OMG-DARPA Workshop on Compositional Software Architectures
Marriott Hotel, Monterey, California
January 6-8, 1998
http://www.objs.com/workshops/ws9801/cfp.htm
Sponsors
========
* The Object Management Group (OMG), the software
industry's largest consortium, is focused on open
interoperable componentware framework interface
technology.
* The Defense Advanced Research Projects Agency (DARPA)
funds revolutionary software research in the U.S.
Objectives of the Workshop
==========================
The workshop will focus on the interactions between:
* componentware software architectures
* web and distributed object architectures
Fundamental concerns face organizations developing and
maintaining large, enterprise-critical, distributed applications.
* Application development teams spend too much time coping
with the complexities of their chosen middleware, services,
tools, web, and programming environments. An application's
choices of underlying middleware products become pervasive
and irrevocable commitments of the application.
* Complex distributed application logic must be delivered
via the web, and today there are object model and service
architecture mismatches.
* The goal of assembling applications from reusable
components is still elusive because business applications
require system-wide properties like reliability,
availability, maintainability, security, responsiveness,
manageability, and scalability (the "ilities").
Assembling components and also achieving system-wide
qualities is still an unsolved problem. As long as the
code that implements ilities has to be tightly interwoven
with code that supports business logic, new applications
are destined to rapidly become as difficult to maintain
as legacy code.
Componentware did not exactly set the world on fire five years
ago. Now we have new languages, maturing visions of compositional
architectures (CORBA, WWW, ActiveX, ....), the web as a
distributed system with a low-entry barrier, and emerging
middleware service architectures. Do we have the critical mass
to jump start the component-based software cottage industry?
Even if the technology enablers are there, what is needed to
establish an effective componentware market? What are the
remaining barriers?
The workshop is intended to bring together a mix of leading
industry, government, and university software architects,
componentware framework developers, researchers, standards
develpers, vendors, and customers to do the following:
* better understand the state of practice of industry
componentware initiatives (ActiveX, OMG's OMA/CORBA,
Java, W3C) and how far they go in solving problems of
composability and plug-and-play.
* better understand how software architectures play a
role in integrating web and object service architectures
and in building systems that can maintain architectural
properties (e.g., composability, scalability,
evolvability, debugability).
* identify key technologies, directions and convergence
approaches and characterize open research problems and
missing architectural notions.
Structure and Content of the Workshop
=====================================
The workshop will consist of a set of invited presentations
and topic-centered breakout sessions.
Topics of interest (but not limited to):
* State of practice in componentware and software
architecture - e.g., views from Microsoft, Netscape,
JavaSoft, OMG/ODP, and the software architecture R&D community.
* State of practice in web + distributed object
integration - e.g., views from Netscape, Visigenc, Iona,
JavaSoft, W3C, Microsoft, web-objects architecture R&D.
* Characterizing the problem. What do large application
developers and enterprise software architects want?
How do they avoid building more unmaintainable legacy
applications? How do they build applications with fifteen
year life cycles on middleware products that change annually?
How do they architect systems so that both functionality and
architectural -ilities can be upgraded over the application's
life cycle? Approaches to evolution of software.
* Composing components. What are examples of minimal
common infrastructures that enable component
composition. Are we there yet with respect to plug
and play? Problems with componentware approaches
(devils' advocate positions) and solution approaches
(counter arguments) - e.g., footprint, too many interfaces,
uncontrolled evolution. Economy of componentware.
Is componentware a silver bullet or a nightmare or
yet-another-technology?
* Composing object services. How can we compose object
services? could you make a competitive OODB from naming,
persistence, transactions, queries, etc.? Implicit interfaces
and wrappers. What behavioral extensions can be added
implicitly to a system? Mechanisms like POA, interceptors,
before-after rules to guard objects to insure they are acted on
by implicit operations. Terminology - e.g., loose or tight
coupling, granularity, frameworks.
