Executive Summary
The enclosed paper will explore some issues drawn from sociological and psychological research
that have implications for the gathering of software requirements related to usability of software. The
issues are divided into “problems”, and “suggestions for improvement”.
The first of these problems is the instability of preferences, which suggests that new or radical
approaches to a problem will often be met with resistance from an audience being asked to describe their
feelings about the approach because they do not know how to describe those feelings; they will prefer
things about which they are better able to articulate their feelings. Since many usability approaches offer
user interfaces or interactions that are new or radical compared to their predecessors, this is something to
be mindful of.
Another problem of interest is the intersection of the “paradox of choice” and the “long tail” of a
market. The “paradox of choice” is that an abundance of choice contributes to paralysis in decision
making and less satisfaction in the final decision made, while the “long tail” describes the many smaller
submarkets within a market that, individually, are insignificant when compared to the critical mass, but
that, in aggregation, represent a significant market. Neither are satisfied by the existing market. By
reducing larger markets to smaller submarkets, the “paradox of choice” is addressed and it is possible to
sell to the “long tail” of the market.
The final problem addressed is that of the conflicting requirements of commercial interests and
user interests. The example of Digital Rights Management (DRM) is used to show how commercial
interests can make a problem excessively complex and possibly turn users to black markets simply
because the illicit alternatives are much easier to use.
The first suggested improvement is the use of a technique called “personas”, derived from
usability research, which builds a model of a user that has a personality and real goals with the hope that
the persona will be sympathized with when building software. Goals are often much different than
functional requirements, yet the satisfaction of goals will lead to passionate users, which is a very
desirable type of user to have.
Secondly, positive aesthetics and visceral reactions can offset usability problems by connecting
with the user at a different level than software usually does. A product that is attractive in this sense will
entice a user to try something, to make them want to keep coming back, and will make a user more
forgiving of usability shortcomings that may be essential (rather than accidental) problems of
contemporary software development.
Third, it may be possibly to partially negate the issue of preferences by making a product that is
sufficiently freeform that a user may not need to be held to an interaction structure that is too rigid. Two
modern tools that utilize this technique are identified.
Fourth, the use of weblogging to actively participate in a product’s user community keeps
developers in touch with the people they are developing for, and ensures that they are shown as being
responsive to customer concerns. The ways in which search engines rank Internet web pages are causing
customer discussion about products to appear alongside product web pages in search result listings, and
some products can be and have been marketed using this technique.
Finally, the possibility of open source being used to address the problem of commercial interests
is discussed, and reasons are given that it may not be a solution to the problem. However, a solution may
lie in a combination of closed and open source code.
Contemporary issues in gathering software usability requirements 1
Contemporary issues in gathering
software usability requirements
Matthew Buckley-Golder
University of Maryland University College (student)
Contemporary issues in gathering software usability requirements 2
AbstractThere is more to a software product than its functional requirements. The usability of such products has an impact on the initial experience of a user with the product, their ongoing experience, and the likelihood that they will be a repeat customer or user of a product that you have developed. In addition to usability, there are characteristics of the current market and operating environment which should be considered in order to arrive at a definition of a product that is more likeable and more responsive to the social and psychological needs of its users and customers. Going beyond the functional needs of a user community in order to create passionate users will more likely foster a sense of loyalty and contribute to ongoing patronage or employment.
Contemporary issues in gathering software usability requirements 3
Contemporary issues in gathering software usability requirements
Introduction
The focus of the following discussion will be on some of the issues facing the gathering
of usability requirements for contemporary software, with varying applicability to the
development of custom software and of commercial off-the-shelf (COTS) software. In particular,
it will focus on the psychological and social aspects of interaction with software related to
requirements gathering, rather than the technical aspects, and in doing so suggest a direction for
future usability focus. Some of the technical aspects of usability were described in an earlier
paper (Buckley-Golder, 2002). Additionally, some suggested improvements will be given which
will offset some of the identified problems and help in the gathering of usability requirements in
the future. Paying attention to these “softer” aspects of software use will appeal to users at a
different level and will have a positive effect on the overall user experience, which is difficult to
achieve with functionality alone.
