JAB's Software Quality Registration Program:
The Controversy Continues
An August 31 JAB-ANSI joint press release proclaimed that an agreement
had been reached in the software accreditation dispute. But follow-up research
by Computing Japan shows this may not be the case. by Steven Myers
Since July, Computing Japan has been monitoring a dispute between the Japan Accreditation Board for Quality System Registration (JAB) and the American National Standards Institute (ANSI) over a JAB proposal to implement a quality registration program for all software sold in Japan. (See "A JAB at US Software," August, page 12; and "A JAB in the Dark," Sept./Oct., page 15.)
On August 31, we received a press release issued by JAB and ANSI declaring that an agreement had been reached following discussions between representatives from the two organizations. Several newspapers duly reported the settling of the dispute, and with all of the issues apparently resolved at that point, the controversy seem resolved. High-ranking officials from both sides, including US Ambassador to Japan Walter Mondale and MITI Vice-Minister of International Affairs Yoshiro Sakamoto, praised the efforts of ANSI and JAB in reaching a "mutually beneficial solution."
When Computing Japan contacted the two organizations to ask about the agreement in mid-September, however, there seemed to be a surprising degree of confusion and misunderstanding about the specifics of the accord. The dispute, it would appear, is not yet settled.
Background
The first public announcement of JAB's software quality registration system proposal was made in April 1995. Under the proposal, software vendors would have the "opportunity" to seek third-party registration of their internal quality systems. This process would entail having a JAB-accredited registrar present inside the company for several weeks to observe and assess the level of quality-control used in the development and manufacturing of all software products.
Several US and European companies quickly voiced strong opposition to the plan, and in early July the American companies sent a letter to US Trade Representative Mickey Kantor requesting trade action against the proposal. Among those signing the letter were Louis Gerstner, Jr., Chairman and CEO of IBM; Lewis Platt, Chairman and CEO of Hewlett-Packard Company; and Gary Tooker, Vice-Chairman and CEO of Motorola, Inc.
Representatives from the American Embassy and the American Electronics Association (AEA) visited the JAB office on several occasions to formally state their opposition to the plan and to explain their position. On behalf of the American software industry, the representatives expressed concern over the following points of the proposal:
* The process for evaluating system quality would be outlined in a special guideline document published by JAB, rather than in accordance with the guideline documents of the International Standards Organization (ISO).
* All registrars would have to be accredited by JAB. Foreign-accredited registrars seemingly would not be recognized, thus posing a significant constraint on companies not resident in Japan.
* Once the voluntary standard was implemented, JAB could attempt to implement the program as a "required" standard, meaning that all software sold in Japan would have to originate from a registered organization.
* There is significant risk that auditors could obtain proprietary information and know-how information during the auditing process.
The JAB response
On July 31, Computing Japan visited the JAB office and spoke with Senior Manager Shigeyoshi Takashi and Manager Fumihiko Akasaki about JAB's response to the foreign criticism generated by the proposal. The JAB representatives insisted that all guidelines drafted for the program are in complete compliance with ISO principles, as defined in document ISO 9000-3, and do not contain regulations that are peculiar to Japan. Although it was true that registrars must be accredited by JAB, Takashi stressed that any organization from any country could apply for accreditation. "Above all," assured Takashi, "the program is strictly voluntary -- no company will be required to undergo auditing and registration if they do not so desire."
Concerning the risk of auditors leaking sensitive corporate information, Takashi stated that this risk is no different than that faced by companies in other industries where quality-control auditing is more common. "If an auditor is found to be leaking proprietary information," he continued, "then his organization will lose its auditing license."
According to Takashi and Akasaki, the whole controversy was due to simple misunderstandings that had been blown out of proportion by the press; the issues could easily be resolved through face-to-face discussion. The two managers expressed indignation that none of the other writers covering the dispute had actually visited the JAB office or talked in detail with JAB representatives. "We were very surprised and disappointed when [AEA representative] John Stern walked out of our office after only 20 minutes, without truly listening to our side of the issue, and then proceeded to issue a very biased, one-sided press release," complained Akasaki.
Quizzed further, though, Akasaki gave some puzzling and even disturbing answers to questions about JAB's motives in implementing the program. When asked why a software quality registration program was needed in Japan, he replied by citing the fact that none of the PCs in his office were able to handle OS-level commands in Japanese. "I've always found it strange that all of the commands are in English," he said. "I'm Japanese, and not so good at English, so why can't they make a system that handles Japanese commands? It's because the OS is owned by America. Japanese companies can't modify the American OS without paying huge royalties." (Just how the quality registration program would address that particular "difficulty" is still not clear.)
