Re: Report on preliminary decision on TLS 1.3 and client auth

Kyle Rose <> Thu, 22 October 2015 23:47 UTC

Return-Path: <>
Received: from localhost ( []) by (Postfix) with ESMTP id 96E471A8F4E for <>; Thu, 22 Oct 2015 16:47:52 -0700 (PDT)
X-Virus-Scanned: amavisd-new at
X-Spam-Flag: NO
X-Spam-Score: -6.389
X-Spam-Status: No, score=-6.389 tagged_above=-999 required=5 tests=[BAYES_00=-1.9, DKIM_SIGNED=0.1, DKIM_VALID=-0.1, DKIM_VALID_AU=-0.1, FM_FORGED_GMAIL=0.622, HTML_MESSAGE=0.001, RCVD_IN_DNSWL_HI=-5, SPF_HELO_PASS=-0.001, SPF_PASS=-0.001, T_RP_MATCHES_RCVD=-0.01] autolearn=ham
Received: from ([]) by localhost ( []) (amavisd-new, port 10024) with ESMTP id nltCEkR2yQpK for <>; Thu, 22 Oct 2015 16:47:50 -0700 (PDT)
Received: from ( []) (using TLSv1.2 with cipher DHE-RSA-AES128-SHA (128/128 bits)) (No client certificate requested) by (Postfix) with ESMTPS id 28B011A8ADD for <>; Thu, 22 Oct 2015 16:47:49 -0700 (PDT)
Received: from lists by with local (Exim 4.80) (envelope-from <>) id 1ZpPWd-0007Ii-9F for; Thu, 22 Oct 2015 23:44:03 +0000
Resent-Date: Thu, 22 Oct 2015 23:44:03 +0000
Resent-Message-Id: <>
Received: from ([]) by with esmtps (TLS1.2:DHE_RSA_AES_128_CBC_SHA1:128) (Exim 4.80) (envelope-from <>) id 1ZpPWY-0007Hn-R3 for; Thu, 22 Oct 2015 23:43:58 +0000
Received: from ([]) by with esmtps (TLS1.2:RSA_ARCFOUR_SHA1:128) (Exim 4.80) (envelope-from <>) id 1ZpPWX-0002Bp-4M for; Thu, 22 Oct 2015 23:43:58 +0000
Received: by ioll68 with SMTP id l68so108950668iol.3 for <>; Thu, 22 Oct 2015 16:43:29 -0700 (PDT)
DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed;; s=google; h=mime-version:in-reply-to:references:date:message-id:subject:from:to :cc:content-type; bh=elBMhIo/qNUcEtTLtvfZ710RK7YjrF0M9PoivIqYXMI=; b=ojZcNAd4rYal0qDkE51PNMi4FqeGjaljmJJlHr0IgUXHwZ9es5IGJYWUYOkCG1TbaG yNPEnVDiHjWE8ghUby1plloXs9Si3TcBlW/6BMGBx0YTys9+rgSA/FGFMQyxIQHQiQo/ O4FYxWetlu2xTNVhcnXM4NSQ2v9iiaphydk+o=
X-Google-DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed;; s=20130820; h=x-gm-message-state:mime-version:in-reply-to:references:date :message-id:subject:from:to:cc:content-type; bh=elBMhIo/qNUcEtTLtvfZ710RK7YjrF0M9PoivIqYXMI=; b=OM9DfvpdSruKdwKe9WTSsfP47mmvD8yweraZdMiGGpVgQJQz+HFX8hCb8PETJV8Ihh 5qZW2mGOBO1Cuj15oQVLg/g4DkMwLgkwWFIPdv5FI/lhWwbsoqdEFngqWrhpN5V48VjG JmC/F1vanTAZwOEygzDOR7h3NUGlxRMsLKcRBLhLP/MYvA/HPJUAADqBlCKj7ZhYnSNV PZvTqX9hdI/Ak1qISqbrxusaswXa5/eyJLpKr8u8Drhmpjga1I4yHP10IYqumjueCm+b +gd0YVLu2a031wb7hg2FhSQcJaphtU932IGJ8s1c9HCaNsstUsuZyaxMY+Yy5G56c0fg su4Q==
X-Gm-Message-State: ALoCoQlDr25DTGb4mgAgJvdRyXQPA93yui0tK7FxBZnXyli0fmDA1eUSX7Fl66T98fszyeddLByT
MIME-Version: 1.0
X-Received: by with SMTP id l201mr18948663iol.156.1445557408908; Thu, 22 Oct 2015 16:43:28 -0700 (PDT)
Received: by with HTTP; Thu, 22 Oct 2015 16:43:28 -0700 (PDT)
X-Originating-IP: []
In-Reply-To: <>
References: <> <> <>
Date: Thu, 22 Oct 2015 19:43:28 -0400
Message-ID: <>
From: Kyle Rose <>
To: Martin Thomson <>
Cc: HTTP Working Group <>
Content-Type: multipart/alternative; boundary="001a11405de2dc628b0522ba0f41"
Received-SPF: pass client-ip=;;
X-W3C-Hub-Spam-Status: No, score=-4.7
X-W3C-Hub-Spam-Report: BAYES_00=-1.9, DKIM_SIGNED=0.1, DKIM_VALID=-0.1, DKIM_VALID_AU=-0.1, HTML_MESSAGE=0.001, RCVD_IN_DNSWL_LOW=-0.7, SPF_PASS=-0.001, W3C_AA=-1, W3C_WL=-1
X-W3C-Scan-Sig: 1ZpPWX-0002Bp-4M 87148de5f9f2c5c46430167a2b43ff8f
Subject: Re: Report on preliminary decision on TLS 1.3 and client auth
Archived-At: <>
X-Mailing-List: <> archive/latest/30398
Precedence: list
List-Id: <>
List-Help: <>
List-Post: <>
List-Unsubscribe: <>

> I wouldn't interpret this as a defense of the client certificate UX in
> browsers.  But I don't expect that to change significantly, our UX
> people have a lot of work to do, most of it much more important than
> this.

I wasn't even actually talking about the browser UI (though I guess I would
like Firefox to actually "remember this decision" for client certificates,
which it doesn't seem to do even when I check that box). I'm more talking
about the UX suggested by your first paragraph, in which the server accepts
the handshake and provides a better error. Given the solutions proposed to
the client authentication problem, I suspect that's what we'll end up
doing, bugs in application authorization logic be damned. Using client
certs as a firewall for permission-to-talk does seem like a hack: having a
simple TCB up to the point of client authentication seems like a better
solution all around.

Tl;dr: don't interpret my previous email as a defense of that use case.