Problems
This section will discuss some of the problems that occur when the concern of usability
meets with the pre-usability approach to developing software.
Instability of preferences
Wilson (1991) undertook a study which analyzes the effect of having people explain their
reasoning on the preferences that they have. The findings suggest that asking someone to explain
their reasoning about why they like something can cause someone to have different (and less
authentic) preferences than someone who is asked for what they want without having to explain
Contemporary issues in gathering software usability requirements 4
why they want it. By asking someone to consider their preferences rather than just asking for
their preferences, they change their preferences. In addition, forcing people to analyze their
preferences too heavily can reduce their satisfaction with their choices.
The study also recognizes that some people may come into a situation already knowing
why they feel the way they do about something that they are knowledgeable about, and having
such people (such as experts) analyze their preferences would not have as much effect as on
someone who has not considered their preferences about something in advance. However, as
Gladwell (2004) identifies, this still exposes people to initially react negatively to something
that is radically different from what they know; if someone is asked to analyze their preferences
about something so unique that they are not able to articulate their reasoning, they are likely to
prefer something that they are able to articulate reasoning about. His example cites Herman
Miller’s Aeron chair, an incredibly popular office chair that is loved by its users but which has an
unusual appearance. It was initially shot down in a focus group because the participants were
asked to describe why they liked its aesthetics but could not come up with the words – it did not
fit their profile of an office chair. In response they compared it to things that they thought it
looked like, which were usually things with negative implications; the focus groups were
unsettling for the product designers, but they pushed ahead and the chair is now the biggest
selling office chair in history. If the focus group results had been taken at face value, the
company would have missed out on this success.
For the above reasons, it is difficult to take preference at face value, particularly when the
object or concept being evaluated is significantly new or unprecedented. This argument supports
Contemporary issues in gathering software usability requirements 5
the use of prototyping for gathering requirements, and also the incorporation of the results of
usability research into the requirements and design of a software product. Usability research
applied to a new product that is to replace an old product will often result in a radically different
user interface that the user may find foreign compared to what they know, and therefore initially
unfavorable. However, usability research provides proven ways of making software easier to use,
and prototypes will help to demonstrate the new interface so that users can become comfortable
with it and perhaps develop a concrete preference for it.
Paradox of choice and the long tail
In The Paradox of Choice, Schwartz (2004) describes how an abundance of choice
contributes to paralysis in decision making and less satisfaction in the final decision made. He
contends that the psychological effect of loss is greater than the psychological effect of gain, and
that the opportunity costs of declining an increasing number of choices in favor of the one final
choice amount to a loss that makes us dissatisfied with the choice that we worked so hard to
make.
This pertinent issue is relevant to the be-all feature sets and user interfaces that are often
tacked on to today’s increasingly complex software. In an effort to be everything to everyone,
software often tries to tackle too many individual problems in a single product and then makes
all of the solutions to those problems available to every user of the product. The results are
problematic:
Contemporary issues in gathering software usability requirements 6
The selection of software packages for a particular domain is large because all products
of a particular class try to solve the same set of problems
It is likely that none of the products will precisely meet the user’s needs because they are
trying to solve someone else’s problem, and that “someone else” might not even exist
because it is an aggregate of individual needs; a fallacy of composition.
The prospect, therefore, that the selection of a COTS product will exactly meet the
software requirements of an organization or individual is very unlikely, unless the requirements
are restricted from the outset by what is available on the market. The COTS product may contain
a superset of the software requirements, but the addition of features beyond what is required by a
particular usage scenario detracts from the product by creating an overwhelming selection of
features that must be chosen from, not only during product selection, which is where the paradox
of choice is so apparent, but also during use of the product. If products were more tailored to
smaller subsets of customers, the product selection process would be easier because the number
of ideal products to select from would be smaller, as would the feature sets, and the use of the
products would lead to less mental friction1 because fewer options would be available within the
software’s user interface.
