-A A +A

CS2013-Strawman-HC-Human-Computer Interaction (DEPRECATED)

9 posts / 0 new
Last post
sahami
CS2013-Strawman-HC-Human-Computer Interaction (DEPRECATED)

The CS2013 Strawman Report comment period is now closed. Please see the CS2013 Ironman report for the latest CS2013 draft to comment on. Forum to comment on "HC-Human-Computer Interaction" Knowledge Area in the CS2013 Strawman report. We ask that comments related to specific text in the report please specify the page number and line number(s) of the text being commented on. Line numbers are provided on the far left-hand side of the each page.

merkle
Comments on the HCI KA Description
field_vote: 
0
Your rating: None

[From 2 to 10]
I will first comment the current description. At the end, I propose an alternative.

2 Human–computer interaction (HCI) is concerned with designing the interaction between people
3 and computers and the construction of interfaces to afford this interaction.

OBS: I understand that HCI description should avoid the dichotomy “between” people and computers, as if they were apart. Computations are carried by users, through the use of computers in action, through computations, broadly understood. Some tasks would never be conceived if there were not computers. Computations are always situated and historical. Human actions are the target, not the computer itself. For a Portuguese reader, it seems that there is a strange parallel between “desining” and “the construction”. (Why not “constructing” or “building”). The term design seems also too narrow, and could be complemented buy a reference to “evaluation”.

4 Interaction between users and computational artifacts occurs at an interface which includes both
5 software and hardware.

OBS: If that were truth, Human-computer interfaces would be enough to describe Human-Computer Interaction. Today, we see a myriad of other trends, such as CSCW, interaction design, software ergonomics, web design, tangible interfaces, ubiquitous computing, etc. Interfaces/Systems were no longer used by isolated individuals, but collectively, socially, though networks and in distributed ways. Therefore they are thicker than "a interface" is supposed to be understood, and should not be flattened out. They are also multiple, and multimodal. The HCI KA indeed include topics such as collaboration and privacy, which are difficult to inscribe solely at a single interface.

5 Interface design impacts the software life-cycle in that it should occur
6 early; the design and implementation of core functionality can influence the user interface – for
7 better or worse.

OBS: User centered process models tend to avoid linear models. If cyclic and participative, terms such as “early” or “end user” are problematic. The term “impact” also implies that the computers come first and the use latter, what is a clear contradiction with the HCI KA.

8 Because it deals with people as well as computers, as a knowledge area HCI draws on a variety
9 of disciplinary traditions including psychology, computer science, product design, anthropology
10 and engineering.

OBS: In times of broadening computing, as this draft states, other dimensions and areas should at least be acknowledged.

FROM A NON-NATIVE SPEAKER, HERE IS MY SUGGESTION FOR THE DESCRIPTION [From 2 to 10]:

Human–Computer Interaction (HCI) is concerned with evaluating and designing interactions, construed as tasks or activities carried by people with the aid or support of computational systems, interactive devices, and associated practices.

For better or worse, the design and implementation of the core functionalities of computer systems can and usually influence what they do as machines, as well as their user interfaces, and therefore can have deep repercussions on the human actions or work they intend to support. The earlier HCI concerns are included in a product life-cycle, or the clearer the process is human centered, the greater the chance that a system will fulfill the users' needs satisfactorily or plentiful.

Therefore, the construction and test of computer interfaces, which include both software and hardware, and the design and evaluation of the tasks and activities they support, demand the consideration of cultural, social, organizational, cognitive and perceptual issues, among others, and draw on a variety of disciplines and interdisciplinary traditions, such as sociology, anthropology, psychology, graphic and product design, human factors, as well as information systems, computer science and engineering and related areas, and are contingent to the domains the target systems are involving.

safincher
Responding to your comment
field_vote: 
0
Your rating: None

Thank you very much for taking the time and effort to read and respond to the "strawman" draft. We have re-written the HCI description to incorporate the spirit of your suggestions. I hope that you will find it improved in the "ironman" version.

dcdl
Feedback on KA: HC- Human Computer Interaction
field_vote: 
0
Your rating: None

I have posted some comments that I believe apply to all KAs within the CS2013 Strawman Report-Chapters 1-5 thread. I would like to make reference to those comments here since they apply to all KAs.

