OODL SIG is Now On-Line

The following is a copy of a recent announcement sent to all OODL SIG members.

Thanks to Harvey Alcabes for setting up our administrative and Applelink group addresses, and to Steve Strassmann for generously agreeing to maintain the Internet portion of OODL SIG's membership list while the SIG is starting up.

Here is how to send messages to OODL SIG's membership.

FROM THE INTERNET:

To send messages to OODL SIG's Internet membership use:
oodl-sig-in@cambridge.apple.com

To send messages to OODL SIG's Applelink membership use:
oodl-sig@cambridge.apple.com

To send messages to OODL SIG administration use:
OODL.SIG@applelink.apple.com

For add/drop subscription requests use:
oodl-sig-request@cambridge.apple.com

FROM APPLELINK

To send messages to OODL SIG's Internet membership use:
oodl-sig-in@cambridge.apple.com@internet#

To send messages to OODL SIG's Applelink membership use:
oodl.sig$ (note the dollar sign)

To send messages to OODL SIG administration:
oodl.sig (note no dollar sign)

For add/drop subscription requests:
oodl.sig (note no dollar sign)

Watch for notices because these addresses may change.

OODL SIG has been busy this last month. Jeff Stulin has been compiling our membership list and member profiles (over 70 members currently). Kent Sandvik has been designing an electronic SIG newsletter, and Howard Oakley has started an enhanced MCL interface class library project. The SIG will soon be asking for membership help with these and many other projects but first, a little business.

OODL SIG is constructing three "foundation" documents. The first presents OODL SIG's goals, methods, and basic philosophy. The second, its operational procedures for deciding which goals to work for and the methods for achieving them. The third will be a formal (legal?) document describing OODL SIG's organizational structure.

The first document is included below for your review, the second will be sent late next week, the third, well . . . we are in no hurry for the third. Let's enjoy our informality while it lasts.

We would prefer not to "micro manage" the first two documents. They are our ideals about OODL SIG, but reality will dictate what we can do, so they need not be perfect. If you have any major philosophical problem with them, then either present it to the SIG membership, else let it be (or just send a note to the SIG administrative address).

We would like to conclude the commenting period for the first document by June 26th. Once we legitimize the foundation documents (at least the first two) we will be ready for some real work.

Thank you all for your patience while we got our act together. Here is the first foundation document:

Statement

This document describes the definition, purpose, scope, goals, and methods of a new organization hereafter known as OODL-SIG.

Definition

An OODL, Object Oriented Dynamic Language, is any computer programming language which strongly supports both object oriented programming and dynamic runtime variable typing. It is also expected, as a practical matter, that an OODL development system will support quick turn around via incremental compilation and linking. Examples of OODLs are Common Lisp with the Common Lisp Object System (CLOS), Dylan, and Smalltalk.

Comment

We will improve this definition later, let's not have a big discussion about it now.

OODL-SIG purpose

It is not an exaggeration to say that the physical and financial health of our society, as well as the individuals within it, depends, to an increasing extent, on the effectiveness of computer software. Thus, the founders of the OODL-SIG, have a personal as well as professional interest in the improvement of software development technology. We know well what is at stake, and are not satisfied with the current trends in software development.

We believe that the OODL approach to software development can greatly improve the timeliness, cost effectiveness, flexibility, and reliability of software. We recognize that historically, due to performance concerns, it has not been practical to use OODLs for "shrink-wrapped" application delivery. We believe, however, that hardware and compiler technology have advanced to the degree that, contrary to general opinion, OODLs will soon be the best tools for development and delivery of most applications software products.

We form OODL- SIG in order to promote and advance the OODL method of software development. To this goal OODL-SIG will be primarily an evangelical, information-exchange, and education organization. In particular it will: promote the use of OODLs to the general programmer population, assist current OODL users in becoming as productive as possible, assist beginning OODLers in their learning process, and encourage the creation of OODL development environments which can deliver high performance "shrink wrapped" applications.