There has been movement to address this problem in realms where the stakes are very
high and where it would be impossible to produce a one-size-fits-all product that worked for
even a single customer; for example, the products of SAP and PeopleSoft. But, this movement
should extend to other, smaller spaces. The advent of the Internet has opened up the opportunity 1 “Mental friction” (or “cognitive friction”) is the amount of mental effort required to complete a task. By optimizing the choices and actions offered by software in a way that is compatible with what the human brain expects, friction is reduced (for example, by being consistent in the way the software behaves). High friction reduces the quality of the user experience (Cooper, 1999).
Contemporary issues in gathering software usability requirements 7
to address the needs of the “long tail” of markets, which describes the many smaller submarkets
within a market that, individually, are insignificant when compared to the critical mass, but that,
in aggregation, represent a significant market. Now that the sale of software is not restricted by
advertising space in trade publications and the shelf space in retail stores, it is much easier to
address this “long tail” by offering more diverse products. Combined with “blogging”, discussed
in more detail later, it is now feasible to have others market your product if the product is
considered worthy by the community. If this approach catches on, it may well erode the critical
mass and create opportunities for software more suitable to individual need.
Commercial interests vs. usability
Commercial interests can interfere with usability, and make some usability requirements
impossible. A prime example can be seen in the area of Digital Rights Management (DRM).
DRM is an authorization mechanism attached to digital content which determines how a user is
permitted to use that content. Unfortunately, many of these restrictions destroy the advantages of
making the content available in a digital format, and make the content significantly more
difficult to use.
When you buy a book, you are free to read it wherever you want, scan it, photocopy it,
and share it with friends. When you buy a CD, or a movie, you are free to take it wherever you
want, watch it on whatever player you want, share it with friends, make a copy for your car or
your portable player. Regardless of the legality of any of these scenarios, there is nothing
stopping you from doing it and permits a degree of freedom that is removed with DRM. After
paying nearly full price for a digital copy of a book, you must print the book (at your own
expense), which will be heavy and cumbersome to carry around, and you may only be able to
Contemporary issues in gathering software usability requirements 8
print a book once, depending on how the author decided that you should be able to use the
content. You will be restricted as to how many devices you can copy it to (if you want to carry it
digitally on a device). You will not be able to copy text from the book to the clipboard, which
would be one very useful advantage of a digital copy, particularly when citing the book as a
reference. You may not even be able to own the book indefinitely; it may have controls which
force it to stop working after a period of time defined by the seller.
In addition to the behavioral protection, there is also the question of whether or not you
will be able to continue to read your book on a device other than the one you originally copied it
to; a license sometimes ties content to a specific device. If the device fails, or you lose your
license, you cannot access the content anymore and there is usually no way to get it back. Not
only are you leaving the world of the physical for a world of symbols (digital copies), an act that
has its own usability problems regardless of the application, but you are placing unnecessary
restrictions on how you can work within the symbolic world.
The restrictions imposed by DRM do not have analogs in the physical world and,
furthermore, the restrictions are not in place to satisfy the goals of the user; in fact, they are in
place to explicitly defeat the goals of the user, so much so that sophisticated users will spend
much time reverse engineering a piece of software to make it work in a way that the authors
would never have been able to make it work because of their conflicting obligations (Borland,
2005). Likewise, users are not likely to be happy when they lose all of the DRM-protected music
because of the concentration of content on a small, fragile device that was designed to never .
Again, sophisticated users will find a workaround (Breen, 2005), but what does this do to the
Contemporary issues in gathering software usability requirements 9
user experience? Fancy marketing aside, the Apple iPod is a small, easily lost device that is
fragile in the context in which it will be used, yet by design it refuses a user the opportunity to
use the natural backup relationship between personal computer and portable device to act as a
backup for the expensive music that they encourage their users to purchase. All of a sudden, the
idea of 5,000 songs at your fingertips does not seem as enticing now that you realize this music
would cost you almost $5,000 (songs on Apple's iTunes music store are $0.99 each) to purchase
and, if you experience a PC failure, you have no way of getting that music from your iPod back
on to your PC. It is much easier to destroy the content of 500 compact discs when they are on a
mobile, handheld device than when they are sitting on shelves in your house. A truly remarkable
feature of this situation is that you are punished for acquiring your content legally – most of the
music that would be available at the iTunes music store is available through illegal means for
free through illicit sites such as Kazaa, yet would not have any of the usability challenges that are
solely related to the implementation of a DRM scheme.
