This page has been translated using TexTra by NICT. Please note that the translation may not be completely accurate.If you find any mistranslations, we appreciate your feedback on the "Request form for improving the automatic translation ".

4th Review Meeting for Revision of Service Design Guidelines

Last Updated:

Overview

Date and

Friday, January 30, 2026, from 10:00 a.m. to 12:00 p.m.

US>

Online meetings

Agenda

  • 1. Explanation of the purpose of the meeting
  • 2. Collection of opinions on each guideline
  • 3. Summary of Results

Material

Handouts

  • Overview of the Guidelines Subject to Discussion * Members only
  • Draft guidelines * Limited to members

List of Attendees

Members (in the order of 50 Japanese syllables / Honorifics omitted)

  • Bunya Akasaka (Senior Researcher, Research Division for Human Society Expansion, National Institute of Advanced Industrial Science and Technology; Visiting Associate Professor, Japan Advanced Institute of Science and Technology; and Visiting Associate Professor, Tokyo University of Agriculture and
  • Yoshihiro Ito (President, NPO Information Gap Buster)
  • Mikie Oi (Integrated Design Laboratory, Mitsubishi Electric Corporation)
  • Teppei Okamoto (Associate Professor, Faculty of Distribution and Information Science, Ryutsu Keizai
  • Nobuko Kitazaki (Professor, Department of Visual Communication Design, Faculty of Art and Design, Musashino Art University)
  • Yoko Nakao (Designer (Inclusive Design))
  • Ikuhide Nakayama (Associate Professor, Faculty of Management, Ritsumeikan University)
  • Atsushi Hasegawa (Professor, Faculty of Design and Architecture, Musashino Art University, President of Socket Co., Ltd.)
  • Natsuko Higa (Representative Employee of Mesh Work LLC, Specially Appointed Associate Professor of the University of Yamanashi)
  • Yasuyuki Hirai (Professor, Graduate School of Art and Engineering, Kyushu University)
  • Naotake Hirasawa (Professor, Department of Social Informatics, Otaru University of Commerce)
  • Shinichi Fukuzumi (Deputy Team Leader, RIKEN Center for Innovative Intelligence Research)

Digital Agency (Secretariat)

Technical Review Meeting Service Design Task Force

(*) As I could not attend the meeting due to a schedule conflict, I will collect individual opinions at a later date. (The content of individual opinions is included in this agenda.)

Summary of the Proceedings

Summary of Opinions from Members

1. Explanation of the purpose of the meeting (Q & A)

No questions asked.