The meeting that wasn't
In August, it was announced that JAB had agreed to hold a meeting with ANSI -- by then acting as representative for those opposed to the JAB proposal -- to seek agreement on how the proposal should be implemented. The meeting was tentatively scheduled for mid-September and was to include a small delegation of experts from JAB and JISC (Japan Industrial Standards Committee) on one side and from ANSI on the other. JAB agreed to delay launching the accreditation program until the disagreement was resolved.
On August 31, however, selected media received a surprising press release stating that not only had JAB and ANSI representatives already met, but they apparently had drafted and signed an agreement that was acceptable to both sides. Under this agreement, it was said, JAB would adopt a revised program and forego use of the supplemental guideline document that had been drafted for the original program. Said Dale Misczynski, corporate vice president and director of quality and standards for Motorola, who had headed the ANSI delegation: "We have resolved these issues much more quickly than either Dr. Ohtsubo (Executive Director of JAB) or I considered possible. As a result, the meeting of experts that we had originally planned for mid-September is no longer required."
The elusive agreement
Finding it odd that the press release contained no mention of any of the other points originally raised by the ANSI delegation, Computing Japan contacted JAB the following week to request more information about the recently announced accord. In response to our questions, Toshitaka Toukai, manager of general affairs for JAB, claimed that, aside from agreeing to drop the supplemental guidelines, absolutely nothing had changed from the original plan. JAB was proceeding as scheduled, he declared.
Toukai went on to blast the American side for launching an all-out media campaign that he charged was designed to discredit the JAB proposal. "The most significant result to come from this meeting," he said, "is that in the face of this aggressive American media campaign, we were able -- without making any strange compromises -- to make America to admit that the JAB Software Program was indeed in compliance with ISO principles, and thus made them accept the fact that this program is going to go into effect." Toukai added that JAB would hold an explanatory briefing sometime in October or November, and that the Accreditation Program would begin shortly afterward.
When Computing Japan contacted AEA representative John Stern to verify the JAB interpretation of the agreement, Stern was incredulous. "No, that is completely incorrect! If that's really what they're saying, then JAB is going to have a huge trade conflict on their hands. This whole thing is going to escalate again, and I really don't think they want that." According to Stern, the Japanese side agreed not only to drop the supplemental guidelines, but also to recognize foreign registrars and to ensure that the program would not become a government standard in the future. These points are not mentioned in the signed document, Stern conceded, but they were agreed to orally during the discussions, and recorded in the extensive notes taken from this meeting.
Another representative present at the discussions, though, who spoke on condition of anonymity, said that the whole issue of recognizing foreign registrars is not really significant anyway (for the US, at least), since there exists a "Memorandum of Understanding" between JAB and ANSI stating that each side will recognize the other's registrars. The representative went on to say that the main goal of the ANSI delegation was to have the supplemental guideline document dropped from the program -- since it reflected JAB's unique interpretation of the ISO standard -- and in that they succeeded. A software quality registration program that conforms to the ISO parameters, he said, is not a problem.
So what really happened?
When informed of the comments made to Computing Japan by JAB's Toukai, several non-Japanese experts who had been present at the meeting expressed surprise and concern. "Their account of what took place during our discussions is completely inaccurate," said one. "They're trying to make it look like the American side had misunderstood the situation and was forced to back down, when in fact it was the other way around." What actually happened, he claimed, was, "We pointed out that the documents they were basing their program on were outdated and contained numerous misinterpretations. After that, they were then apologetic and cooperative, and agreed to drop all of their own guidelines."
One member of the ANSI delegation, Dr. Robert Orr, told Computing Japan that some of the JAB statements appeared to represent a face-saving attempt, while others sounded as though JAB was indeed trying to hedge on the recent agreement. "Some of the things they said really happened, but others are a bit off. It's hard to say anything definite at this point," said Orr. "We are alerting Dale [Misczynski, head of the ANSI delegation] to this matter immediately, and will wait to see what his reaction is."
Watching and waiting
In any event, this is where the situation stands as we go to press (late September). It seems the issues are not quite as settled as the initial JAB-ANSI press release made them out to be. And in spite of the August 31 joint agreement, no one that Computing Japan spoke with seems to be able to say -- or is willing to state for the record -- exactly what type of program JAB has in the works.
It all may well turn out to be a tempest in a teapot, but that will not be clear until this controversial software accreditation program officially begins. Perhaps by then the two sides will be able to agree about what they have agreed to.ç