Suggestions for improvement
This section will discuss some recent developments that will help toward gathering
usability requirements, or that will impact the gathering of usability requirements in the future.
Usability research: personas
Goodwin (2001) and Cooper (2003) suggest the use of a "persona" to guide decisions
about software features, interactions, navigations, and visual design. A persona is a user
archetype -- more than a model -- which have goals and patterns of behavior that can be used as
a reference when developing software. Goals go beyond job functions because they address
human needs. One goal of a persona might be to fell intelligent and not be made to feel stupid
Contemporary issues in gathering software usability requirements 10
without cause; humans don't like to unnecessarily be made to feel stupid, but a typical corporate
environment would not encourage people to express that they feel stupid and software rarely
addresses this goal; much software does make people feel stupid. Combined with the code of
silence, this problem manifests itself in employee well-being, which is one of the factors that
many corporations, in their new-age outlook, are trying to address. Goals can be life goals,
experience goals, or end goals. End goals account for most of the goals in software development
-- for example, to be more productive, or to be more organized. Experience goals describe the
experience that a user wants to have when using the product – to not feel stupid, for example, or
to be confident that online financial transactions are secure.
A persona isn't simply a list of activities and tasks. It's a flow of their day. Personas have
skills, attitudes, a work environment, and the aforementioned goals. Personas are given a face,
and are augmented with small amounts of narrative that may not be directly related to the
development of software, but provide a human face to the people who will use the software. In
order to be relevant, personas shouldn't be reused between products.
Fully-developed personas can be used by the development team to help answer questions
about how users will use their product without having to constantly query the user, particularly
about issues to which they are not likely to be forthcoming about due to the corporate culture.
To have (positively) passionate users is perhaps the best result you could hope to have
because they will experiment with your software and find new ways to use it and then promote
your product to others through their online and offline social networks. Personas will help you
Contemporary issues in gathering software usability requirements 11
produce passionate users because they address the true goals of the users, which cannot be
derived from the software’s functional requirements. A particularly poignant expression of this
concept is given by Sierra (2005).
Aesthetics and visceral reactions
Users will be much more tolerant of problems in a product that is aesthetically pleasing
and that appeals to them viscerally. While it is never a good idea to be all style and no substance,
some well-placed style can make up for the shortcomings in a version 1.0 product, for example.
Style may be the difference between your product being branded for replacement and your team
getting a chance to make it better in version 2.0. In addition, good aesthetics make a user want to
try a product and provoke positive emotions each time they look at the product. A user in a better
emotional state will be more willing to try, retry, use, and have patience with a product.
An excellent example of the value of aesthetics and visceral reactions lies in Apple’s
iPod. The iPod is a product that is essentially an overpriced plastic case containing electronics,
software, a monochrome LCD display, and a miniature hard drive. There were many other
products that matched this description on the market before the iPod was introduced; indeed,
some were arguably better products. The difference is that Apple created a stylish product that
had a story and a sense of community around it. The iPod’s predecessors were not stylish, nor
did they have a story. The products usually came in an uninspiring box, had little marketing
support, and were targeted at people who already knew what it was and had a preconceived
notion of what they would do with it. The response from its users was so great that the users
themselves were creating “remixes” of Apple’s TV commercials and sharing them with friends.
An industry has been created for the development of iPod accessories. People wanted to be seen
Contemporary issues in gathering software usability requirements 12
wearing the signature Apple iPod earbud headphones, so much so that even in the face of a 20%
increase in underground subway crime in New York City, partially attributed to iPod thefts
(Donohue, & Sacks, 2005), and highly publicized instances of “iPod mugging” (Evans, J, 2004),
many users will not give up their iPod status symbol. This frenzy continues, despite quality
problems2, its rather limited software environment3, its lack of features compared to similar
products from other vendors (Gavin, 2005), and the DRM issues described later in this paper.
Since the iPod is a system, of which software is a significant component – it provides the
interface with the device both at the device and the personal computer level – it therefore follows
that the same positive reaction must also be possible for products that are chiefly software.