The comments are posted under the following subjects:
First of All: Thank You.
Matching of Topics and Learning Outcomes for Core-Tier-1 and 2.
Unique Labeling or Numbering of all Topics and Learning Outcomes.
Improved Correspondence between Learning Outcomes and Topics.
Consistent Labeling for Levels of Understanding.
On the Importance of the Learning Outcomes in Core-Tier1 and 2.

Comments for this KA:

Page 69: I think that this KA should be labeled HCI rather than HC. In this new report there are several other labes with 3 letters.

Page 69, Line 31: I would add as Core-Tier1: Case Studies of expensive and catastrophic errors and the role of HCI in them.

safincher
Responding to your comment
field_vote: 
0
Your rating: None

We've tidied up all the small errors and taken your advice and renamed the KA "HCI" - as you pointed out, it did seem perverse to retain the two-letter identifier here where the three letter acronym is in such common use.

Thanks!

sahami
Additional comments (received via email)
field_vote: 
0
Your rating: None

As I approached this I wondered about:

• Evaluating a system to find out how best I learn?
• Interfaces for supporting online education, game playing, safe system development (with obvious alarms), secure (info not leaked through interfaces).

I guess they are covered but not explicitly. There is scope here for making this more explicit and thereby highlighting the outward facing nature of the curriculum.

KLM = ?

See HC/Designing Interaction. The learning outcomes of this section are curios. Surely outcomes 2, 3, 4 should be renumbered 1, 2, 3 and the first one dropped.

There are places where the learning outcomes are qualified by [knowledge], [Evaluation], etc and in the HC section round brackets ( and ) are used rather than [ and ]. Consistency?

safincher
Responding to your comment
field_vote: 
0
Your rating: None

Thanks for this. We've expanded acronyms and tidied up the learning outcomes. With regard to your ocmment on the "Designing Interaction" learning outcomes. There are only 3 learning outcomes for Designing Interaction.
I think you may have mis-read the outcomes for Programming Interactive Systems as if they were here?

gerrit
ACM SIGCHI comment on the HCI part of the Strawman Curriculum
field_vote: 
0
Your rating: None

Reflections on HCI Strawman Curriculum: Summary & Recommendations
Gerrit C. van der Veer, ACM-SIGCHI President

Rather than reflecting on separate phrases in the Strawman curriculum, we developed a consistent vision on the whole document, based on an empirical study and subsequent analysis.

The following is this report.
In order to comply with the rules, we have indicated the sections most related to our remarks, but we strongly suggest to consider our vision as a whole, including he appendix that reflects our empirical foundation for our comments.

(re: page 1, line 1:)

The ACM SIGCHI Executive committee is sponsoring a 2-year project on Human-Computer Interaction Education spearhead by Elizabeth Churchill, Jenny Preece, and their student assistant Anne Bowser. Based on insights from the HCI community garnered through this research effort to date, the SIGCHI Executive Committee would like to offer the comments on the proposed Strawman curriculum.

Main Compatibilities and Recommendations

Three aspects of the curriculum are particularly compatible with the results of our research:

1. The two core courses cover foundational theory and design topics and methodologies. This pairing reflects our understanding of typical HCI course structure and supports central aspects of HCI education for computer science students.

2. An emphasis on design principles and processes in both HC/Foundations and HC/Designing Interaction reflects important competencies. “Principles of good design,” found in HC/Foundations, and “Basic two-dimensional design fundamentals,” found in HC/Designing Interaction, are especially pertinent.

3. The three levels of mastery defined as knowledge, application, and evaluation are highly compatible with HCI pedagogy, which approaches education as an applied and iterative process. The learning outcomes of both HC/Foundations and HC/Designing Interaction imply that students will apply their knowledge through projects, which reflects the structure of many existing HCI courses.

With that said, we have three primary suggestions for improving the HCI portion of the Strawman curriculum:

1. Design and empirical research methods are central within HCI as a discipline. We recommend increasing the number and variety of methods covered in HC/ Designing Interaction. These methods should span the entire software design and development cycle and include methods for ideation, prototyping, and evaluation.

2. Qualitative methods are at least as important as quantitative methods. We would like HC/ Designing Interaction to have a general qualitative methods requirement and also list specific topics such as interview, observation, and/or ethnography.

