KEYNOTE ADDRESS BY THE HONORABLE EMMETT PAIGE, JR. ASSISTANT SECRETARY OF DEFENSE COMMAND, CONTROL, COMMUNICATIONS AND INTELLIGENCE TO THE TRI-Ada '94 CONFERENCE BALTIMORE, MARYLAND NOVEMBER 8, 1994 Good Morning, It's nice to be speaking to a group of professionals -- some who share my appreciation for the benefits to be gained through the use of the Ada programming language, and some with lingering doubts. This is an exciting time in the evolution of the Ada programming language. The new version, Ada 9X, will be here soon, and there are indications of growing interest in its use within the Federal, industrial, and academic sectors. My commitment to Ada is based on facts as I see them from my perspective. The majority of the cost to DoD for software is in the maintenance phase. We have known software maintenance would be our biggest challenge for at least a couple of decades. We developed Ada to address these challenges. The facts indicate that Ada reduces maintenance costs, while increasing productivity, quality, and most importantly reliability. The payoff is there, and we are reaping it where Ada has been used. Based on hard data, there is no doubt that Ada has been a wise choice. It continues to make good business and technical sense for the Department of Defense as well as elsewhere. When I spoke to some of you at the Washington Ada Symposium, I stated the DoD couldn't afford to continue to write systems in 44 languages. The number "44" was based on the number of languages being used within just one of our fee-for-service software development activities. When asked, "Why 44 languages?" the managers will tell me that they are responding to their customers' demands. When I asked them, "What's your preference relative to languages?" They answered that their preference would be to use a single language. So it's a mixed bag. For your information, the DoD instruction on programming languages provides a list of approved high order languages. I assure you it doesn't contain 44 languages. Being a responsive person myself, I can understand the managers wanting to be responsive to their customers. However -- and this is a big however -- as this administration's official designated to be the policeman for defense software systems, I assure you that I cannot and will not condone the continuation of programming in languages that are -- at a minimum -- in violation of defense policy, and are costly to the taxpayers, and in the long run doing our customers a disservice. After the Washington Symposium, I tasked the Ada Joint Program Office to conduct a language survey to determine just how many languages are currently used in fielded DoD systems. I want to know if there has been a reduction or increase since the late 1970's. I should be getting the final results of that survey by the end of the year. However, I will share with you some preliminary findings. The fielded systems we have looked at continue to work very effectively. They are a monument to American ingenuity and its technical superiority. They have made significant contributions which have enabled us to survive as the world's only superpower. These systems, some of which are over 15 years old, rely on a variety of proprietary languages. This translates into high maintenance costs. Some of these systems are now undergoing major upgrades, and Ada is successfully being integrated into them as they are reengineered and updated to add new features and functionality. Based on our experience, these improvements will lower our maintenance costs and improve reliability. And that's why we developed Ada in the first place. Our goal was not just to invent another software wheel, but rather to attack the high costs of maintenance, which we believed would eat us alive if we didn't do something. We must continue to take aggressive action; the bear of software maintenance costs remains at our door. We cannot afford to be tied to proprietary solutions. Our focus is to tap the commercial base whenever we can, and to leverage it to reduce our costs of ownership. Ada allows us to do this in a sensible and scientific manner. The defense of our nation is dependent upon reliable, cost effective software. I suggest to you that the private sector has the same profound dependency on software. Ada is the lowest cost, most maintainable, portable, and reliable solution to the software challenge available today. It has the tools we need to develop reasonable solutions to the wide variety of applications we pursue within the DoD. I wish I could predict it would become the economical choice for those commercial sectors that have much in common with the type of requirements we have in DoD, such as - Medical Instrumentation, - Manufacturing, - Automotive, and - Financial services industries. These are among the primary industries upon which the well-being our nation and its economy depend. In these days when our warfighters face peril around the globe, I want our troops to have confidence in the systems that support them. They need systems based upon solid software engineering concepts. They need systems engineered with reliability and maintainability in mind. Most of all, they need systems that can be rapidly reprogrammed or assembled from interchangeable parts as the threat changes. Our warfighters deserve good software, and we are committed to ensuring that they get it. That's why I am so adamant about Ada, as I am still convinced that it is superior to all the other languages. Abraham Lincoln said, "He who has a right to criticize, is he who has the heart to help." I've heard a lot of criticism, so a year ago I invited the community to help. We conducted the first of several planned Ada Dual-Use Workshops. The purpose of the workshop was to