[Aggsrv] updated charter proposal

Peter Saint-Andre <stpeter@stpeter.im> Fri, 22 February 2013 18:44 UTC

Return-Path: <stpeter@stpeter.im>
X-Original-To: aggsrv@ietfa.amsl.com
Delivered-To: aggsrv@ietfa.amsl.com
Received: from localhost (localhost []) by ietfa.amsl.com (Postfix) with ESMTP id B63E721F8A90 for <aggsrv@ietfa.amsl.com>; Fri, 22 Feb 2013 10:44:51 -0800 (PST)
X-Virus-Scanned: amavisd-new at amsl.com
X-Spam-Flag: NO
X-Spam-Score: -102.534
X-Spam-Status: No, score=-102.534 tagged_above=-999 required=5 tests=[AWL=0.065, BAYES_00=-2.599, USER_IN_WHITELIST=-100]
Received: from mail.ietf.org ([]) by localhost (ietfa.amsl.com []) (amavisd-new, port 10024) with ESMTP id UaC4ekKnruwC for <aggsrv@ietfa.amsl.com>; Fri, 22 Feb 2013 10:44:50 -0800 (PST)
Received: from stpeter.im (mailhost.stpeter.im []) by ietfa.amsl.com (Postfix) with ESMTP id 9B81B21F89A3 for <aggsrv@ietf.org>; Fri, 22 Feb 2013 10:44:50 -0800 (PST)
Received: from [] (unknown []) (Authenticated sender: stpeter) by stpeter.im (Postfix) with ESMTPSA id 86710403CD for <aggsrv@ietf.org>; Fri, 22 Feb 2013 11:52:24 -0700 (MST)
Message-ID: <5127BC9B.8000203@stpeter.im>
Date: Fri, 22 Feb 2013 11:44:43 -0700
From: Peter Saint-Andre <stpeter@stpeter.im>
User-Agent: Mozilla/5.0 (Macintosh; Intel Mac OS X 10.7; rv:17.0) Gecko/20130216 Thunderbird/17.0.3
MIME-Version: 1.0
To: aggsrv@ietf.org
X-Enigmail-Version: 1.5
Content-Type: text/plain; charset="UTF-8"
Content-Transfer-Encoding: 7bit
Subject: [Aggsrv] updated charter proposal
X-BeenThere: aggsrv@ietf.org
X-Mailman-Version: 2.1.12
Precedence: list
List-Id: "Aggregated Service Discovery \(aggsrv\)" <aggsrv.ietf.org>
List-Unsubscribe: <https://www.ietf.org/mailman/options/aggsrv>, <mailto:aggsrv-request@ietf.org?subject=unsubscribe>
List-Archive: <http://www.ietf.org/mail-archive/web/aggsrv>
List-Post: <mailto:aggsrv@ietf.org>
List-Help: <mailto:aggsrv-request@ietf.org?subject=help>
List-Subscribe: <https://www.ietf.org/mailman/listinfo/aggsrv>, <mailto:aggsrv-request@ietf.org?subject=subscribe>
X-List-Received-Date: Fri, 22 Feb 2013 18:44:51 -0000

Hash: SHA1

The proponents have sent me an updated charter for the aggsrv
initiative. The text can be found below, and I've also updated
http://trac.tools.ietf.org/wg/appsawg/trac/wiki/AggSrv accordingly.


Aggregated Service Discovery Working Group

Problem Statement:

Service providers and enterprises commonly offer a variety of
application services delivered over multiple protocols.  A user will
often consume these services from several endpoints, requiring service
discovery or manual configuration for each service at each endpoint.
Some of these services leverage existing standards-based discovery,
such as DNS, DHCP, or UDDI, while others may rely on some form of
proprietary discovery.  Still others may not support any form of
discovery, requiring the manual entry of service access information.
As the quantity and variety of these services grow, it becomes
increasingly onerous for administrators to manage the disparate
discovery mechanisms, and burdensome on users to provide configuration
where discovery is not supported.

It is useful to first consider whether the issues described here might
be addressable through the direct use or extension of existing
discovery protocols.  DHCP [1], for example, is widely deployed and
supports extension for the discovery of new types of services.  Many
DHCP extensions already exist for the discovery of such services as
DNS, NTP, NIS, LDAP, and others.  The DHCP protocol, however, is
limited by a dependence on local network broadcast, lack of support
for structured data, and an extension mechanism not intended for
unregistered customization.  This, coupled with a relative dearth of
application-level APIs supporting DHCP service-specific extensions,
makes DHCP an unlikely solution for the issues we face here.

Another option would be to leverage DNS through DNS-SD [2].  The DNS-SD
extension is expressly designed to support Internet-scale service
discovery using standard DNS queries and existing record types.
However, it remains a significant limitation that the discovered data
is global and cannot be made a function of information provided in the
request.  For example, an enterprise with several IMAP servers, each
servicing the same email domain but hosting different subsets of
employees, could not employ DNS-SD for email service discovery using
that one domain.  It is also important to consider that we wish to
establish a solution that is broadly and uniformly usable across a
wide array of platforms and environments.  It is an unfortunate fact
that browser-based clients often lack the necessary APIs and trust
necessary to make explicit DNS queries for particular types of records.

In terms of new ideas, similar discovery standardization efforts
already under way include WebFinger [3], which is intended to address
generalized discovery for any type of URI identifiable resource, and
Simple Web Discovery [4], which is more specifically related to the
discovery of services.  The former provides an interesting framework
within which aggregated service discovery could operate, however by
itself there is insufficient guidance and structure to address the
specific needs of service discovery use cases.  Simple Web Discovery,
on the other hand, while expressly intended for the discovery of
services, tends to focus specifically on service location and is not
well suited for aggregate discovery of dissimilar service types.


The Aggregated Service Discovery working group will standardize an
extensible protocol supporting the discovery of multiple services with
a single operation and with limited initial information (e.g. a
well-known URL, or email address).  A central objective shall be the
establishment of a protocol and message format which may be readily
adopted by a wide variety of endpoints, minimizes client startup time
by reducing network roundtrips, and takes into consideration the
various technical challenges faced by clients in the wild, including
security, firewalls, same-origin policies and limited platform APIs.
Equally important, the working group will focus on ease of deployment,
and support for both standard and non-standard services types.  The
barriers to adoption should be made as low as possible without
compromising the security and integrity of the discovery process.

In the interest of meeting the above objectives, this working group
will develop a core message format based on JSON, and protocol based on
HTTP and REST (using [5] as the starting point).  Initially, the group
will focus on a draft defining an extensible aggregated discovery
document structure, and operations for discovery document retrieval.
The group will not necessarily define new formats and protocols, and
will consider existing technologies where possible.

Potential follow up work may include an additional draft for defining
a complimentary protocol for local area network service discovery.
This draft would define a means by which aggregated discovery
documents may be obtained by clients in a fully automated manner,
possibly based on mDNS or DHCP.  However, the group would need to
recharter in order to add such a draft to its deliverables.


Jun 2013 - Accept protocol and format document as Working Group item
Oct 2013 - Start Working Group Last Call on protocol and format document
Dec 2013 - Submit protocol and format document to IESG


[1] RFC 2131 - DHCP
[2] RFC 6763 - DNS-SD
[3] draft-ietf-appsawg-webfinger
[4] draft-jones-simple-web-discovery
[5] draft-daboo-aggregated-service-discovery



- -- 
Peter Saint-Andre

Version: GnuPG/MacGPG2 v2.0.18 (Darwin)
Comment: Using GnuPG with Thunderbird - http://www.enigmail.net/