2 iPods have had significant problems related to short battery life. Apologist user communities have emerged to deal with the problem: http://www.ipodbatteryfaq.com/
3 You must download special software for Windows, which only works with the iPod despite the existence of OS functionality that provides the same function as the software. iPod also does not support the Windows media file format, which nearly all other competing players do support (Thurrott, 2004).
Contemporary issues in gathering software usability requirements 13
Figure 1 A comparison between the emotional appeal of Delicious Library and Movie Collector Pro
The iPod was not an accident. Apple has a long-standing history of creating emotionally
appealing software for their Macintosh personal computers; an example can be seen in the visual
comparison of catalog software between Delicious Library (a Macintosh product), and Movie
Collector Pro (a Windows product) (figure 1). Even at this abstract view of the two programs, it
is far easier to imagine how the Macintosh-based product functions. The products are
functionally equivalent, but the Delicious Library product is much more visually appealing and
encourages emotional response that the Windows product does not. Another example exists in a
Contemporary issues in gathering software usability requirements 14
comparison between Microsoft PowerPoint and Apple Keynote (figure 2); both of these
applications are Macintosh applications, but PowerPoint has origins in the Windows
environment. The clutter, ambiguity, and confusion in the PowerPoint can be overwhelming. A
third example lies in the comparison of any Windows videoconferencing software with Apple’s
iChat AV videoconferencing software4. This is “only software”, yet when new Apple software
products are revealed to its user community at trade shows like MacWorld (Apple Corporation,
2005), the audience breaks into cheer in a scene reminiscent of a rock concert. As a result of
Apple’s understanding of the importance of aesthetics to its user community, the Macintosh
community is very defensive of Apple in public, and voluntarily evangelize the benefits of their
preferred platform to anyone who will listen, and even those who won’t. The abundance of cheap
offerings based on Microsoft’s Windows platform, combined with the higher price tag of
Apple’s offerings means that, when someone chooses an Apple, they choose it with intent and
purpose.
4 http://www.apple.com/ichat/
Contemporary issues in gathering software usability requirements 15
Figure 2 A comparison between Apple Keynote and Microsoft Powerpoint
The important role of aesthetics and visceral appeal is a relatively new revelation to the
scientific community. Donald Norman, a well-respected authority on the design of useable and
functional products and author of the seminal book The Design of Everyday Things (Norman,
1990), was an evangelist of usable function over all else, based on his studies of cognition.
Recent developments in the understanding of the brain have revealed how closely emotion and
cognition are intertwined, and his new approach, which recognizes the importance of visceral
and aesthetic appeal, is well-documented in Emotional Design (Norman, 2004). Norman also
comments on the possibility of reflective design, which allows a product to complement an
Contemporary issues in gathering software usability requirements 16
individual’s self-image and be a part of their memories and stories. In essence, the value of
emotional appeal is illustrated when we perceive a newly washed and polished car a driving
better, and cheap wine as tasting better when tasted from expensive glasses. The challenge is to
capture this same benefit in the requirements of software.
Partially negating preferences
The problem of expressing what is desired in software for the purposes of a requirements
phase might be a reason that some of the most popular tools for content creation are freeform
tools. Such tools allow the user to work freely without interference from the software, and
without conformance to a structure that may not make sense to them. Two classic examples of
such tools are the word processor and the spreadsheet. Regardless of one’s expert knowledge of
the specific tool, almost anyone can get to work and be productive with either of these tools with
little or no training.
Recently, some impressive tools have arrived on the market which extend the freeform
nature of the word processor and the spreadsheet in order to tackle more complex tasks. Two
examples are Mindjet’s MindManager and Microsoft’s OneNote. After learning a few simple
concepts, MindManager can be used to build ideas, document meetings, or organize research.
OneNote allows you to take ad-hoc notes in a digital form, whether by keyboard, by pen, or by
voice, on the spur of the moment and organize them into a structure at a later time, at your
convenience. The freeform, liberal nature of both tools lets you document free-flowing
information without having to make the information conform to a pre-determined structure. Such
software can be used as intended, or you can use it for whatever you want to use it for. There are
Contemporary issues in gathering software usability requirements 17
few preconceptions trying to shape the way in which you use the tool, and they are backed up by
healthy “undo” features so that there is no harm in experimentation.