3. HCI increasingly incorporates subjects, topics, and epistemologies from the social sciences. We recommend emphasizing cognitive and social models over physical capabilities in HC/ Foundations to reflect this growing incorporation of social science concerns and perspectives. Specific topics of interest include anthropological theory/methods, sociological theory/methods, and social psychology/methods.

Electives

(P3, L 65:)

HC/ Programming Interactive Systems
• Related research: Our research participants consider interaction design, the ability to build high-fidelity prototypes, and information architecture to be important topics in HCI. Participants believe that learning how to operationalize requirements through scripting and programming helps students to develop logical and procedural reasoning skills and a deeper understanding of technology development processes, and offers a foundation for collaborative design/development.
• Recommendations: Universities should offer this valuable course to non-majors as well as computer science majors to support the interdisciplinary nature of HCI.

(p4, L 95:)

HC/ User-centered design and testing
• Related research: Our research points to the perceived importance of ideation and evaluation methods in user-centered design. By covering topics ranging from requirements gathering to prototyping and evaluation, computer science students are exposed to crucial HCI material and skill development.
• Recommendations: An excellent course, this offers core material (e.g., techniques for gathering requirements and presentation of requirements) that are core to HCI material and, as such, should added to HC/ Designing Interaction.

(P4, L120:)

HC/ Design for non-mouse interfaces
• Related Research: Mobile and tablet interfaces are considered to be as important as the more traditionally studied desktop experience. Many also consider shared, embodied, and large interfaces to be important. Similarly, new input modalities such as sensor and location are considered to be as important as keyboard.
• Recommendations: Preserve this timely elective, and consider including new input modalities and methods of data collection to complete a “new interactive technologies” course.

(P5, L 144:)

HC/ Collaboration and communication
• Related research: Our research participants consider social media, social computing, and social network analysis important. Many also advocate the importance of topics in Computer-Supported Cooperative Work (CSCW) such as online collaboration, ‘smart’ spaces, and social coordination aspects of workflow technologies.
• Recommendations: Consider adding a CSCW component covering the topics listed above to examine collaboration and communication in both work and recreational environments.

(P5 L 165:)

HC/ Statistical methods for HCI
• Related research: Statistics, empirical research methods, and quantitative methods are deemed centrally important to HCI.
• Recommendations: Universities should offer this valuable course to non-majors as well as computer science majors to support the interdisciplinary nature of HCI.

(P6 L186:)

HC/ Human factors and security
• Related research: While our research participants prioritize other topics in HCI over issues related to security, our review of papers in recent SigCHI associated conferences indicates that usable security is a growing research focus.
• Recommendations: Preserve this timely elective, but include a stronger focus on the user-centered aspects of secure systems design/development.

(P7 L 211:)

HC/ Design-oriented HCI
• Related research: Respondents consider interaction design and graphic design important, and believe that HCI education should include both theoretical elements of design and practical, studio-based design courses.
• Recommendations: Universities should offer this valuable course to non-majors as well as computer science majors to support the interdisciplinary nature of HCI.

(P7 L 237:)

HC/ Mixed, Augmented or Virtual reality
• Related research: A growing niche of HCI students, professors, and practitioners consider augmented reality, virtual reality, and game studies important.
• Recommendations: While this elective does not reflect the immediate priorities of the HCI community it does represent an emerging area of research that should be supported.

(P2 L 14:)

Tier 1: HC/ Foundations

HC/ Foundations covers foundational theory by addressing the contexts for HCI, the user-centered design and development process, and various factors that influence design.
Contexts for HCI include important interfaces like mobile and more general interfaces such as webpage and business applications. We recommend adding additional interfaces such as tablet, poster, and tabletop to the curriculum and also addressing the history of HCI.
The user-centered design and development process is central to HCI, and HC/Foundations covers important topics such as empirical research and agile/iterative design. However, the topics listed with “different measures for evaluation” do not match the results of our research. We would recommend replacing these topics with a more in-depth look at usability heuristics and the principles of usability testing.
The remaining course material covers factors that influence design. Social models, cognitive models, design principles, and accessibility are all considered important by a majority of our research participants. A minority believe that it is important to study physical capabilities.
We therefore recommend emphasizing cognitive models, social models, and principles of good design over physical capabilities in the proposed HC/ Foundations course. Specific topics of interest may include anthropological theory, sociological theory, social psychology and social science research methods.