Scope

The SIG is open to users of all OODLs, and to all OODL development platforms.

Comment

As a practical matter, since our initial membership is Macintosh based, our early efforts will be focused on Macintosh based products. Membership demand will probably change this.

Goals and Methodologies

Below are OODL-SIGs goals and suggested methods which I list in no particular order.

Goal: To promote the use of OODLs to the general programmer population.

Most programmers have seen OODLs only in a scholastic setting and are under the impression that they are "weird", perhaps interesting, but worthless for real world applications. We must change this perception.

Methods

To be effective we must become credible. Here's how:

• By becoming a community
  We become a community by having an active SIG, by publishing a newsletter, by having meetings, by having the membership participate in group projects, and by having a group address for communications.
• By the publication of articles
  Education is the way to change how developers think about OODLs. We educate by publication. We publish in Frameworks, in the OODL-SIG newsletter, in Byte magazine, and in any respected publication that we can get a piece about OODLs or about our organization published. We must actively seek out those members of our organization capable of creating accurate, responsible, and interesting articles, and encourage them to publish.
• By Example
  The best way to convince developers that OODLs are for real is by example; that is by demonstrating real world programs. We must survey our membership for examples of programs that are under active use, collect these in a case book, and promote the existence of these programs.
• By Creation of a sample program
  The SIG should have a rule that for an OODL to be taken seriously, a sophisticated real world program developed with that OODL, including source code, must be made available to OODL-SIG and its members.
• By being in the news
  The SIG should actively seek publicity for itself, for OODLs, and for programs developed using OODLs.
• By creating an "Idealized OODL Environment" description.
  One clever, inexpensive way to change developers perceptions of OODLs would be to create, as a challenge, an "Idealized OODL Environment" document, to which all SIG promoted OODLs would be compared.

This document would say nothing about any particular OODL language. Rather, it would describe things like delivered application performance, user interface development, cross platform availability, turn around time, hardware requirements, documentation quality, etc.

As an example consider MCL. I think that most programmers unacquainted with MCL would be shocked to learn about its fine real world performance, instant turn around, and modest hardware requirements.

The document will also be a good vehicle to let OODL system designers know what is most important to us.

Goal: Encourage the creation of OODL development environments which can deliver high performance "shrink wrapped" applications.

In my opinion there is not yet any OODL that is practical for delivering "shrink wrapped" applications. I would be surprised, however, if there is not at least one within two years.

Methods

We must be vocal in telling OODL manufacturers what we need. To do this we must agree on what we want, and speak out with a unified voice.

As an example consider the responses to the MCL group's call for MCL 2.1 features. If we as a body could agree on a prioritized list of new features, they would probably take us very seriously.

Goal: To assist current OODL users in becoming as productive as possible.

We probably can not improve on the individual technical assistance offered by the MCL group or through MacAppTech$, but we can help with the:

Methods

• Publication of detailed technical articles
  As an example consider how much benefit could be accomplished by an article on maximizing MCL performance (e.g. declarations, in line defuns, macros etc.).
• Compilation of technical responses
  The MCL discussion group is great, but once the messages are sent they are history. We need a "technical librarian" who would edit and maintain a list of technical questions and the responses. I do not volunteer.
• General dissemination of information
  Is there a browser for MCL? I bet the answer is several, but I'll be damned if I know where to get one. OODL-SIG could be a repository for information about things like this.

Goal: To assist beginning OODLers in their learning process.

I think this is the area in which the SIG can have its greatest impact.

Methods

• Publication of tutorial articles and sample programs.
  Whenever we can, anywhere that we can.
• Regional OODL introduction seminars.
  It would not be expensive to create a canned "Introduction to OODLs" seminar and to present it, on a regular basis, at various locations world-wide.
• Creation of an "Introduction to OODLs" document.
  The SIG should create an honest document detailing what OODLs are, what their advantages and disadvantages are, examples of the successful use of OODLs, road maps to leaning various OODL environments, and realistic OODL leering curve estimates. I think this should be a #1 OODL-SIG priority.