The popularity of such tools, and the passion displayed by their users, is a cue toward
building the requirements of more usable products. Sadly, as illustrated by the instability of
preferences, if a user is not accustomed to this approach to building software, they will not
request it. Users will only ask for what they know and will not realize the benefits of a great
product until after they have used it. For this reason, leadership on the part of the development
team is required in order to bring these benefits to the table.
Join the conversation with weblogs (“blogs”)
At first glance, the act of keeping and maintaining a personal weblog (“blogging”) seems
like it is simply a different take on the personal website. In actual fact, there is an underlying
technology which makes it much more powerful, called a “TrackBack”. When a blogger uses a
blogging tool to update her blog with a posting, other bloggers have the ability to create a
TrackBack link to that posting. Likewise, if she is referring to another blog in her posting, she
will create a TrackBack link to the posting being referenced. This in itself seems rather basic, but
combined with the way search engines like Google rank their search results5 (Google, 2004), it
means that a product that is heavily discussed in the “blogosphere” will have a high ranking in
search results returned for that product. It is for this reason that software companies must pay
attention to what their customers are saying about their products. If customers do not like your
product and are actively discussing their distaste, it is quite possible that their comments will
appear above or alongside your own company’s website in Google’s search results. Similarly, if
5 In addition to the keyword being searched for, Google’s ranking algorithm gives higher priority to websites that are more frequently linked to by other sites.
Contemporary issues in gathering software usability requirements 18
your product is loved then it will receive similar attention, although we are all too familiar with
the fact that customers are generally more willing to complain about something they hate than
give praise about something they like.
High-tech products such as software, in particular, will receive more attention in the
blogosphere because of the ability of technical people to utilize leading-edge techniques such as
blogging to participate in virtual communities. And even to say this is perhaps a bit unfair,
because blogging is not solely or even mostly the domain of high-tech enthusiasts now that tools
are available that make it as easy as sending a web-based e-mail. Regardless, users of Google,
irrespective of their technological sophistication, will still see what is being said about products
when they use their search engine! In addition, most blogs allow comments to be made in
response to blog postings. Such comments provide valuable, real customer feedback that you
would not receive if you were not part of the larger conversation.
Some companies, such as Microsoft, have recognized the importance of this new
conversation technique and encourage their own employees to create blogs and participate in the
user community6,7 and discover what their users like and dislike a bout their products. Other
companies, such as ClearContext8, receive a lot of publicity because of the positive experiences
of bloggers. A Google search for their company name reveals these experiences, and what’s
better is that they are real people having a conversation, not advertising-biased newspapers, or
company spokespeople!
6 http://blogs.msdn.com/7 http://channel9.msdn.com/8 http://www.clearcontext.com/
Contemporary issues in gathering software usability requirements 19
Why not open source?
Open source has been included as a solution because it would be one of the obvious
apparent solutions to the problem of conflicting requirements stemming from different
commercial and user interests. Essentially, because open source is written by the user community
for the user community, it would seem to follow that commercial interests would not have nearly
as much weight as in commercially-developed software. However, open source has a number of
problems which may make it even worse than the commercial offerings at serving the usability
requirements of typical user populations. Firstly, open source software is written collaboratively
by experts for experts. Programmers participate in open source projects because they want to
improve the software for themselves and for like-minded people; such people are very far
removed from the mindset of the typical end-user and do not have the same requirements. This
disparity between open source developers and typical end-users is especially evident in the open
source efforts to offer a compelling alternative to Microsoft’s Windows desktop operating
system. Such efforts generally demonstrate a visual appearance that resembles Windows, yet fail
to create a powerful overall user experience that complements the workflow and sensibilities of a
typical end-user. The open source community has been much more successful in the server
operating system market because users of such products are more analogous to the open source
developers themselves.
Open source has the additional problem of having to function within a world based on
commercial interest. An excellent example exists in the Digital Versatile Disc (DVD) market.