2. Collection of opinions on the guideline revision policy, 3. Collection of opinions on the overall guideline

<DS-630.1 User Research Guidelines>
  • 2.1.2 "User Research": Digital methods such as analysis of SNS utterances and reviews should be added, and the scope of disclosure, possibility of scraping, and points to note such as consent should be clearly indicated.
  • Workshops should be added to the sample methods to indicate applicable conditions to complement observations and interviews.
  • 2.1.7 "Risk": In addition to physical, property and privacy risks, social impacts such as labeling and the promotion of prejudice should be included and examples of mitigation measures should be provided.
  • 2.1.8 "Benefits": In addition to economic and intellectual benefits, the acquisition of rights and the resolution of problems of Special Care-required Participants should be added so as to be consistent with 9.1.2 "Cases in which survey participants are expected to gain generalizable knowledge."
  • 2.1.9 "Representation": The term "social standing" is ambiguous and should be clarified with indicators such as social class and socioeconomic status (SES).
  • 3.5 "Cases where exemptions are possible": Simplification of procedures, rather than uniform exemptions, should be limited. Minimum requirements (e.g., simplified notification methods and records) should be exemplified.
  • 3.5.1 "Operating in a Safe Environment for Education and Training": We would like to clarify the requirements for safety (limited scope, handling of records, freedom to leave) and the limit of simplification.
  • 4.1 "User survey at project start-up and initial response": The intention, necessary information, and influence on decision making of the survey should be clarified, and its connection to planning and requirements definition should be emphasized.
  • 4.2.3 "Points to Consider regarding Contracts with Outsourcees": It is necessary to go beyond confirmation of acquisition and management of consent to implementation (feasibility study, alternative proposal, and improvement) requested by requestors.
  • Compliance clauses, audit authority, correction obligations and training requirements should be written into contracts.
  • We would like to reinforce the significance of risk management, reference flow of partial application, and AI evaluation after 4.3 "Service and business planning".
  • Chapter 4: Segmentation itself is useful. I think it is necessary to illustrate the relationship between headings and reconstruct them on the axis of "why, which phase, how to proceed, and how to choose".
  • The method should be accompanied by selection criteria, and the storage period and disposal method should be specified in the consent form. It is also recommended to indicate the variety of analysis, the limit of AI, alternatives in the event of operational failure, and the standard of difference in scale.
  • 5.4 "Obtaining consent" to 5.7 "Omission of written consent": Clarify the principles and exceptions of written consent. The title of 5.5 should be changed from "Requirements for obtaining consent" to "Conditions for handling data", and the form of consent and conditions for handling (anonymization, openness, confidentiality) should be separated. It is also desirable to clearly state the premise for applying 5.7.
  • 5.4 "Obtaining consent": Some methods allow for verbal or electronic consent, flexible timing, and standardize the recording method only.
  • We would like to add that approval by the Ethics Committee may be required, the sample size, the summary of the procedure, and the consultation service.
  • In addition to the decision-making process, it is recommended that a checklist (high-workload work, physical interventions, and sensitive information) be prepared, and that the supervisor escalate the matter to the ethics committee when appropriate. It is also recommended that the scope and applicable criteria be specified.
  • It is recommended to show the flow chart for determining the necessity and to distinguish the handling of responsibility and records between voluntary implementation and outsourcing.
  • It is recommended that the right to discontinue, privacy, confidentiality, and the right to leave be clearly stated not only for the persons concerned but also for sign-language interpreters, helpers, family members, and other supporters, and that the handling of information on those present be organized.
  • 7.1.1 "Consideration for the vulnerability of Special Care-required Participants": It is advisable to make it a requirement to be able to explain the reason for selecting target participants. Please change "research" to "survey".
  • With regard to the fairness criteria, the setting of a target attribute ratio based on population statistics will be examined, and the limit will also be included in the agenda for continued discussion.
  • With regard to affirmative action and minority inclusion, it would be better to adjust the design and recruitment so that it does not depend on the ratio, but can also include very few and complex characteristics, and to allow preferential recruitment.
  • 8.1 "Data protection" & 8.2.1 "Data disclosure": It is desirable to clearly indicate specific measures (access control, storage period, encryption, prohibition of removal, audit log, disposal procedure). It is desirable to note re-identification risks after anonymization and prevention measures (suppression, threshold, differential privacy, etc.).
  • 8.3 "Consideration in analysis using AI": We would like to stipulate that attention should be paid to stereotype amplification, human interpretation should be made mandatory, consent should be obtained after explaining the use of AI and risks, and unauthorized tagging should be prohibited.
  • 9.3.1 "Requirements for approval by a legal representative and consent of survey participants": If identity verification is possible, the consent of the individual should be obtained as a rule. If consent is not possible, it is preferable to clearly state that confirmation should be made with the legal representative. It is necessary to devise ways to avoid misunderstanding that a proxy is essential only based on the presence or absence of disability.
  • It is desirable to include a section to disseminate information through groups and places of use, to give consideration to resistance to the selection of specific conditions, and to describe examples of cross-sectional needs (e.g., touch panels and visually impaired people and parents with children).
  • 9.1.2 "If it is likely to provide generalizable knowledge to survey participants": It is necessary to be consistent with 2.1.8 "Benefits" (clarify the scope of benefits for both individuals and communities).
  • We would like to present specific examples of threats to anonymity and participation of users with various characteristics (forced face detection, real-name linking, metadata leakage) and workarounds. It is also recommended to show the response to standards and process assessment of the output.
