Gen-ART LC review of draft-ietf-ecrit-location-hiding-req-01

[Ben Campbell <[e-mail]>]

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-ecrit-location-hiding-req-01
Reviewer: Ben Campbell
Review Date: 20090507
IETF LC End Date: 20090511
IESG Telechat date: (if known)

Summary:

This document is almost ready for publication as an informational RFC.  
There are some minor clarity issues where the reader is left to infer  
some things that could be more explicit.

Major issues:

None



Minor issues:

-- 1.1, last paragraph:

Can you expand on how withholding information information needed for  
call routing concretely differs from withholding information from  
emergency personnel? I assume there is more to this than the intent of  
the ISP. Also, by saying an ISP is "not interested", I think the point  
is to say that they have legal obligations to disclose to emergency  
personnel, regardless of any interest otherwise, right?

-- 1.2, first paragraph:

I think this leaves out what I assume to be the actual problem  
statement, which is we need a way that an ISP/IAP can hide location  
info from the user agent of the VSP in such a fashion that it is still  
available for PSAP routing, correct? I can infer that pretty easily,  
but I don't see where it is explicitly stated in one place.

Is there a case where an ISP is simply unable to provide location  
information? I assume that would be out of scope for this document,  
but it should be stated as such.



-- 1.3, fourth paragraph:

This paragraph could be more clear--how does the PSAP having  
credentials meet a requirement to _hide_ information? I infer the  
assumption is that the caller does _not_ have the necessary  
credentials. If so, it would be better to state it explicitly

-- Fifth paragraph:

is compatibility with LoST a requirement?

-- Req-B

Is it appropriate for this document to put requirements on the ISP/ 
IAP? Or do you mean to say they MUST be _able_ to support this, while  
hiding information location from the VSP and/or UA?

-- Req-C

I don't really understand what is being said here. Is the point to say  
that they must be able to validate that the URI identifies a "bona  
fide" emergency service contact, and that a call to that URI actually  
routes to the right place? How does this interact with the later  
requirement that the entities need not be SIP aware?

-- Req-D

this is stated as a requirement on the ISP rather than a statement  
about the solution. I _think_ you are saying there is a requirement to  
be _able_  to provide location info to the PSAP while withholding it  
from the caller. Is that correct?. Also how does "by value or by  
reference" interact with the previous statement concerning LoST  
requiring LbyV?

-- Req-5

How does the requirement that the ISP/IAP not need to know SIP  
interact with the statement in Req-D that the ISP must be able to  
determine if a call is being routed to a bona-fide location service?  
Also, does Req-5 imply a requirement to work with non-SIP VoIP services?

-- Req-6

What does it mean for a PSAP boundary to have holes?

-- Req 12:

"Minimal impact" is vague--can you add clarifying text to make this  
more concrete?

-- Req 15: 

Is that really a requirement, or just an observation of fact?

-- Security Considerations:

I'm a little skeptical of this statement that this does not raise  
additional considerations. For example, would you consider that a  
human might be endangered because an ISP wanted to reserve location  
information as a "for pay" service a security consideration, in that  
it requires the solution to be more fail-safe than other protocols? On  
the other hand, is the need to keep the UA from inferring location  
when an ISP wants to hide it a security consideration?

Nits/editorial comments:

-- Abstract, paragraph 2:

It's not clear to me that the document described architectural  
impacts. It refers to architecture, but I don't see explicit  
statements about how the architecture breaks if the ISP is not willing  
to disclose.

-- 1.1, list item "3."

Please expand VSP on first use.

-- Req-A:

I don't think the requirement is to be able to withold location from  
"any entity it wishes", since that would include the PSAP, etc.

-- Req-2: "jurisdiction of the PSAP"

Geographical jurisdiction?

-- Req-10:  The solution MUST allow the end host to determine PSAP/ 
ESRP URLs prior to the call, for all emergency services.

Who is the "end host"?

-- 3.3, first bullet:

Is it appropriate to have "MUST"s in a section on "desirable  
properties"?

-- Third bullet:

That's an implementation detail. I think you mean to say something to  
the effect of the presence of NATs SHOULD not break the mechanism.






-- idnits reports the following (which I include without prejudice):

idnits 2.11.11

tmp/draft-ietf-ecrit-location-hiding-req-01.txt:

   Checking boilerplate required by RFC 5378 and the IETF Trust (see
   http://trustee.ietf.org/license-info):
    
----------------------------------------------------------------------------

   ** It looks like you're using RFC 3978 boilerplate.  You should  
update this
      to the boilerplate described in the IETF Trust License Policy  
document
      (see http://trustee.ietf.org/license-info), which is required from
      December 16, 2008.  Version 1.34 of xml2rfc can be used to produce
      documents with boilerplate according to the mentioned Trust  
License
      Policy document.

   -- Found old boilerplate from RFC 3978 Section 5.1 on line 22.

   -- Found old boilerplate from RFC 3978 Section 5.5 updated by RFC  
4748 on
      line 377.

   -- Found old boilerplate from RFC 3979 Section 5 paragraph 1 on  
line 388.

   -- Found old boilerplate from RFC 3979 Section 5 paragraph 2 on  
line 395.

   -- Found old boilerplate from RFC 3979 Section 5 paragraph 3 on  
line 401.