(P2 L 42:)

Tier 2: HC/ Designing Interaction

HC/ Designing Interaction covers design fundamentals, methods in the design process, help & documentation, handling human/system failure, and user interface standards.
Our research participants believe that design in general is important and mention design topics such as visual design, graphic design, and studio-based design methods.
A strong training in ideation, design, development and evaluation methods is central to HCI. Our survey respondents also considered “general qualitative methods” a key topic in HCI, including: proficiency with interviews; observation; paper/low-fidelity prototyping; usability testing; experimental methods; field study/ ethnography; brainstorming; and scenarios/ storytelling. A smaller number of participants consider task analysis to be central. We recommend amending this course to offer a greater emphasis on qualitative approaches including the qualitative aspects to generating task analytic models.
While help and documentation is a worthy epilogue to any course on the design and development process, our research indicates that this is considered to be a good general practice for software engineering and technology development, not necessarily an HCI-specific topic area.
A majority of our participants consider heuristics, one component of user interface standards, important. This supports its inclusion in HC/ Designing Interaction.

(P1 L 10??)

Appendix: Supplemental materials

This document includes the raw data that informed our reflections on the Strawman curriculum. Survey data is presented first, interview data is presented second, and a review of current course in HCI is presented third.
HCI Education Survey Data

We asked 338 participants to reflect on 118 topics.

The ten most important topics are:

Survey Item % considering important
1 General qualitative research 89%
2 Interaction design (as method) 89%
3 Mobile 89%
4 Interviews 87%
5 Interaction design (as topic) 86%
6 Prototyping (general) 86%
7 General empirical research and methods 86%
8 Observation 85%
9 Paper/ low-fidelity prototyping 85%
10 Empirical research 85%

Data used in the evaluation of HC/ Foundations:

Topic in HC/Foundations Related survey item % considering important
Scope of HCI Mobile 89%
Scope of HCI History of HCI 54%
User-centered design process Empirical research 85%
User-centered design process Agile/ iterative design 77%
Social models Ethnography 79%
Social models Sociology 42%
Social models Cultural studies 36%
Cognitive/ social models Psychology 50%
Cognitive models Cognitive science 60%
Principles of good design Design (general) 70%
Accessibility/ Interfaces for different populations Universal design 55%
Physical capabilities Ergonomics 42%

Data used in the evaluation of HC/ Designing Interaction:

Topic in HC/ Designing Interaction Survey Item % considering important
Design fundamentals Design (general) 70%
Methods in the design process General qualitative methods 89%
Methods in the design process Interviews 87%
Methods in the design process Observation 86%
Methods in the design process Paper/ low-fidelity prototyping 85%
Methods in the design process Usability testing 82%
Methods in the design process Experimental methods 79%
Methods in the design process Field study/ ethnography 79%
Methods in the design process Brainstorming 77%
Methods in the design process Scenarios/ storytelling 75%
Methods in the design process Task analysis 66%
User Interface Standards Heuristic evaluation 65%

Data used in the evaluation of Electives:

Elective Item % considering important
Programming Interactive Systems Interaction Design 89%
Programming Interactive Systems High-fidelity prototypes 74%
Programming Interactive Systems Information Architecture 57%
Design for non-mouse interfaces Mobile 89%
Design for non-mouse interfaces Tablet 76%
Design for non-mouse interfaces Desktop 76%
Design for non-mouse interfaces Shared interfaces 58%
Design for non-mouse interfaces Embodied interfaces 58%
Design for non-mouse interfaces Large interfaces 53%
Design for non-mouse interfaces Sensor 70%
Design for non-mouse interfaces Location 68%
Design for non-mouse interfaces Keyboard 68%
Collaboration and communication Social media 53%
Collaboration and communication Social computing 53%
Collaboration and communication Social network analysis 30%
Collaboration and communication CSCW 51%
Statistical methods for HCI Statistics 71%
Statistical methods for HCI General quantitative methods 84%
Statistical methods for HCI General empirical research and methods 86%
Design-oriented HCI Interaction design 89%
Design-oriented HCI General design 70%
Design-oriented HCI Design elements and practice 69%
Mixed, Augmented, or Virtual Reality Augmented reality 29%
Mixed, Augmented, or Virtual Reality Game studies 27%
HCI Education Interviews: The Computer Science Perspective

