[freehaven-cvs] Removed commented out lines for submission to CACM

[syverson@xxxxxxxx]

Update of /home/freehaven/cvsroot/doc/cacm06
In directory moria:/tmp/cvs-serv14105/cacm06

Modified Files:
	netauth.tex 
Log Message:
Removed commented out lines for submission to CACM



Index: netauth.tex
===================================================================
RCS file: /home/freehaven/cvsroot/doc/cacm06/netauth.tex,v
retrieving revision 1.1
retrieving revision 1.2
diff -u -d -r1.1 -r1.2
--- netauth.tex	21 Jul 2006 16:40:45 -0000	1.1
+++ netauth.tex	2 Aug 2006 18:40:04 -0000	1.2
@@ -160,14 +160,6 @@
 deterrent~\cite{leyden05verizon}; since that time, Verizon has agreed to a
 settlement.
 
-%Sometimes gathering identifying user information is done under the guise of
-%providing a privacy policy and personalized service when the primary goal is
-%actually to build a profile for sale or for direct exploitation by the
-%provider.
-%
-% ***Perhaps true, but it's also aside and political -PFS
-
-
 In general, binding the identity of users to how they are connected to the
 Internet is not only impossible, but also undesirable.  The vast majority of
 Internet users have dynamically assigned addresses, and while they may use the
@@ -214,20 +206,6 @@
 user-controlled credential management independent of IP address.  This can help
 permit remote access without resorting to IP tunneling.
 
-%but it is only one aspect of the problem with IP-address authentication.
-%IP spoofing can bypass such simple controls. It is thus somewhat
-%ironic that the legitimate user must employ more extensive security
-%overhead to obtain authenticated authorization than the attacker.
-%Another irony is that, when IP authentication is the only available
-%means to access, legitimate users have incentive to access
-%subscriptions through the more expedient means of abusing routing
-%information. Thus attackers gain authorized access, and legitimate
-%users become attackers.
-%
-% ***Removed the above as per Scott's comment about spoofing TCP.
-% I think anything honest we could say about IP spoofing would be
-% subtle and only apply to pretty sophisticated attackers. --PFS
-
 If legitimate users of credit systems have incentive via easier authorization
 of their transactions to route from an IP address associated with their home
 location, they reveal via routing information their interactions with merchants
@@ -260,43 +238,6 @@
 traversing their wires adheres to certain normative
 requirements~\cite{palfrey06}.
 
-%Enforceability by an ISP implies that ISPs have an
-%effective way of determining
-%the human identities of all users of their networks, which disallows many
-%services and proxies.  
-%In addition, the notion
-%that an ISP can discern and actively reduce impropriety on the part of its
-%customers creates additional incentive to rely upon ``secure networks''.
-%the same myth championed by IT managers who choose to rely too heavily
-%on firewall technology.
-%Even in settings where the userbase is subject to close scrutiny and
-%background checks, good IT managers know better than to rely too
-%heavily on firewall technology.
-%In public networks, such over-reliance
-%is a misdirected solution to real problems.
-%
-% ***I was trying to fix the above paragraph and gradually came to
-% the conclusion that I don't believe it, which I was missing because
-% I do agree with its sentiment. Enforceability need not imply that
-% an IPD can determine human identities. If you could automatically
-% shut down access when certain behavior arose, you wouldn't have to
-% care who was doing it, or even if they could just login again as
-% someone else. That's just an example.  -PFS
-
-%Authors of anti-spam proposals often use the word ``commercial'' to restrict
-%the scope of the messages to which a proposed policy applies.  However, once
-%we restrict our tactics to directly fighting spammers themselves rather than
-%those who fund them, then the ``commercial'' qualification becomes irrelevant.
-%The mechanisms necessary to achieve the degree of enforceability sought
-%preclude both (a) freedom of speech qua anonymity, and (b) adherence to the
-%principles that afford the Internet its usefulness.
-%
-
-% ***And the above paragraph seems to me off to the side of our focus on
-% IP address. It perhaps fit better before some other stuff was cut.
-% Also, talking about free speech makes it
-% more ``position paperish'' --PFS
-
 \section*{Function Creep and Expedience}
 
 Making use of the routing infrastructure itself to protect participants from
@@ -388,20 +329,6 @@
 deterrence do not lead to its abandonment, then the incentives it provides
 to network attackers ultimately should.
 
-%In a sense the technology is
-%largely there
-%
-%  If the alternatives to using
-%network location information in this fashion are not just theoretically better
-%but also realistically deployable and usable, then the best time for
-%considering such alternatives might be now.
-
-% conclusions: open questions, remove / change arguments, structure
-
-% here is a potential problem emerging with this technology
-
-% unless these problems are addressed, these two technologies are headed for a
-% clash, etc.
 
 %%%%%%%%%%%%%  BIBLIOGRAPHY HERE %%%%%%%%%%%%%
 

***********************************************************************
To unsubscribe, send an e-mail to majordomo@xxxxxxxx with
unsubscribe freehaven-cvs       in the body. http://freehaven.net/