"Violating Fit's Law" [sic]: Fitts' Law Beyond HCI Education

Abstract

Fitts' Law is, more often than not, thought of as standard knowledge in the field of Human-Computer Interaction. However, in many occasions, interface components are being built by professionals with little or no knowledge of HCI principles (e.g., software menus are often being built by software developers with no designer input). How much appreciation of Fitts' Law exists among developers, and how does it affect our collaboration with them and the software they build and we use? To help answer this question, a survey among 65 designers and developers was conducted, as well as an analysis of the Computer Science undergraduate curricula from 94 internationally high-ranked universities. The contribution of this paper is twofold: firstly, treating Fitts' Law as an Open Educational Resource and, secondly, the analysis identified gaps that extend beyond HCI Education and fall in the sphere of Epistemology.

Full text

“Violating Fit’s Law” [sic]: Fitts’ Law
Beyond HCI Education
Abstract
Fitts’ Law is, more often than not, thought of as
standard knowledge in the field of Human-Computer
Interaction. However, in many occasions, interface
components are being built by professionals with little
or no knowledge of HCI principles (e.g., software
menus are often being built by software developers
with no designer input). How much appreciation of
Fitts’ Law exists among developers, and how does it
affect our collaboration with them and the software
they build and we use? To help answer this question, a
survey among 65 designers and developers was
conducted, as well as an analysis of the Computer
Science undergraduate curricula from 94 internationally
high-ranked universities. The contribution of this paper
is twofold: firstly, treating Fitts’ Law as an Open
Educational Resource and, secondly, the analysis
identified gaps that extend beyond HCI Education and
fall in the sphere of Epistemology.
Author Keywords
Fitts’ Law; HCI; Education; Epistemology
ACM Classification Keywords
H.5.m. Information interfaces and presentation (e.g.,
HCI): Miscellaneous. K.3.m. Computers and Education:
Miscellaneous.
Paste the appropriate copyright/license statement here. ACM now
supports three different publication options:
• License: The author(s) retain copyright, but ACM receives an
exclusive publication license.
This text field is large enough to hold the appropriate release statement
assuming it is single-spaced in Verdana 7 point font. Please do not
change the size of this text box.
Every submission will be assigned their own unique DOI string to be
included here.
Evangelos Kapros
Learnovate Centre, Trinity
College Dublin, The University
of Dublin
Unit 28, Trinity Technology &
Enterprise Campus, Pearse
Street, Dublin 2, Ireland
evangelos.kapros@learnovate
centre.org

Introduction
We all know what Fitts’ Law is. Or do we? The answer
may depend on who “we” are, and on how we
understand scientific laws. Fitts’ Law is a sine qua non
in HCI Education. However, many—if not most—
practitioners in the field would not have undertaken
traditional HCI Education: they would originate from
either non-technical backgrounds such as (non-applied)
psychology and design, or from engineering and
computing. The interdisciplinarity of HCI practitioners
has offered a great deal to the field, and many courses
in the aforementioned fields offer elective HCI modules;
however, it is not expected that a graduate of these
courses would be necessarily aware of what HCI is.
Working in groups where the majority of members have
little or no awareness of HCI can hinder communication
around a design problem—or a proposed solution
thereof—, and can potentially lead to designing for poor
user experiences.
Finding a balance between the desired
interdisciplinarity of a group and a mismatch in
understanding is no easy task. Difficulties are manifest
in a large sub-field of HCI, which is software
construction. The designer/developer workflow is a
constant example of workplace friction, and a rich
source of inspiration for imaginative start-ups that
produce tools promising to ease the workflow. While an
increasing number of organisations values design more
than in the past and the number of start-ups with
designer founders is reportedly increasing, the
conceptual mismatch in the designer/developer
workflow is still present.
In this paper, the aim is to consider Fitts’ Law as an
example of the designer/developer mismatch. Do
designers and developers perceive Fitts’ Law in
different ways? If yes, is this a result of HCI Education?
If not, what can be still said about the apparent
conceptual mismatch?
Related work
Fitts, after having worked for a number of years with
the U.S. Air Force on the psychological aspects of
aircraft displays [9], generalised his work on controlling
the amplitude of movement. His experimental work was
seminal and has resulted in a series of laws [10]. The
law describing how the response time of controlling a
target is proportional to the distance from and the size
of the target has been named the “Fitts’ Law” and been
considered as a law of paramount importance for the
field of Human-Computer Interaction [24].
Figure 1. A figure from the original Fitts’ experiment (Public
Domain) [10].
Fitts’ Law has played a big role in HCI research, as it
has been viewed through multiple lenses. It has been
shown that the Law holds for tablets; for hand-gesture
detection; for tracking certain eye movements; for