devise a strategy based on your inputs to energize Ada and increase its commercial use. A lot of positive events have happened as a result. I would like to tell you about some of these accomplishments. The input from the participants at the Dual-Use Workshop -- which included representatives from government, industry, vendors and academia -- was hard-hitting and constructive. Participants told us that the Department needed to continually reconfirm its commitment to Ada. They told us that we needed to accelerate the release of Ada 9X and get it to market quicker. They told us we needed to make strategic investments in education, bindings and tools, research, and technology transfer. They told us we needed to market Ada better and get the message out that it was a success. Most importantly, they told us we needed to provide leadership and execute a far-reaching and cohesive strategy. These recommendations were heard, analyzed, and acted upon. The Ada Joint Program Office (AJPO) used your recommendations in their plan which Mr. Riefer briefed to me in December of last year. With my concurrence, he immediately began executing this plan. This plan aggressively pursues progress along five strategic lines. They are to: - Increase marketing, - Establish partnerships, - Provide support and incentives, - Reinforce commitment, and - Strengthen AJPO activities. A great deal of progress has been made in each of these strategies. Let me review the more significant of these with you and you will understand why I continue to be upbeat when it comes to Ada. In the area of marketing, many of you have see the Ada success stories that the Ada Information Clearinghouse has published in cooperation with SIGAda. The 21 commercial stories, published to date, highlight the successes that Ada has made in the fields of banking, commercial aviation, telecommunications and the like. The AJPO is also publishing success snapshots on-line weekly. The current distribution list of over 200 people includes most of the senior leadership within the Department of Defense. Our commercial marketing campaign is also well underway. Many of you have seen the advertisements in such publications as Computer World, Dr. Dobbs' Journal, and Software Magazine. The posters we've developed have been a great success, including the one with the surfer on it. The marketing information package we've developed, which contains a catalog of vendors, has been distributed to over 300 interested people within our target areas. Most important, the Ada vendors have piggybacked their advertising along similar lines by emphasizing increased reliability with the use of Ada within safety-critical markets. We have also taken the initiative and started spreading the word at non-traditional trade shows and conference events. We were at the Computer Science Conference and COMDEX and several other non-DoD conferences. At Object World, we not only exhibited, but a Navy command won a major award for an Ada development. The Navy was presented "The best object-oriented system using a non-object-oriented language" award for the TYCOMS Readiness Management System. It was programmed using the AdaSAGE tool. AdaSAGE also stimulated another award. Government Computer News just presented Mr. Al Wynn an award to recognize his efforts at the Department of Energy for AdaSAGE support to the DoD. Actually, the Department of Energy can thank the Marine Corps for exposing them to Ada. I want to publicly acknowledge the Marine Corps' Major Dave Wood for working with DoE's Idaho National Engineering laboratory to develop AdaSAGE. Major Wood exhibits the "Can do" Marine attitude. Major Wood worked with the DoE to develop AdaSAGE in 1987. By 1990 it was being used by every Service within DoD, and in 1991 it caught hold in industry. The Partnerships initiative in the Dual-Use Plan consists of partnerships with educational institutions, government agencies, and industry. The AJPO has pursued the academic partnerships very aggressively. We were told at the workshop that we needed to overcome a number of barriers to the teaching of Ada within our universities and colleges. With this in mind, the AJPO has entered into the following educational partnerships: - To make the Ada tools affordable, we are partnering to develop an Ada 9X textbook with an inexpensive compiler environment shrink-wrapped to it. This textbook will sell for about $50 in the university bookstores. - We have also developed the Ada 9X GNU translator. This software is available free to educators from the Free Software Foundation. It is being used throughout the world to get familiarity and experience with Ada 9X. - We have also awarded 60 university contracts to help universities move to Ada. The course material and teaching artifacts developed under these awards are being cataloged and will be made available to the general public via the ASSET reuse library. - Another 15 to 20 university grants are in process of being awarded to take Ada into engineering and business schools in addition to traditional computer science departments. - A partnership with the Air Force Academy was developed to convert its Department of Astronautics from FORTRAN to Ada. - A survey of the Service schools and academies was completed this summer, and an action a plan is being formulated to increase the use of Ada in these schools. - Based upon this survey, a best-of-breed courseware CD-ROM was developed to provide Service school educators with educational resources. We plan on updating the CD-ROM later this year. It too will be made available to the general public. We are also pursuing government partnerships. ARPA has stepped up to Ada under a number of separate initiatives. We are partnered on a research effort