Standardization supports the globalization of markets, so it doesn’t face resistance from policy makers and industry as back-door regulation. Nonetheless, standards aren’t built quickly. An example is HFES 200, Human Factors and Ergonomics Society project to develop a software user interface standard. It has been under construction for 11 years.
In August The Ergonomics Report™ interviewed cognitive scientist Paul Reed, Ph.D., the Chair of the HFES Technical Standards Committee on Software User Interfaces, which has the task of seeing HFES 200 established as an American National Standard using American National Standards Institute (ANSI) procedures. The most pressing question was what makes HFES 200 worth 11 years-plus of toil.
Dr. Reed replied that the technical design recommendations in it “provide detailed design guidance for improving software usability and accessibility [that represents] the consensus of dozens of international experts.” It is based on available empirical evidence and established practices for improving the usability and accessibility of software for the widest possible range of users, he said, adding that software users and producers will both benefit from this “consistent set of technical design recommendations.”
“In the Beginning … “
Dr. Reed, a Bell Laboratories scientist for 18 years who is now director of a Denver-based company that specializes in advanced electronics packaging, has been involved with the project from its start. “The real genesis traces back 21 years ago to the summer of 1986 when a meeting was called in Washington DC concerning the new software ergonomic standards being developed in Europe,” he explained. In 1986-1987, software ergonomics projects were undertaken in Germany – by DIN – Deutsche Institute fur Normung – and with ISO 9241, a project initiated by the Geneva-based International Organization for Standardization (ISO). The latter has been a construction zone since 1986.
“The initial formation of the HFES Technical Standards Committee on the human factors engineering of software user interfaces … was in response to the European activity,” he added. “It is fair to say that from 1986 to 1996 “a principle focus for the committee was contributing to and participating in ISO 9241, and the set of software ergonomic standards that were developed and approved within ISO.” He noted that consensus was achieved within HFES in the 1990s “that we should pursue a national standard. The specific issue of software accessibility emerged in the mid 1990s as an area of critical need, and that represented fundamentally new technical design guidance that had not been addressed substantively in any of the ISO 9241 activities.”
HFES 200
The full name of the project is, “HFES 200 – Human Factors Engineering of Software User Interfaces.” As described on the HFES website, it has five parts:
Part 1, the introduction, gives an overview of the content, explains relationships among the individual parts and provides guidance on the relevance of individual parts to the development process so designers can understand where and when to use the parts.
Part 2 covers accessibility. It provides recommendations on features and functions of computer operating systems, drivers, application services, other software layers on which applications depend, and applications that increase the accessibility of applications for users with disabilities. It points out that hardware is not specifically addressed by any recommendations, but that many hardware assistive devices could utilize recommended functions provided by operations system and application software. It notes that Part 2 has been extensively harmonized with ISO 9241-171 Software Accessibility Committee Draft released in the first quarter of 2005.
Part 3 covers interaction techniques and incorporates material from ISO 9241 Parts 13 through 17, and is compatible with those ISO standards.
Part 4 covers interactive voice response. It consists of completely new material that has not appeared in ISO 9241 standards.
Part 5 covers the visual presentation and use of color, incorporating material from ISO 9241 Part 12 and including new recommendations on the use of color.
ANSI Requirements
ANSI approved the project proposed by the HFES, specifying the “canvass method” for achieving the consensus on the technical recommendations necessary for the eventual approval of the standard. According to Dr. Reed, stakeholders who may be directly or materially affected by the proposed standard make up the canvass committee. Producer, User, and General Interest stakeholder categories each have 5 representatives on the HFES 200 Canvass Committee for a total of 15 members.
The choice of these categories ensures that there is an equitable distribution of stakeholder interests, he explained.
He noted that the selection of the canvass committee began with a call for volunteers in the ANSI publication and in industry publications. The announcement requested individuals in the three categories to contact the HFES to apply for membership on the committee.