touchpads; for (some) motion kinetics; or for screen-
edge pointing [1, 2, 14, 16, 22]. It has been shown not
to hold for some radial menus [11]. It has been viewed
as a time/error tradeoff, as a research and design tool
in HCI, or as a research tool for the perception of user
performance [13, 15, 18]. Others have focused on the
specific variations of the formula that expresses Fitts’
Law [7]. A multitude of academic research articles and
blog-posts have been written about it.
Figure 2. A search for “Fitts” in the ACM Digital Library is
indicative of the popularity of Fitts’ Law.
With regard to the field of Education, previous research
has built upon traditional HCI for Education work to
show that research projects around experimenting with
laws such as Fitts’ one are beneficial to CS
undergraduate students [20].
More broadly speaking, some previous work describes
case studies where HCI has been integrated, in one
way or another, in a CS curriculum [4, 5, 6, 8, 12, 17,
19, 21]. The need for interdisciplinarity has been
emphasised; specifically, the joint ACM/IEEE CS 2013
curriculum especially recognises the need to “provide
students with the flexibility to work across many
disciplines” and to cover various knowledge areas in
introductory courses (it specifically includes HCI in both
Tier 1 and Tier 2, which is great) [23].
Thus, a study on the actual situation in CS curricula
seems to be necessary; is the recommendation of the
CS2013 curriculum to include HCI in Tier 1 Core
modules actually being implemented? Such a study is
presented below.
Fitts’ Law beyond HCI Education
A study on the inclusion of HCI in CS curricula should
try to address the following two issues. Firstly, do HCI-
related concepts appear in Tier 1 modules? Secondly,
what are the implications of not including HCI in core
CS?
HCI Education in CS Curricula
To investigate if HCI-related concepts appear in CS core
modules the following approach was followed.
Firstly, a list of universities was compiled by
triangulating the results of three prominent university-
ranking providers 1. The top 75 universities in CS from
the 2014 Shanghai Academic Rankings, the top 50
universities from the 2014 QS University Rankings in
CS and Information Systems, and the top 50
universities from the 2014-2015 Times Higher
Education ranking in Engineering and Technology. This
list yielded a list of 94 universities from 11 countries.
For the sake of simplicity (no political connotations
should be drawn from this!), universities from the EU
and Switzerland were filed under “EU” as they use the
structures of the European Higher Education Authority
(EHEA) and, similarly, universities from China and Hong
Kong were filed under “China”.
1 Disclosure: I participated in the 2014 QS World University
Rankings Survey as a respondent.