DVD discs are protected with a private-key encryption system called Content Scrambling
System (CSS) (Kesden, 2000). Since the encryption is based on a secret key, the secret key must
Contemporary issues in gathering software usability requirements 20
be protected within the player’s source code, which is impossible when the source code has full
transparency. The result of this was that an open source DVD player was not possible until the
encryption scheme was broken. But, distribution of such software is then a violation of the
Digital Millenium Copyright Act (DMCA). An open source product can therefore not support
playback of encrypted DVDs, regardless of the product’s usability requirements, and would
likely have similar problems adding a DRM scheme. A hybrid solution that consisted of closed
and open source components would likely resolve this issue, but it would require a collaboration
between parties who would be reluctant to work with each and meet each other’s demands.
Conclusion
Ongoing research outside of the software engineering field can change the way in which
software engineers understand the motivations of their users for the better. As a result, should
they choose to pay attention to such research, software engineers can better connect with the
motivations that their users have, and offer solutions to problems that they have that do not fit
neatly into a category of functionality. This is one of the fundamental differences between
software engineers and computer scientists – engineers are accountable to the public and should
produce products that make the output of computer scientists fit for consumption by the public.
By reducing the mental friction involved in dealing with software at all stages of the product
development, including product selection, the user will be more satisfied with their experience
and the chances of them returning for more of the same great experience are increased.
ReferencesAnderson, C. (2004, October). The long tail. Retrieved March 27, 2005 from http://wired-
vig.wired.com/wired/archive/12.10/tail.htmlRelevance: this article was used to describe the “long tail” to support the use of the term in the paper.
Contemporary issues in gathering software usability requirements 21
Borland, J. (2005, March 22). 'DVD Jon' reopens iTunes backdoor. Retrieved March 23, 2005 from http://news.com.com/DVD+Jon+reopens+iTunes+back+door/2100-1027_3-5630703.htmlRelevance: this article was used to support the idea that advanced users will reverse engineer your software to do things that you try to prevent them from doing. Not only will they be dissatisfied with your software at the outset, but they will make their findings available to the wider, less advanced users.
Breen, C. (2005, January 31). Two-way street: moving music off the iPod. Retrieved March 23, 2005 from http://playlistmag.com/help/2005/01/2waystreet/Relevance: this article was used to support the idea that advanced users will reverse engineer your software to do things that you try to prevent them from doing. Not only will they be dissatisfied with your software at the outset, but they will make their findings available to the wider, less advanced users.
Buckley-Golder, M. (2003, November 20). Is usability testing enough?. (MSWE601 research paper). Retrieved April 2, 2005 from http://www.buckley-golder.com/mbg_ut.docRelevance: I wrote this paper for MSWE601 and have made it available online. This paper describes some of the technical approaches to usability, so I included the reference in this paper to support my evaluation of social and psychological issues (rather than reinvent the wheel).
Cooper, A. (1999, June). 14 principles of polite apps. Visual Basic Programmers Journal, 1999(6), 62-66.Relevance: this article is written by the “father” of interaction design and gives a good overview of the idea of “mental friction”, which is briefly described in my paper. This article provides more understanding of the term..
Cooper, A. (2003). About face 2.0: the essentials of interaction design. Wiley: Indianapolis, IN.Relevance: this book is one of the primary textbooks providing a foundation for the field of interaction design. Cooper displays a genuine interest in the problems that users have in dealing with today’s software and provides taxonomies for some previously unstructured problems (such as software’s “undo” functionality, which I refer to in my paper).
Evans, J. (2004, March 30). ‘iPod mugging’ latest media frenzy. Retrieved March 20, 2005 from http://www.macworld.co.uk/news/index.cfm?newsid=8309Relevance: this article was used to support the assertion that iPods are so desirable that they are responsible for creating new categories of crime.
Gavin, P. (2005, January 21). iPod Mini vs. Zen Micro. Retrieved March 31, 2005 from http://www.mp3.com/iPod+Mini+vs.+Zen+Micro/stories/1149.htmlRelevance: this article shows that there are products having more functionality than the iPod that are reviewed more favorably, yet iPod users continue to upgrade to the next iPod. There is something more to a product than functionality (aesthetics, usability, community) that will make users passionate about a product.