The “equitable distribution” was managed by inviting key stakeholders to participate in the process. “In some cases we invited key accessibility technical experts and leading industrial players to participate on the Canvass Committee to ensure proper representation of stakeholder interests, in accordance with ANSI procedures,” said Dr. Reed. “
The public also has a say. The ANSI procedures require a public announcement in the ANSI national publication notifying “everyone in the United States public that this document is being reviewed,” said Dr. Reed. “Anyone has an opportunity to review the document and submit comments.”
He said the main steps in the process are the distribution of the project document to all canvass members, the announcement of the document distribution for public review in the ANSI publication, Standards Action, and the receipt of ballots and comments from canvass members and public reviewers.
Building Consensus
The complexity of the procedures for building consensus also slows the process. The voting criteria require that two-thirds of non-abstaining canvass members must check Yes on the ballot for an issue to go to the next step, according to Dr. Reed. The process requires “canvass member and public reviewer approval of the disposition of all comments received from all reviewers, and satisfactory adherence to ANSI required procedures as determined by the ANSI Board of Review.”
The biggest obstacle to wrapping up the project, he explained, has been “the satisfactory resolution of the hundreds of complex and detailed technical comments that have been received from reviewers. Because many of the design recommendations involve design trade-offs and functions that span operating system, middleware, and application layers of software, it has been tremendously challenging to address and resolve detailed technical comments to the satisfaction of all participants when diverse comments originate from software developers, users with disabilities, and international participants.”
We are still completing ANSI Canvass procedures, he responded when asked about the present status of the HFES 200 project. While an overall "approve" vote was obtained in the latest round of balloting, he said, “ANSI procedures require the HFES to have all comments addressed and distributed to the HFES 200 Canvass Committee. … That process must be satisfactorily completed before the HFES can submit the document to ANSI for final approval.
The crucial next phase faced by the committee, he said, is to have all of the critical comments received from any source in hand before the second ballot. “We need to have all of those comments addressed and resolved, with the comment dispositions explicitly documented in a document that will be distributed to everyone. That will be distributed back to all of the canvass members and all the public reviewers.” He emphasized that ANSI procedures require that every person has the opportunity to review all of the comments that were submitted. “And they have an opportunity to change their ballot – that’s true with every ballot undertaken using the canvass method,” he added.
“It’s possible,” he replied when asked if this could mean more delays.
Coordination of HFES 200 with its ISO counterpart has slowed the process, but it is an area where there is clear achievement. “We have expended tremendous effort to closely coordinate and harmonize with the ISO document and it is fair to say that with our most recent document, when it was released, it had every single technical recommendation had been compared with its ISO counterpart, and our reviewers determined that the design guidance was completely consistent with no conflicting design recommendations in the ISO and ANSI documents.”
Asked if ISO 9241-171 and HFES 200 Part 2 (Accessibility) might be able to function eventually as a single universal standard, Dr. Reed was emphatic that they will be two separate standards from two separate organizations. “The technical content level of the recommendations – the technical content – is virtually identical. However, it is critical to highlight that in ISO they have determined that they wish to include ‘shall’ statements that become mandatory for some of the technical design recommendations. In ANSI, in stark contrast, we have only “should” statements
(recommendations) and no mandatory ‘shall’ statements (requirements).”
Given the industry positions and preferences I’ve observed over the past 20 years, he explained, I would have to say that … key United States industry players would have resistance to any type of mandatory regulation.”
He notes strong industry support for harmonization between ISO and ANSI because companies with global interests “wish to be able to develop their products to a single set of design recommendations rather than having one set of standards applying to the United States and a separate different standard applying to Europe and beyond.”
Asked to predict when HFES 200 will become an American National Standard, Dr. Reed replied, in so many words, that he couldn’t say. “There is an inherent uncertainty about when we could expect to receive conclusive feedback from the HFES 200 Canvass Committee membership regarding the disposition of all comments.”
On its website HFES sounds a confident note: “Based on largely favorable responses from ballots received during the first round of balloting, the committee is cautiously optimistic that the second ballot will result in approval of HFES 200 as an American National Standard.”
Sources: Dr. Paul Reed; HFES
This article originally appeared in The Ergonomics Report™ on 2007-08-17.