Figure 3: A graph of the—skewed!—distribution of the top
ranked universities for CS per country. Triangulating rankings
from three different providers composed the list.
For these 94 universities, their CS curricula were
located through their websites. One undergraduate
Bachelor’s or Major in CS program per university was
chosen to be included in the study. Computer
Engineering, Information Systems, Interaction Design,
or Multimedia Design programs were excluded, as they
are out of scope for this study.
Then, for each CS course the following information was
collected. Firstly, if the course offers an introductory
course to CS, in which various fields of computing are
represented. Possible answers are {yes, no,
programming}, where “yes” means that an
Introduction to CS exists and covers various CS fields,
“no” that such an introduction to CS does not exist, and
“programming” that an introduction to problem solving
exists, but is entirely focused on algorithmic thinking
and using a programming language. Secondly, data
were collected with regard to the course including HCI
topics in the introductory CS modules, as an obligatory
HCI module, as an elective module, or not at all.
Limitations in data gathering resulted in analysing 82
out of 94 programs. Some programs offered
information only in the native language of the country
where they are offered: machine translation helped
successfully with some of these, but not all curricula
pages could be located. Another limitation was the lack
of detailed descriptions in some courses, which may
have mentioned that they offer an introductory CS
course, but its syllabus is not publically available.
Finally, one website was not available due to technical
failures during data collection.
All effort was put into gathering the latest information,
namely for the 2014-2015 academic year. An apparent
weakness of this method is that it may identify current
patterns is curricula, but the direction towards these
curricula will be developed is not clear. Thus, a
university that currently considers moving to the
CS2013 curriculum, and one that does not, will appear
the same, if their current offerings are the same.
However, considerations to change the curriculum
towards CS2013 are rarely announced in public, making
such a data collection almost impossible.
The list of universities and the full data are available at:
http://bit.ly/1JYF03A
The results of the data analysis are presented in the
next section. Before that, the following sub-section
describes the second part of the study around the
implications of excluding HCI from core CS curricula.

Designers’ and Developers’ Perception of Fitts’ Law
To define all implications of excluding HCI from Tier 1
CS courses in no easy task. For this reason, this study
is limited in identifying a small, but important, subset
of the HCI body of knowledge: its iconic Fitts’ Law.
The rationale originates from a real case in Learnovate
Centre around an educational software application. On
an in-house re-design of a content-composition
application for Windows 8.1 devices, a discussion arose
among our designers and developers around the design
and implementation of the navigation menu. The
original impression was that the native menu (see
Figure 4) had to be implemented, but eventually a
different mechanism was designed and implemented.
Figure 4. The navigation menu on the Microsoft Surface tablet device. This illustration uses a
mockup of the Windows 8.1 operating system.. Your images should be higher quality, so
readers can clearly see the details.
A gesture on the bezel of the
device, that is a drag down from
the bezel into the screen reveals
the navigation menu. The low
affordance of this menu, the small
height of the device’s top bezel,
and the limited bezel width that is
at reach each moment may hinder
its use.
Right-clicking when using a mouse
reproduces the same behaviour.

However, during the discussion around the native
menu, various web searches revealed that many
developers had a flawed perception about what Fitts’
Law is. This flawed perception seemed to have affected
the implementation of menus by the said developers.
Specifically, a perception that bad menu design
somehow “violates” Fitts’ Law seems to be particularly
popular. The discussion about how a menu “violates” or
“breaks” [sic] Fitts’ Law is taking place in popular
developer blogs and forums, and even in corporate
developer forums (in start-ups and multi-nationals
alike).
To validate the above, a survey related to Fitts’ Law
and menu design was designed and circulated to
designers and developers.
The survey consisted of a short introduction to Fitts’
Law, a question about whether the menu of Figure 4
“violates Fitts’ Law” [sic], and questions about the
occupation of the respondent, their gender, age, and
ethnicity. The demographic questions were not asked to
validate a pre-existing hypothesis, but rather to allow
for better data analysis.
The survey was circulated to various user groups
through email and social media (Twitter, and LinkedIn
SIGCHI and IxDA Groups).
The next section presents the findings from the CS
curriculum analysis and the survey.
Findings
HCI in CS Core Curricula
Out of the 82 analysed CS undergraduate curricula,
only 11 (~13%) follow the ACM/IEEE CS 2013
recommendation, and 4 (~5%) other universities offer
alternative options that also incorporate HCI in core
modules. The 11 universities that include HCI in their
Tier 1 employ different methods, ranging from briefly
mentioning HCI in their introductory CS courses, to
offering obligatory HCI modules. Alternative HCI
incorporations include designing UIs at projects, and
there is also a singular case where a CS program offers
modules from a Design Academy in its core
curriculum—not as a “dual degree” option (the Hebrew
University of Jerusalem offers a CS program with
Bezazel). A list of these institutions is below in Table 1.
University
Country
Has HCI in CS
Core?
Massachusetts
Institute of
Technology
USA
Yes
Nanyang
Technological
University
Singapore
Yes
The Hong Kong
University of Science
and Technology
Hong
Kong
(China)
Yes
University College
London
EU
Yes
University of
Washington
USA
Yes
University of
Copenhagen
EU
Yes
University of Tokyo
Japan
Yes
Australian National
University
Australia
Yes
>