* Architectural properties. What are ilities, i.e., some
property added to a system that is independent of the
functionality of that system. How to insert them into
componentware architectures? Say you had a system doing
something. How would it look different if ility X was added
or removed? Is there some kind of architectural level where
the boundary between the visibility/hiddeness of the ility
changes? What is needed in the architecture in order to add
ilities?
* Scaling componentware architectures. Frameworks, patterns,
configurations, inter component protocols. Examples of
composition involving heterogeneous data sources.
Federation - do we have to federate the services when we
have ORBs on 40,000,000 desktops? what can we say about the
federation pattern? end-to-end, top-to-bottom
ilities like optimization, QoS, security, ...
* Adaptivity of componentware architectures. Tailorability,
evolvability, assured services and graceful degradation,
survivability.
* Web object models, metadata and registry/repository
in Internet/Web. How do DOM, XML, PICS-NG, RDF, and the
many metadata proposals relate to object and agent
definition languages?
* Convergence of ORB and Web architectures. (Why) are
both camps doing the same things differently? How to avoid
recreating the same set of services like versioning on both
architectures.
Outcomes
========
* Explicit outcome - a workshop report to be available
on the web soon after the meeting. The workshop report
will include:
o position papers
o workshop breakout session reports
o summary of research issues, proposed architectural
frameworks, key research directions, i.e., key
conclusions in a nutshell.
* Implicit outcome - possible direction changes and
convergence among technologies as different groups
(e.g., OMG, W3C, ODP, IETF) understand what others
bring to the party.
Position Papers
===============
Position papers (around three pages) on a topic related to
the workshop theme should be sent to Craig Thompson
(thompson@objs.com) by November 14, 1997. Preferred format is
HTML but .txt, .rtf, .ps., .pdf, or .doc are acceptable.
All position papers will be posted on the web.
The workshop committee will announce accepted papers on or
before December 7, 1997.
Logistics
=========
The workshop will be limited to around 60 participants.
The workshop will start at 8 a.m. on January 6 and end at
noon on January 8. Breakfast will be provided on days 1, 2
and 3; lunch on days 1 and 2; cash bar reception on day 1.
The meeting hotel is the Monterey Marriott, located two
minutes' walk from Monterey Bay Fisherman's Wharf, at 350
Calle Principal at Del Monte Blvd., Monterey, CA 93940,
(408) 649-4234, http://www.marriott.com/marriott/MRYCA/index.htm.
The room block under the name "Compositional Software
Architecture Workshop" will be held until December 15, 1997.
Negotiated rates are $119/night (up to 15 reservations
at Government per diem of $71).
The workshop fee is $225 payable by check or credit card
to MCC. Participants can reserve hotel rooms and pre-register
by completing the attached registration form and sending it
to Felisa Legaspi, MCC/WCL, 2099 Gateway Place, Suite #450,
San Jose, CA 95110. For questions, contact Felisa Legaspi at
(408) 573-4130-tel, (408) 573-4140-fax, legaspi@mcc.com.
Directions: From the Monterey Airport/Hwy. 68: Take the
Monterey Fisherman's Wharf exit. At the first stoplight,
turn right on Aguajito. Continue on Aguajito until it ends
at Del Monte. Turn left on Del Monte and continue straight
for three stoplights. At the third stoplight, get in the left
turn lane and continue straight on Del Monte to the hotel.
Taxi from Monterey Airport is $8. Valet parking is $12/day.
There is a public garage 1-2 block from the hotel.
Important Dates
===============
* November 12, 1997 - position papers due
* November 14, 1997 - submitted position papers posted on the web
* December 7 - announce accepted papers (on or before)
* December 15, 1997 - last date for room block reservations
Workshop Committee
==================
* Randy Garrett, DARPA
* Barry Leiner, consultant
* Ted Linden, MCC
* Andry Rakotonirainy, DSTC
* Richard Soley, OMG (chair)
* Craig Thompson, OBJS
* Gio Wiederhold, Stanford
Organizers
==========
* Ted Linden, Microelectronics and Computer Technology
Corporation (MCC)
* Craig Thompson, Object Services and Consulting, Inc.