Gladwell, M. (Speaker). (2004). Human nature. Retrieved March 2, 2005 from http://www.itconversations.com/shows/detail230.htmlRelevance: this presentation (audio format) provides a case for the instability of preferences, and was the gateway to my finding the Wilson & Schooler paper.
Goodwin, K. (2001). Perfecting your personas. Retrieved March 30, 2005 from http://www.cooper.com/newsletters/2001_07/perfecting_your_personas.htm
Contemporary issues in gathering software usability requirements 22
Relevance: this article was written by an associate of Alan Cooper’s (see above) on the subject of personas. Personas are also covered in Cooper (2003), but this article is more concise.
Google. (2004). Google technology. Retrieved April 10, 2005 from http://www.google.com/technology/Relevance: this page explains Google’s PageRank technology, which gives preference to search results that are heavily linked by other pages. This is relevant to the “blogging” discussion because it shows that bloggers (whose discussions are heavily cross-linked) can affect what information appears alongside company links in search results, possibly harming the reputation of a bad product.
Jobs, S. (2005). Macworld San Francisco 2005 expo keynote [video clip]. Retrieved April 2, 2005 from http://www.apple.com/quicktime/qtv/mwsf05/Relevance: this video clip was referenced because it illustrates the enthusiasm and passion that users of Apple’s personal computers have for their chosen product.
Kesden, G. (2000, December 6). Lecture 33 (notes). Carnegie Mellon University course 15-412. Retrieved December 6, 2000 from http://www-2.cs.cmu.edu/~dst/DeCSS/Kesden/Relevance: this presentation was used to support the idea that open source software products have challenges if they are going to provide functionality (and hence meet requirements) which requires source to be closed or protected.
Norman, D. A. (1990). The design of everyday things (1st Doubleday/Currency ed.). New York: Doubleday.Relevance: this book is an important book to the design community, but it also shows Norman’s prior emphasis on functionality over appearance in order to provide context for his new emphasis on emotional design in Norman (2004). Norman is well-respected and his views therefore carry substantial weight in the product design community.
Norman, D. A. (2004). Emotional design: Why we love (or hate) everyday things. New York: Basic Books.Relevance: this book illustrates Norman’s conversion to the belief that aesthetics, visceral appeal, and reflective design can add to the value of a product. His new views are based on contemporary research about cognition.
Sierra, K. (2005, February 16). What users really want. Retrieved April 9, 2005 from http://headrush.typepad.com/creating_passionate_users/2005/02/what_users_irea.htmlRelevance: this article provides a brief, if crass, example of a goal vs. a functional requirement.
Thurrott, P. (2004, September 9). Windows Media Player 10 review. Retrieved March 24, 2005 from http://www.winsupersite.com/reviews/wmp10.aspRelevance: this article was used to show that there are much more open media platforms available beyond Apple’s iPod ecosystem, yet iPod continues to be successful. It is more evidence that functionality and openness is not the only thing that matters. In fact, nearly all iPod competitors plug-in to the Windows Media platform, yet Apple has more than 80% market share.
Sacks, E., & Donohue, P. (2005, March 29). Music to their fears. Retrieved April 2, 2005 from http://www.nydailynews.com/news/story/294378p-252061c.html
Schwartz, B. (2004). The paradox of choice: why more is less. HarperCollins: New York, NY.Relevance: this book is the basis of my comments about the paradox of choice, which essentially is that as we have more choice, we have more alternatives to choose from, which means more opportunity cost for
Contemporary issues in gathering software usability requirements 23
picking one of the alternatives. As a result, we are less satisfied with the choice that we do make. He suggests voluntary constriction of our own choices to improve the situation, and a settlement for “good enough” over “the best” for trivial matters.
Wilson, T. D., & Schooler, J. W. (1991). Thinking too much: introspection can reduce the quality of preferences and decisions. Journal of Personality and Social Psychology, 60(2), 181-192.Relevance: this research article explains that asking someone why they like something can change their expression of preference; someone who knows that they will have to explain their preference will prefer something that they know how to explain