University
Country
Has HCI in CS
Core?
University of
Queensland
Australia
Yes
Yale University
USA
Yes
Rice University
USA
Yes, but focuses
on Game Design
Georgia Institute of
Technology
USA
Students can
take HCI while
on CS ‘Thread’
Columbia University
USA
Potentially in
projects
The Ohio State
University
USA
Potentially in
projects
The Hebrew
University of
Jerusalem
Israel
In CS program
with Bezazel
Academy of Art
and Design
Table 1. The 11 top-ranked universities that include HCI as
part of their core, Tier 1 CS undergraduate offering, as well as
the 4 universities that offer alternative paths in their CS
curriculum that may include some HCI in a core module.
The majority of curricula, that is 42 (~51%) of them,
offer HCI as an elective module, and do not mention
HCI in their core. However, in all 42 curricula of this
kind, it remains unclear why the student is expected to
elect HCI as a module: the field is not introduced at any
point, and it is unclear what mechanisms are in place to
motivate students elect HCI. Thus, students may
choose to attend the elective HCI module for
pedestrian, rather than academic reasons (e.g., a
lecturer of their liking, easier access to previous exams’
solutions, or similar).
Other institutions, more specifically 24 universities
(~29%), do not include HCI at all in their CS courses,
neither mentioned in an “introduction to CS” course nor
as an elective course. Moreover, an institution has an
HCI course in their course, but it is not offered in 2014-
2015 (see Figure 5).
Figure 5: The majority of top-ranked universities do not
mention HCI as a field in their core CS undergraduate
curriculum.
A geographical analysis of the results did not identify
specifically interesting patterns. In general, in the USA,
Australia, and Europe a variety of offerings has been
adopted, while Canadian, Chinese, Korean, and
Taiwanese institutions tended to follow one model
(offering elective HCI courses, or no HCI at all).
However, this difference may have been influenced by
the larger sample size for some countries than for
others, and should not be considered definite.
In conclusion, approximately 82% of the top-ranked
universities do not actively motivate future computer
scientists to further explore HCI, and 29% do not even
offer basic knowledge about what HCI is. Only 18% of
the institutions facilitate an informed choice with regard
to studying HCI.

