Re: [http-state] Ticket 3: Public Suffixes

Adam Barth <ietf@adambarth.com> Sat, 16 January 2010 22:16 UTC

Return-Path: <adam@adambarth.com>
X-Original-To: http-state@core3.amsl.com
Delivered-To: http-state@core3.amsl.com
Received: from localhost (localhost [127.0.0.1]) by core3.amsl.com (Postfix) with ESMTP id AC0F13A67FD for <http-state@core3.amsl.com>; Sat, 16 Jan 2010 14:16:58 -0800 (PST)
X-Virus-Scanned: amavisd-new at amsl.com
X-Spam-Flag: NO
X-Spam-Score: -1.94
X-Spam-Level:
X-Spam-Status: No, score=-1.94 tagged_above=-999 required=5 tests=[AWL=0.038, BAYES_00=-2.599, FM_FORGED_GMAIL=0.622]
Received: from mail.ietf.org ([64.170.98.32]) by localhost (core3.amsl.com [127.0.0.1]) (amavisd-new, port 10024) with ESMTP id 4PKor7Yd7aCu for <http-state@core3.amsl.com>; Sat, 16 Jan 2010 14:16:57 -0800 (PST)
Received: from mail-px0-f186.google.com (mail-px0-f186.google.com [209.85.216.186]) by core3.amsl.com (Postfix) with ESMTP id E433D3A67E7 for <http-state@ietf.org>; Sat, 16 Jan 2010 14:16:57 -0800 (PST)
Received: by pxi16 with SMTP id 16so1327801pxi.29 for <http-state@ietf.org>; Sat, 16 Jan 2010 14:16:52 -0800 (PST)
MIME-Version: 1.0
Received: by 10.142.2.24 with SMTP id 24mr2841322wfb.139.1263680212184; Sat, 16 Jan 2010 14:16:52 -0800 (PST)
In-Reply-To: <op.u6m3g1jnqrq7tp@acorna.invalid.invalid>
References: <7789133a1001160001h62d203b3w76e175ec22d55e6@mail.gmail.com> <op.u6mioszyqrq7tp@acorna.invalid.invalid> <7789133a1001161018h4d70b5cal6edfa3978c478fe4@mail.gmail.com> <op.u6m3g1jnqrq7tp@acorna.invalid.invalid>
From: Adam Barth <ietf@adambarth.com>
Date: Sat, 16 Jan 2010 14:16:32 -0800
Message-ID: <7789133a1001161416q5cade775r313cb8325862fc88@mail.gmail.com>
To: "Yngve N. Pettersen (Developer Opera Software ASA)" <yngve@opera.com>
Content-Type: text/plain; charset="ISO-8859-1"
Content-Transfer-Encoding: quoted-printable
Cc: http-state <http-state@ietf.org>
Subject: Re: [http-state] Ticket 3: Public Suffixes
X-BeenThere: http-state@ietf.org
X-Mailman-Version: 2.1.9
Precedence: list
List-Id: Discuss HTTP State Management Mechanism <http-state.ietf.org>
List-Unsubscribe: <https://www.ietf.org/mailman/listinfo/http-state>, <mailto:http-state-request@ietf.org?subject=unsubscribe>
List-Archive: <http://www.ietf.org/mail-archive/web/http-state>
List-Post: <mailto:http-state@ietf.org>
List-Help: <mailto:http-state-request@ietf.org?subject=help>
List-Subscribe: <https://www.ietf.org/mailman/listinfo/http-state>, <mailto:http-state-request@ietf.org?subject=subscribe>
X-List-Received-Date: Sat, 16 Jan 2010 22:16:58 -0000

On Sat, Jan 16, 2010 at 11:08 AM, Yngve N. Pettersen (Developer Opera
Software ASA) <yngve@opera.com> wrote:
> On Sat, 16 Jan 2010 19:18:48 +0100, Adam Barth <ietf@adambarth.com> wrote:
>
>> On Sat, Jan 16, 2010 at 3:39 AM, Yngve N. Pettersen (Developer Opera
>> Software ASA) <yngve@opera.com> wrote:
>>>
>>> Alternative 4: Use a DNS based heuristic by only allowing cookies set to
>>> domains that have an IP address defined. If there is no IP address,
>>> remove
>>> the domain attribute
>>
>> One problem with this approach is that a number of major ISPs,
>> including COMCAST, return IP addresses for all non-existent domains.
>
> I know, and it could become a signifcant problem with the system, depending
> on how those systems work.
>
> The question is though: Do they do this for domain.com if www.domain.com
> exists?
>
> I know ISPs and others do this for unknown domains, but do they do it for
> registered domains, or the public suffixes that actually have domains under
> them?
>
> The situation in which the heuristics will operate is that
> server.domain1.domain2.tld tries to set a cookie for domain2.tld. Would a
> ISP highjack the name domain2.tld to display ads, even if domain2.tld is a
> public suffix?

I don't know, and I expect it varies from ISP to ISP.  In general,
though, whether or not a domain resolves is a fragile heuristic.
These folks have strong business incentives for creating these
records.  Even if their current behavior is ok, we don't have any
guarantees that it will remain so.

>>> Alternative 5: (which require an extensive change of the cookie
>>> specification)
>>
>> As Daniel Stenberg says, extensive changes to the spec are off the
>> table in phase one.  However, we should revisit this approach in phase
>> two.
>
> Could elements of the cookie-v2 spec be used? For example to initiate a move
> towards a better domain designation method?

I don't think it's a good idea to defocus phase 1 by mixing in ideas
for changing the protocol.  The charter specifically forbids us from
introducing new syntax or semantics.

Adam