MIME-Version: 1.0
In-Reply-To: <CABkgnnXv6KtUSEj_+rNiPTLd78QX+M0L5k_2ipfSCjnbmp_o7Q@mail.gmail.com>
References: <CABkgnnUu6aE0socrxXm6L11T5F0cdHL-Y5K0deQudOorwEeVqg@mail.gmail.com>
 <r470Ps-10132i-E0E190ABCD214523B790DE7F83C37914@Williams-MacBook-Pro.local>
 <CABkgnnXv6KtUSEj_+rNiPTLd78QX+M0L5k_2ipfSCjnbmp_o7Q@mail.gmail.com>
From: Victor Vasiliev <vasilvv@google.com>
Date: Tue, 12 Dec 2017 19:40:59 -0500
Message-ID: <CAAZdMacrTJPhsjTv0+gFNwmhVTE02stY55uE4Vvpf9kRChWqkg@mail.gmail.com>
To: Martin Thomson <martin.thomson@gmail.com>
Cc: Bill Frantz <frantz@pwpconsult.com>, "tls@ietf.org" <tls@ietf.org>
Content-Type: multipart/alternative; boundary="001a113e317c7b7a7c05602e05e0"
Archived-At: <https://mailarchive.ietf.org/arch/msg/tls/KxD9DhZaCBi4pVOy_0OQ0K1qQ9k>
Subject: Re: [TLS] I-D Action: draft-ietf-tls-certificate-compression-01.txt
Precedence: list

--001a113e317c7b7a7c05602e05e0
Content-Type: text/plain; charset="UTF-8"

On Mon, Dec 11, 2017 at 6:49 PM, Martin Thomson <martin.thomson@gmail.com>
wrote:

> Certificates are pretty wasteful, outside of the keys themselves.
> There has to be some significant gains to be had.  I think that we
> have discussed generating a dictionary that would be useful for
> certificates, so if we do that we won't know the full answer yet (I
> see no mention of that in the draft, so I guess that I might be in
> dreamland).


Indeed.  I've presented some numbers on this back in Chicago:


https://datatracker.ietf.org/meeting/98/materials/slides-98-tls-certificare-compression/

There is currently no pre-shared dictionary in the draft, since deciding
what to put into that dictionary is somewhat of a hard question (both
from the technical and from the ecosystem perspective).  I'm still
working on making one, but the current plan is to not block the draft on
this, since the simple scheme is already quite effective, and adding it
is a matter of adding another compression algorithm to the list.

--001a113e317c7b7a7c05602e05e0
Content-Type: text/html; charset="UTF-8"
Content-Transfer-Encoding: quoted-printable

<div dir=3D"ltr"><div class=3D"gmail_extra"><div class=3D"gmail_quote">On M=
on, Dec 11, 2017 at 6:49 PM, Martin Thomson <span dir=3D"ltr">&lt;<a href=
=3D"mailto:martin.thomson@gmail.com" target=3D"_blank">martin.thomson@gmail=
.com</a>&gt;</span> wrote:<br><blockquote class=3D"gmail_quote" style=3D"ma=
rgin:0px 0px 0px 0.8ex;border-left:1px solid rgb(204,204,204);padding-left:=
1ex">Certificates are pretty wasteful, outside of the keys themselves.<br>
There has to be some significant gains to be had.=C2=A0 I think that we<br>
have discussed generating a dictionary that would be useful for<br>
certificates, so if we do that we won&#39;t know the full answer yet (I<br>
see no mention of that in the draft, so I guess that I might be in<br>
dreamland).</blockquote><div><br></div><div>Indeed.=C2=A0 I&#39;ve presente=
d some numbers on this back in Chicago:</div><div><br></div><div>=C2=A0 <a =
href=3D"https://datatracker.ietf.org/meeting/98/materials/slides-98-tls-cer=
tificare-compression/">https://datatracker.ietf.org/meeting/98/materials/sl=
ides-98-tls-certificare-compression/</a></div><div><br></div><div><div>Ther=
e is currently no pre-shared dictionary in the draft, since deciding</div><=
div>what to put into that dictionary is somewhat of a hard question (both</=
div><div>from the technical and from the ecosystem perspective).=C2=A0 I=
9;m still</div><div>working on making one, but the current plan is to not b=
lock the draft on</div><div>this, since the simple scheme is already quite =
effective, and adding it</div><div>is a matter of adding another compressio=
n algorithm to the list.</div></div></div></div></div>

--001a113e317c7b7a7c05602e05e0--