Fitts’ Law Developers’ Perception Survey
Out of the 65 respondents of the survey, 19 (~29%)
replied that the menu of Figure 4 “violates” Fitts’ Law,
and 31 (~47) replied that it does not. 15 respondents
(~23%) typed in a response in the “Other” field, of
whom only 5 (~8%) respondents questioned the notion
of what it means to violate Fitts’ Law altogether (Table
2).
Resp
onde
nt
Job
Function
Gender
Age
Ethnicit
y/ies
#16
Designer
Male
45-54
White
#28
Developer
M
35-44
Prefer
not to
answer
#57
Researcher
Female
25-34
White
#60
Designer
Male
25-34
Hispanic
#64
Designer
F
35-44
White
Table 2. Demographics for the 5 respondents who questioned
the notion of “violating” Fitts’ Law altogether.
Respondent #64 summarised nicely that: “I do not
know how Fitt's law can be violated if it is a function
that returns time.”
Respondent #28, the only developer who questioned
the “violation” notion, gave a practical example to
explain his rationale: “Say Zig's law is, a person who
chooses a center seat in a theater walks further than a
person who chooses an aisle seat. One person chose an
aisle seat. Is Zig's law violated? Absolutely not, a
person who chooses a center seat walks further than a
person who chooses an aisle seat. The law does not
apply to what decisions can or cannot made, it applies
to the consequences of those decisions. The particular
design decision made by Microsoft does not change the
fact that design decisions have consequences, and all
Fitt's law states is that design decisions have
consequences.” The other 3 respondents in Table 2 had
a similar line of thought.
Other respondents who typed in the “other” field but
didn’t question the “violation”, perceived Fitts’ Law in
different ways. A (male, designer, white, 35-44)
respondent (falsely) replied that Fitts’ Law is only
relevant in drag-and-drop operations, and not when
reaching a target: “Fits' Kaw is about the target (in this
case, where you drop the item) and not about the start
location” [sic]. A (male, white, 35-44) developer replied
that: “because it is an established pattern on mobile
i.e. to drag a menu or utility tray from the 'bezel', it
works.”
In general, developers were more likely to say that the
menu “violates” Fitts’ Law than designers (Table 3).
Moreover, developers were less likely to choose the
“other” option and type in their own answer.
With regard to other demographics, there were 12 male
and 3 female developer respondents, and 13 male and
8 female designers. Both designers and developers
were predominantly white, 11 developers (~69%) and
17 designers (~77%). Designers were slightly older
than developers. Moreover, it is worth noting that
female respondents were more diverse ethnically, and
with regard to age, and predominantly designers (60%
designers, 25% researchers, 15% developers).
Considering ethnicity, 2 out of 20 non-white
respondents (8 Asian/Pacific, 2 Black/African American,
7 Hispanic, 3 preferred not to answer) identified the
Developers
Answer
Respo
nses
Percent
age
Yes
8
44.45%
No
7
38.89%
Other
3
16.67%
Designers
Answer
Respo
nses
Percent
age
Yes
4
18.19%
No
12
54.54%
Other
6
27.27%
Table 3. Number of responses and
percentages per group for the survey
question based on Figure 4: “In your
opinion, does this drag-from-bezel
menu violate Fitts' Law?”
7

issue with “violating” the law, while 3 out of 44 white
respondents did (10% and 7%, respectively).
Conclusion
In conclusion, when presented with a concrete menu
design problem at hand, most designers and developers
who participated in the survey (~87%) did not question
the premises presented to them. Moreover, only 8%
achieved to re-frame the problem in a way that makes
sense from an HCI point of view.
At first glance, this seems to be a very limited result
about a specific law and how it applies to a particular
menu. However, it reveals underpinning misconceptions
about what a scientific law is and how it can be violated
(e.g., a boat floating on water does not violate the law
of gravity).
In addition, 82% of top-ranked universities have not
yet adopted the recommendations of [23] to include
HCI in their core CS undergraduate curriculum.
The effectively active epistemological misconceptions
about scientific laws and the lack of computer scientist
exposure to HCI may hinder the communication and
inhibit the understanding between HCI practitioners or
researchers and computer scientists, i.e., the very
people who, most of the time, implement the solutions
we design.
In the current times when a typical enterprise consists
of 15 people [3], an 8% of people able to re-frame a
problem in a scientifically appropriate way would mean
approximately 1 person per organisation. Given how
group dynamics work, a single person may find it
impossible to re-frame a related problem when
necessary.
Of course, other factors may also exist, and further
research should consider design curricula, life-long
learning, and any related non-education factors.
Ultimately, this paper is not an attempt to list an
exhaustive list of factors, but rather to begin a dialogue
which will link HCI Education to actual CS curriculum
design.
Acknowledgements
This research is supported by the Learnovate Centre at
Trinity College, the University of Dublin. The
Learnovate Centre is funded under the Technology
Centre Programme through the Government of Irelands
state agencies Enterprise Ireland and IDA Ireland.
References
[1] Appert, C., Chapuis, O., and Beaudouin-Lafon, M.
Evaluation of pointing performance on screen edges.
Proceedings of the working conference on Advanced
visual interfaces - AVI ’08, (2008), 119.
[2] Brown, M.A. and Stuerzlinger, W. The Performance
of Un-Instrumented In-Air Pointing. 59–66.
[3] Bureau, U.S. and Statistics, L. Monthly Labor
Review, March 2012: The declining average size of
establishments: evidence and explanations. March
(2012).
[4] Chan, S., Wolfe, R., and Fang, X. Teaching HCI in
IS/EC Curriculum. Eighth Americas Conference on
Information Systems, (2002), 1011–1020.
[5] Cockburn, A. and Bell, T. Extending HCI in the
computer science curriculum. Proceedings of the third
Australasian conference on Computer science education
- ACSE ’98, ACM Press (1998), 113–120.
[6] Douglas, S., Tremaine, M., Leventhal, L., Wills,
C.E., and Manaris, B. Incorporating Human-Computer
Interaction into the undergraduate computer science
curriculum. ACM SIGCSE Bulletin 34, 1 (2002), 211.

