUST  NOT

Spencer (clarity): could this be "URI does not identify a user:"?

    reveal any information about the Target other than location
    information.  This is true for the URI itself, (or in the  document
    acquired by dereferencing), unless policy explicitly permits
    otherwise.

3.4.  Location URI Construction

 Depending on local policy, a location URI may be constructed in  such
 a way as to make it difficult to guess.  Accordingly, the form of  the
 URI is then constrained by the degree of randomness and uniqueness
 applied to it.  In this case, it may be important to protect the
 actual location information from inspection by an intermediate  node.

Spencer (clarity): is this section applicable to all of the scenarios in the previous section, or just to "possession authorization model"? If not applicable to all, you might point that out here.

 Construction of a location URI in such a way as to not reveal any
 domain, user, or device specific information, with the goal of  making
 the location URI appear bland, uninteresting, and generic, may be
 helpful to some degree in order to keep location information more
 difficult to detect.  Thus, obfuscating the location URI in this  way
 may provide some level of safeguard against the undetected  stripping
 off of what would otherwise be evident location information,  since it
 forces a dereference operation at the location dereference  server, an
 important step for the purpose of providing statistics, audit  trails,
 and general logging for many different kinds of location based
 services.

4.1.  Requirements for a Location Configuration Protocol

 C3. Location URI cancellation:  The location configuration protocol
    MUST support the ability to request a cancellation of a specific
    location URI.

    Motivation: If the client determines that in its best interest  to
    destroy the ability for a location URI to effectively be used to
    dereference a location, then there should be a way to nullify  the
    location URI.

Spencer (clarity): sorry, but can't parse this sentence. Something like "If the client determines that a location URI should no longer provide a location, then there should be a way to nullify the location URI"?

 C5. User Identity Protection:  The location URI MUST NOT contain
    information that identifies the user or device.  Examples  include
    phone extensions, badge numbers, first or last names.

Spencer (minor): this is probably a good idea, but I'm not sure it's a 2119 MUST (NOT). How would you recognize this on the wire (do you know what MY badge number is :-)?

    Motivation: It is important to protect caller identity or  contact
    address from being included in the form of the location URI  itself
    when it is generated.

 C7. Selective disclosure:  The location configuration protocol MUST
    provide a mechanism to control what information is being  disclosed
    about the Target.

Spencer (clarity): not obvious who is controlling this disclosure until you get to the motivation sentence - perhaps "a mechanism allowing the Rule Maker to control ..."?

    Motivation: The Rule Maker has to be in control of how much
    information is revealed during the dereferencing step as part of
    the privacy features.

 C8. Location URI Not guessable:  As a default, the location
    configuration protocol MUST return location URIs that are random
    and unique throughout the indicated lifetime.  A location URI  with
    128-bits of randomness is RECOMMENDED.

    Motivation: Location URIs should be constructed in such a way  that
    an adversary cannot guess them and dereference them without  having
    ever obtained them from the Target.

Spencer (clarity): s/ever/previously/ ?

4.2.  Requirements for a Location Dereference Protocol

 D1. Location URI support:  The location dereference protocol MUST
    support a location reference in URI form.

    Motivation: It is required that there be consistency of use
    between location URI formats used in an configuration protocol  and

Spencer (nit): s/in an/in a/

    those used by a dereference protocol.

5.  Security Considerations

 The method of constructing the location URI to include randomized
 components helps to prevent adversaries from obtaining location
 information without ever retrieving a location URI.  In the
 possession model, a location URI, regardless of its construction,  if
 made publically available, implies no safeguard against anyone  being
 able to dereference and get the location.  Care has to be paid when
 distribution such a location URI to the trusted location  recipients.

Spencer (clarity): s/distribution/distributing/ ?

 When this aspect is of concern then the authorization model has  to be
 chosen.  Even in this model care has to be taken on how to  construct
 the authorization policies to ensure that only those parties have
 access to location information that are considered trustworthy  enough
 to enforce the basic rule set that is attached to location
 information in a PIDF-LO document.

_______________________________________________
Gen-art mailing list
Gen-art@ietf.org
https://www.ietf.org/mailman/listinfo/gen-art




_______________________________________________
Gen-art mailing list
Gen-art@ietf.org
https://www.ietf.org/mailman/listinfo/gen-art

Reply via email to