  • Typical use scenarios (short-term verification, exploratory hearings, on-line observations) should be presented for field adaptation, and navigation by purpose, phase and risk should be designed.
  • Regarding the sense of the level of ethics, if the criteria for "necessary and sufficient" are clearly stated and overly strict evaluation is avoided by using a checklist of low, medium, and high risk cases, it may be possible to prevent misunderstanding that "it is okay to be lax".
  • When learning administrative work, it is recommended to improve basic literacy through the introduction version, e-learning, etc., and make it mandatory for staff to understand and review the work, and clarify the division of roles.
  • Removal of information and communication barriers, flat dialogue workshops, imputation sampling and record transparency, and stakeholder involvement and co-design of semantic saturation criteria are recommended.
  • With regard to qualitative surveys, it is recommended that the number of respondents be expanded to 40-50 when minorities are included, and that the criteria for KJ method, multiple analysts, triangulation, and amplification of minority opinions be clearly stated.
  • The term "research participant" should be used consistently, and "test subject" should be avoided. The definitions of "special care-required," "reasonable care," and "ethical care" should be consistent.
  • We would like to redesign the title and cover to be well-coming and strengthen the connection with the 12 service design items. It is recommended to clarify the positioning in the high-level frame and the overall structure diagram common to the three documents, and to incorporate the exploratory viewpoint from the beginning.
  • It is recommended that ethical guidelines for social research and fieldwork (guidelines of the American Anthropometric Association and the Association of Internet Researchers) be used as a frame of reference, along with practical examples (possible negative ethical consequences of actions), and that ethical codes and international policies in the field of marketing research be referenced to reinforce the criteria for applying user research.
  • We would like to show the basic form of user research (ISO HCD diagram), and specify the positioning which is not absolute compliance and the essential matters for each phase.
<DS-670.1 Usability Guidelines>
  • Chapter 3: The "what, to whom, and why" should be clearly stated at the beginning of the chapter, the object and purpose should be defined before the text, and the title should be changed to reflect the nature of planning ahead of implementation, such as "Planning for an Easy-to-Use System."
  • 3.3.8 "Users' Concerns and Expectations about Technologies and Systems": Systematically identify areas of concern such as fairness, non-disclosure of information, and the division of roles between humans and systems, and define the evaluation procedure from problem identification, hypothesis, verification, and correction. It is also recommended to include other than general touch points and to clearly state the evaluation process in an operational manner.
  • 5.3.1 "User Feedback Form": Brief and unclear in relation to the User Research Guidelines. It is recommended to specify how the collection, classification, analysis, and prioritization of feedback from the form will be linked to the user research process (including the treatment of consent and privacy) and to indicate the relationship between the two guides here.
  • 5.4 "Evaluation of AI Systems in Operation": It is commendable that a AI evaluation chapter has been added. In addition, it would be useful to reinforce specific evaluation perspectives and design considerations, such as a system side guide for the next action when an unexpected response occurs, presentation of misperception and uncertainty, and a UI that prompts human confirmation.
  • As a whole, the amount of text is large and the reading hurdle is high. It may be better to introduce tables and figures that list the main points and visually show the relationship between chapters and the flow of decisions to promote understanding and operation.
  • The provision of information is one way, and the perspective of confirming whether the recipient can understand is weak. Understanding (confirmation questions after explanation, repetition of summary, alternative explanation when understanding is difficult) should be incorporated as a design requirement.
  • Clarify the guidance of design (default effect, etc.) and define the concept of tolerance. Review the image of a "rational economic person" that administrative documents tend to assume. Based on the assumption of misreading and skipping, prepare information design and confirmation processes that are less likely to cause cognitive errors. The UK GDS "Good Government Services" reference should promote the sharing of good conditions.
  • It is desirable to clarify the flow of how to cover errors and how to connect them to the final AI by showing a framework that links the three aspects, "AI during use (outcome)," "use errors (in use)," and "quality control system." Descriptions related to quality control should supplement specific points to note, such as the system guiding the next action when an unexpected response is made.
  • Redefine "use error (in use)" not as a person but as a design cause (procedure, information presentation, environmental factor). On the premise that it is difficult to read carefully for a long time, it should be reflected in the task / evaluation design, and the evaluation procedure, pass / fail criteria, and necessity of assistance should be specified.
  • Since face alignment is difficult for users who have difficulty obtaining visual information, it is clearly stated that "installation of facial recognition terminals that are difficult to align if they are not visible" is not acceptable. It is necessary to avoid designers' assumption that facial recognition alone is sufficient.
<DS-671.1 Usability Introduction Guidebook>
  • 2.1.1 "Usability point of view": It is necessary to review the description of "completeness of usage" because it has already been deleted from SQuaRE. The sentence "Usability is also defined as' quality at the time of use' in SQuaRE" should be deleted because it is an error.
  • "JIS X 25010:2013 (ISO/IEC 25010:2010)" should be updated to the latest version, and references to the standard in this chapter should be reworded to ensure that the relationship between usability and quality models is not misinterpreted.
  • 2.5 "Cognition and Bias": Reorganize the chapter structure to deal with the diversity of cognition (e.g., literacy disorder) in a separate section. Clarify healthy person bias and bias in the viewpoint of providers, and add specific guidelines for correction. Unify the term to "user", and define UX as an end-to-end experience. It is recommended to illustrate the positional relationship between service design and usability, and incorporate a viewpoint that captures the usage process in the time axis into the text.