[7] Drewes, H. Only one Fitts’ law formula please!
Proceedings of the 28th of the international conference
extended abstracts on Human factors in computing
systems - CHI EA ’10, ACM Press (2010), 2813.
[8] Fischer, G. Transdisciplinary Education and
Collaboration. Lifelong Learning, (2008), 1–18.
[9] Fitts, P.M. Psychological Aspects of Instrument
Display. 1. Analysis of 270 ‘Pilot-Error’ Experiences in
Reading and Interpreting Aircraft Instruments. (1947).
[10] Fitts, P.M. The information capacity of the human
motor system in controlling the amplitude of
movement. Journal of Experimental Psychology:
General 121, 3 (1992), 262–269.
[11] Friedlander, N., Schlueter, K., and Mantei, M.
Bullseye! when Fitts’ law doesn't fit. Proceedings of the
SIGCHI conference on Human factors in computing
systems - CHI ’98, (1998), 257–264.
[12] Greenberg, S. Teaching human computer
interaction to programmers. interactions 3, 4 (1996),
62–76.
[13] Guiard, Y. and Perrault, S.T. Fitts ’ Law as an
Explicit Time / Error Trade-Off. (2011), 1619–1628.
[14] MacKenzie, I.S. and Oniszczak, A. A comparison of
three selection techniques for touchpads. Proceedings
of the SIGCHI conference on Human factors in
computing systems - CHI ’98, (1998), 336–343.
[15] MacKenzie, I.S. Fitts’ law as a research and design
tool in human-computer interaction. Hum.-Comput.
Interact. 7, 1 (1992), 91–139.
[16] Mandryk, R.L. and Lough, C. The effects of
intended use on target acquisition. Proceedings of the
2011 annual conference on Human factors in computing
systems - CHI ’11, (2011), 1649.
[17] Moore, G. and Lottridge, D. Interaction design in
the university. Proceedings of the 28th of the
international conference extended abstracts on Human
factors in computing systems - CHI EA ’10, ACM Press
(2010), 2735.
[18] Nicosia, M., Oulasvirta, A., and Kristensson, P.O.
Modeling the perception of user performance.
Proceedings of the 32nd annual ACM conference on
Human factors in computing systems - CHI ’14, (2014),
1747–1756.
[19] Pastel, R., Brown, C., Woller-Carter, M., and
Kumar, S. Teaching Human Factors to Graduate and
Undergraduate Computer Science Students.
Proceedings of the Human Factors and Ergonomics
Society Annual Meeting 56, 1 (2012), 595–599.
[20] Pastel, R. Integrating science and research in a HCI
design course. ACM SIGCSE Bulletin 37, 1 (2005), 31.
[21] Rusu, C., Rusu, V., Roncagliolo, S., and Rubio, J.M.
A Systematic Approach to Usability Practices in
Computer Science Curricula. CLIHC, (2007).
[22] Surakka, V., Illi, M., and Isokoski, P. Gazing and
frowning as a new human--computer interaction
technique. ACM Transactions on Applied Perception 1, 1
(2004), 40–56.
[23] The Joint Task Force on Computing Curricula
(Association for Computing Machinery / IEEE-Computer
Society). Computer Science Curricula 2013: Curriculum
Guidelines for Undergraduate Degree Programs in
Computer Science. ACM, Inc, 2013.
[24] Thimbleby, H. Targeting the Fitts Law. Interactions
20, 1 (2013), 12.