to develop a persistent database binding; this is being done via an Ada 9X object-oriented database binding. This project will investigate use of data in a language-independent manner. Another government partnership is with the Department of Energy to improve the human interface of AdaSAGE. I understand the new release of this tool has incorporated the Windows interface requested by the user community both in industry and government. Earlier this year I signed a letter giving permission to use Ada 9X in certain development efforts. Each of the developmental systems will be compiled on a validated Ada 9X compiler before being fielded. We expect to have some of these systems ready to go when the first compilers are ready for validated use, sometime in early 1995. The AJPO is working with DoD activities to support early usage of Ada 9X on several projects. They are: The Joint Advanced Strike Technology Program, The Sustaining Base Information Support System, The SQS-53 sonar, and The Common Applications Support System. Each of these programs will develop software in Ada 9X prior to the release of the first validated compiler, using its own funds. The AJPO will provide training and transition support to help. In addition to these four programs, I am glad to say that Lt Gen Edmonds, the Director of DISA, has identified two Global Command and Control System applications for transition to Ada 9X. The DISA efforts will be accomplished with AJPO transition support during the next 6 months using in-house resources. The lessons learned from these projects will be captured and published. In this way, other software development activities can learn from their experiences and be better able to take full advantage of Ada 9X's features. The third strategy area provides support and incentives for Ada 9X. We have heard over and over, an inhibitor to the use of Ada was the lack of bindings and tools. Abraham Lincoln said, "It has ever been my experience that folks who have no vices, have very few virtues." While Ada had her problems, she also had many distinct and unmatched positive traits. Ada has the power to provide fault tolerant, reliable systems. But Ada has been criticized, not for itself, but for the lack of support tools. We must turn this around with Ada 9X. Bindings for POSIX, Windows, Motif, SQL, and Common Object Request Broker Architecture are in development. The source code for these bindings will be placed in the public domain when they are completed, so anyone can take advantage of the work. A tools survey is being completed to identify the needs of the Ada 9X user community. In addition, a number of tools are being developed to deal with shortcomings and ease the use of Ada 9X. Finally, we are investigating the feasibility of awarding an umbrella contract for Ada 9X compilers. We hope to leverage the buying power of the Department to get compilers for our programs at the most affordable prices. The fourth strategy area is to "Reinforce commitment". If there was ever a doubt about my commitment, I want you to put your mind at rest about that. Mr. Longuemare and I signed out a letter in August clarifying the Department's continued commitment to Ada as it relates to being a Military Standard. Ada 83 is not just Military Standard 1815A. It is also an ISO (International Standardization Organization), FIPS (Federal Information Processing Standard), and an ANSI (American National Standards Institute) standard. Ada 9X will also become an ISO, FIPS, and ANSI standard. Chris Anderson has done an excellent job in shepherding Ada through this process. Approval by ISO for Ada 9X is expected this December. The final strategy is in the activities of the AJPO. These include putting into place validation and evaluation test suites, providing transition support, and interfacing with the public. The validation and evaluation test suites will be distributed along with test scripts, so that our Ada validation facilities will be capable of validating compilers in March of next year. Very soon our Ada 9X Transition Handbook and Ada 9X Transition Plan Template will be cleared for public release. We have held over 20 Ada 9X for Ada Programmers classes so far this year, training over 300 DoD software professionals in the rudiments of Ada 9X. The Ada Quality and Style guidelines published by the Software Productivity Consortium is also being updated for Ada 9X. A monograph series has been spawned to put details of case studies in the hands of practitioners. The Ada Information Clearinghouse Internet host will be accessible via the World-Wide-Web via MOSAIC by the end of the year. This information resource is updated daily and serves as the source of Ada information. It contains more than 3,412 information files and reports. And finally, the Ada Information Clearinghouse Library is now open to the public at 5600 Columbia Pike in Arlington, VA. Mr. Reifer, as the Director of the AJPO, is reinforcing the Department's commitment to Ada every day by aggressively managing over 200 tasks aimed at accomplishing the initiatives the community said were needed to make Ada successful. From where I sit, the future of Ada still looks very promising. Let me challenge you to help. In the late 19th century an American writer, Elbert Hubbard said, "The world is moving so fast these days, that the man who says it can't be done is generally interrupted by someone doing it." The train is moving and momentum is building rapidly for Ada. If we need to do other things, I need to know. I need your help to feed the fires and keep on accelerating. Tell me what you need to be successful in your use of Ada. I'll listen, and consider all suggestions. I am very excited about the prospects. Thank you for your attention.