  • 3.1 "Design and Ethics": In order to go beyond the conceptual explanation, it is recommended to add concrete examples based on cases. It is also recommended to include examples of negative ethical effects that an action can have so that people can understand at a practical level what can be considered as consideration.
  • Chapter 6, "USE ERROR MECHANISMS," through Chapter 9 (related chapters): The description of use errors is biased toward legacy safety analysis and is a little old. It would be better to update this with experts in the next examination.
  • The link between human factors and usability in security (e.g., the relationship between usability and misoperation of authentication procedures) should be shown in concrete examples and evaluation perspectives.
  • We would like to prepare cross-references with standard guidelines (JIS/ISO, X 8341, Z 8520, Z 8522, ISO/IEC 25000 series, etc.) and clarify which requirements of which standards should be met.
  • Human-centered thinking should be reflected in the governance and digital human resource development sections (role definition, education, review system).
  • Introduction (Introduction) ・ As a minimum requirement for each chapter, it is necessary to clearly state the requirement (checklist) of "At least this" in each chapter so that the operation does not vary by vendor and department. I would like to list the mandatory / recommended category, trail, responsible person, and approval gate in the introduction section.
  • The guidebook should be designed so that it can be easily searched by AI, etc. (design of headings, unification of terminology, addition of metadata), and it should be adapted to actual use scenes. The user research guideline should not be published separately, and duplication of content and reference relationships should be optimized on the premise of understanding the revision policy of incorporating the essence into the usability guideline / introduction guidebook.
  • The opinions of the persons concerned with disabilities and minorities should not be an afterthought. The persons concerned should be involved as staff from the early stage of design, and the process of continuous consensus building and verification should be clearly stated in the guidelines. By making the participation of the persons concerned a requirement (subjects, stage of involvement, method of reflecting in decision making, records, etc.), it is better not to end up with a formal hearing.
  • It is necessary to clearly state the premise that the user experience is continuous across jurisdictions and to ensure external consistency even among multiple ministries and agencies. We would like to request that the system be designed and operated so that staff members are also users and the risk of burden on the site is reduced.
  • The "Design Principles" should be restructured as follows: Superordinate Principles → Specific Norms → Application Examples. Previously released documents should be organized by "User Characteristics → Sensory Characteristics → Information Support Equipment," and the relationship between reasonable consideration and special consideration, social model, coordination through dialogue, and the right to obtain the same content at the same time should be specified. First, the design should be enhanced comprehensively to reduce dependence on individual consideration. The volume should be based on "time reduction," and it is recommended to support short-time understanding through generative AI utilization, headlines, and summary of key points. Extreme good practices (e.g., watching subtitles at double speed) should be generalized.
  • On the premise of the cross-sectional nature of administrative services, a team structure (responsible persons, persons in charge of practical operations, and related departments) should be established in principle instead of individual personnel, and roles and decision making should be materialized to the extent possible. It is desirable to give priority to the provision of knowledge at this stage and to shift to a "delivery" design that enhances readability and understanding support (including lectures and events) at the next stage.
Guidelines for Colonies and Others
  • Based on the fact that the Tokyo Metropolitan Government's Service Design Guidelines specify utilization at the planning and conceptual stages, it is desirable that these guidelines also clarify the relationship with the standard guidelines in the preface and indicate "how to partially use" at each stage of designing, operation, and evaluation. In concrete terms, reference flow lines by life cycle and cross-references (designing → requirements and agreements, operation → data protection, evaluation → AI evaluation and improvement) will be placed in the introduction.
  • The expression of "disability" should be unified within the administration based on the views of the concerned parties (disability is on the side of the social system), the discussion of Cabinet Office, and the practice at the training site (kanji expression), and the reason should be clearly stated. Policies need to be consistent, including the pros and cons of hiragana expression.
  • Throughout the three documents, the term "audience / participants" should be used to clarify the policy and ensure consistency, for example, by replacing the term "audience / participants" with "participants." The basis for terminology uniformity (e.g., international trends, Helsinki Declaration) should also be provided.
  • Regarding accessibility, we would like to specify consideration for visual hypersensitivity and auditory hypersensitivity, and include implementation policies such as image generation buttons and guidance in the check items for survey design so that settings such as the black and white inversion (disable) mode of the OS are reflected on browsers and the web. We would like to evaluate the consistency of OS level settings and web UI as accessibility requirements for the research environment and stimulus presentation.
  • In the three documents, the government should differentiate itself from the private sector and clarify its intention for the administration, and present the "core for the ministries and agencies" by highlighting the administrative issues (fairness, accountability, laws and regulations observance, accessibility, reasonable consideration, site restrictions, public nature, trust, etc.).
  • Although FMEA/FTA (Failure Mode and Effects Analysis, Fault Tree Analysis) originated from a mechanical system, it is appropriate from the viewpoint of ensuring reliability, which is indispensable for administrative systems. Even in the current situation where agile and modification after construction are common, it is useful as an assurance logic. In order to convey the importance to the reader, we would like to briefly supplement the background of the description (reason for adoption and positioning), application scene (usage in planning, design and operation), and obtained effects (risk identification and recurrence prevention), and show the connection with the modern development context.

Greater than or