12 Interview participants hold a B.A., M.S., or PhD in Computer Science. These participants agree that the following skills and competencies are important for Computer Science undergraduates:

• An understanding of the iterative design process including individual methodologies (n= 6)
• Major theories and conceptual models in HCI (n= 6)
• Psychology (n= 4)
• Methodologies and epistemologies from the social sciences (n= 3)
• How to evaluate different interfaces and forms of interaction (n= 2)

8 participants are Computer Science Professors, and consider the following important:

• An understanding of the iterative design process including individual methodologies (n= 7)
• Qualitative research (n= 3)
• Elements of “design thinking” such as the ability to “understand the problem and understand the solution at the same time,” gained through studying and practicing design (n= 3)
• Major theories and conceptual models in HCI (n= 2)

The perspective of our interview participants is generally compatible with the content covered in HC/Foundations and HC/Designing Interaction. One notable incompatibility is our participants’ emphasis on qualitative research, which is not currently included in the curriculum.

HCI Education research: review of existing courses

We reviewed 53 courses in HCI from institutions in the United States, Canada, Great Britain, Germany, the Netherlands, and France. These courses were put in non-exclusive categories based on the material that they covered. The vast majority of these courses covered foundational theory or methods, and required students to complete significant projects:

Course Component Number of courses
Foundational theory or methods 50 (94%)
Foundational theory 34 (64%)
Design and/ or empirical research methods 46 (87%)
Project requiring students to build an interface or conduct research 50 (94%)
Build a high-fidelity prototype or functional interface 33 (62%)
Conduct design or empirical research 17 (32%)

safincher
Response to CS2013 HCI Strawman comments
field_vote: 
0
Your rating: None

Gerrit-

We thank the SIGCHI executive committee for their thoughtful engagement
with the IEEE/ACM CS2013 curriculum and their careful comments on the
HCI component. Please pass our thanks to them.

With regard to the three primary suggestions, we very much agree with
the points you make. They are especially important within HCI as a
discipline. However, we are working within the constraints of a Computer
Science undergraduate curriculum, for which 8 “core” hours are
allocated. We have – purposefully – taken a conservative approach,
building bridges where possible to elements that appear elsewhere in the
CS curriculum. It has been interesting (in some ways) to be constrained
to think of only how much (and what) HCI is appropriate for "straight"
CS students. You survey will, of course, speak much more comprehensively
to HCI as a specialist area.

> 1. Design and empirical research methods are central within HCI as a
> discipline. We recommend increasing the number and variety of
> methods covered in HC/ Designing Interaction. These methods should
> span the entire software design and development cycle and include
> methods for ideation, prototyping, and evaluation.

We agree with this, and have included many cross-references to the
Software Engineering KA especially in the “User-centred Design and
Testing” elective unit. It is our hope that some institutions will build
on first year courses in Software Engineering to consider these methods
a natural part of the unit test/system test cycle.

> 2. Qualitative methods are at least as important as quantitative
> methods. We would like HC/ Designing Interaction to have a general
> qualitative methods requirement and also list specific topics such
> as interview, observation, and/or ethnography.

We agree that qualitative methods are important, and have these in the
“User-centred Design and Testing” elective unit. Our hesitation in
including them in core tier one requirement is that many small
institutions will not have the specialist faculty able to teach this.

> 3. HCI increasingly incorporates subjects, topics, and
> epistemologies from the social sciences. We recommend emphasizing
> cognitive and social models over physical capabilities in HC/
> Foundations to reflect this growing incorporation of social science
> concerns and perspectives. Specific topics of interest include
> anthropological theory/methods, sociological theory/methods, and
> social psychology/methods.

We think these are, indeed, special topics of interest. Even in
HCI-enlightened departments many of our colleagues would even so
struggle to have sociological methods (for example) accepted into the
core curriculum, whereas these approaches are bread-and-butter to other
HCI colleagues who work, for example, in Sociology departments or Design
Schools. We recognise there are tensions and that we need to make
choices. We have, in general, tried to err on the side of conservatism
and make choices that reflect mainstream CS departments. At the moment
our 8 "core" hours are very unthreatening, very "safe", well-supported
by textbooks and online resources. In one sense, the fact that we are
able to do this may reflect a certain HCI maturity: some things get to
be boring preliminaries because everyone has to know them/do them.
With regard to the core units, we have followed your suggestion and
incorporated usability heuristics and the principles of usability
testing into “Foundations”; similarly we have added qualitative aspects
of task-analytic models and generalised low-fidelity prototyping in
“Designing Interaction”. The remainder of the material you suggest here
we think is already represented in the elective “User-centered design
and testing”.

With regard to your suggestions on elective topics:

> • HC/ Programming Interactive Systems Recommendations:
> Universities should offer this valuable course to non-majors as well
> as computer science majors to support the interdisciplinary nature of
> HCI.

We certainly hope that this might be the case.

> • HC/ User-centered design and testing Recommendations: An excellent
> course, this offers core material (e.g., techniques for gathering
> requirements and presentation of requirements) that are core to HCI
> material and, as such, should added to HC/ Designing Interaction.

We cannot add this quantity of material without exceeding the 8
allocated hours.

> • HC/ Design for non-mouse interfaces Recommendations: Preserve this
> timely elective, and consider including new input modalities and
> methods of data collection to complete a “new interactive
> technologies” course.

We thank the committee for these suggestions which we have incorporated
into the KA, and indeed have changed the name to match their recommendation.

> • HC/ Collaboration and communication Recommendations: Consider
> adding a CSCW component covering the topics listed above to examine
> collaboration and communication in both work and recreational
> environments.

Although we have avoided the term “CSCW”, the specific topic areas you
identify are valuable, and we have incorporated all these.

> • HC/ Statistical methods for HCI Recommendations: Universities
> should offer this valuable course to non-majors as well as computer
> science majors to support the interdisciplinary nature of HCI.

We hope that this might be the case, although in our experience, we
believe that it is as likely that CS departments will “buy in” such
courses from Psychology departments.

· HC/ Human factors and security Recommendations: Preserve this
timely elective, but include a stronger focus on the user-centered
aspects of secure systems design/development.

In drafting this elective, we were aware that other ACM communities are
also considering specialist curriculum content that is oriented toward
security. Our "straw-man" proposal is in part a place-holder to ensure
that any security electives in the new curriculum do take a
user-centered approach, even if this is not explicitly recognised as an
HCI topic. The one thing we want to avoid is fragmentation of the
security community, into a "technical" (cryptography and systems) stream
and "user issues" (HCI and sociopolitical) stream. We would welcome
assistance from the SIGCHI committee in establishing these bridges, and
ensuring that the ACM security curriculum meets our goals. We would
particularly welcome any material collected during your survey, of
existing security curricula taking this integrated approach.

> · HC/ Design-oriented HCI Recommendations: Universities should offer
> this valuable course to non-majors as well as computer science majors
> to support the interdisciplinary nature of HCI.

We agree. There are many aspects of the ACM Computer Science curriculum
that would be of value to non-majors. Although ACM has little influence
on curricula outside of Computer Science, some of our working group are
actively engaged in the broader dissemination of socially-informed
design thinking across the educational spectrum. We would be very happy
to make contact with others who have effective channels of influence
along these lines.

> · HC/ Mixed, Augmented or Virtual reality Recommendations: While
> this elective does not reflect the immediate priorities of the HCI
> community it does represent an emerging area of research that should be
> supported.

We agree. As in the case of security, our goal in drafting this elective
was to recognise that other sub-fields of CS already teach these topics
without considering an HCI perspective. We think it is likely that an
elective along these lines will appear somewhere in the ACM curriculum,
and we wanted to ensure that it addresses HCI concerns as well as
technical ones.

Finally, we were very pleased to hear of your systematic review of 53
HCI courses. This will be a valuable resource - we didn't realise that
this had been planned, based on the “preliminary report” circulated in
July.

What was the methodology you employed? We would be particularly
interested to know if you had access to planning documents as well as
delivery materials. How were the courses solicited (were they from CS
departments, or more widely distributed)? We have gathered only a very
limited number of course exemplars (such as you yourself completed) and
would be keen to further understand the opportunities and constraints
that inform HCI curricula choices. If you have interviewed the teachers
from these 53 courses, their reflections will be invaluable in shaping
the revised curriculum. If you'd like to pass on this material, we'd be
very happy to build on it further.

- Sally

Log in or register to post comments