USABILITY HEURISTICS FOR XML-BASED WEB

DEVELOPMENT

Marta Fernández de Arriba and José A. López Brugos

Computer Science Department. University of Oviedo. Spain.

Keywords: Usability heuristic, usability inspection methods, Web site evaluation, XML

Abstract: Heuristic evaluation is a usability engineering method for finding the usability problems in a user interface

design. This paper discusses a set of rules to evaluate the level of usability of an XML-based Web site.

Taking advantage that XML separates the contents of a document from its presentation, heuristics grouped

by content and presentation are defined in an analogous way, allowing a suitable evaluation for that type of

applications.

1 INTRODUCTION

Heuristic evaluation is a rapid, low-cost and easy

method used to find usability problems in a user

interface design (Nielsen and Molich, 1990).

Heuristic evaluation is made by a group of

expert evaluators, who apply the set of criteria

established by the heuristic rules. A number between

three and five evaluators is enough, although it may

be increased in the case that usability was a key

factor for the success of the Web site. Nevertheless,

in order to know the optimal number of evaluators, a

cost-benefit analysis must be made in each case

(Nielsen, 1994).

Nielsen describes ten usability heuristics

(Nielsen and Mack, 1994) that can be applied to the

user’s interface of any type of computing system.

Although these heuristics might be potentially

enough to fulfil the evaluation of a Web site, its

application is difficult due to the generic character of

both criteria.

On the other hand, the standards of usability

published by public organizations (ISO 9241, Parts

10-17; ISO 13407, 1999; ANSI/HFES, Parts 1-4)

specify a great number of rules to be followed,

showing the same generic character as in the case of

the heuristics

developed by Nielsen. This amount of

criteria makes its application difficult both from the

Web site developer and its usability evaluator’s

point of view.

This paper presents a set of heuristics,

specifically positioned to the evaluation of the

degree of usability of XML based Web sites.

2 HEURISTIC EVALUATION

PROCEDURE

Heuristic evaluation is performed by having each

individual evaluator inspect the interface alone. Only

after all evaluations have been completed are the

evaluators allowed to communicate and have their

findings aggregated, with the aim of guaranteeing

the independence and impartiality during the process

of evaluation.

During the evaluation session, the evaluator

examines the elements that form the Web page,

comparing them to the list of usability principles

(the heuristics). Each individual session should

never last more than two hours (Nielsen, 1994).

Once finished this session, the evaluator makes a

report specifying the list of usability problems found

in the interface, showing the principles that have

been violated and listing in detail both where and

how have been produced .

The next step consists of the classification of

errors found, according to their importance (measure

that comes from the combination of three factors:

(1) the frequency with which the problem occurs, (2)

the impact of the problem if it occurs, and (3) the

persistence of the problem), so that the most

important usability problems can be found and, thus,

we may decide whether the product will be

published or not before its solution (Nielsen, 1995).

As each evaluator will find a small number of

usability problems, it is advisable to link all the

mistakes found by each of them, so that they make a

121

Fernández de Arriba M. and A. López Brugos J. (2004).

USABILITY HEURISTICS FOR XML-BASED WEB DEVELOPMENT.

In Proceedings of the Sixth International Conference on Enterprise Information Systems, pages 121-125

DOI: 10.5220/0002630901210125

 SciTePress

later evaluation of each of the problems (using a

rising scale from one to ten to classify their

importance). The classification of usability problems

found will be useful as the basis for the decisions

taking on which parts must be redesigned and their

order of application.

3 HEURISTIC MODEL IN LAYERS

Nowadays, most of the content published on the

Web uses the Hypertext Markup Language (HTML)

where content, presentation and structure are

inseparably linked.

Following the Web accessibility guidelines

(W3C, 1999a), the presentation and structure of a

document should be separated. The content (the

information on the site) should be separated from

presentation (the look and feel of the site), so that

future redesigns can easily be developed.

Furthermore, the same content and structure can be

present in different ways to different users, devices

(PC, mobile, ...) and platforms (e.g. through use of

style sheets).

Web sites developed by using XML technologies

are characterized by the structured description of the

content of its documents, completely independent of

the definition of that information presentation format

(W3C, 2000).

The presentation of XML documents can be

specified using technologies such as Cascading Style

Sheets, level 1 (W3C, 1999b), Cascading Style

Sheets, level 2 (W3C, 1998) or eXtensible

Stylesheet Language (W3C, 2001).

Starting from this structure, this paper provides a

model of heuristic evaluation structured in four

layers (Figure 1), with the aim of making the

usability study of these Web sites easier (as shown

in the reports of usability problems found).

The two basic layers of the heuristic evaluation

model (content heuristics and content organization

heuristics) are related with the level of information

content of the Web site (XML document). Thus,

content heuristics evaluate the information of each

individual page of the Web site, whereas the content

organization heuristics evaluate the structure and

homogeneity of all the information in all the pages

of the Web site.

The heuristics´ consistency in presentation layer

is associated with the level of translation of the

XML document to its presentation in the browser,

using CSS or XSL technologies, which is applied to

the study of style and navigation correction of such

documents, as well as the coherence in the

presentation of the set of all the pages that form the

evaluated Web site.

The last presentation layer (heuristics of

presentation in the browser) specifies the evaluation

of usability of the Web site documents view in the

user´s browser, as well as the user´s interaction with

the Web application.

4 HEURISTICS

Heuristics of usability are shown grouped in four

levels corresponding to the four layers of the

described heuristic model.

ICEIS 2004 - HUMAN-COMPUTER INTERACTION

122

H1: Content.

H1.1 All content is clear and intelligible to the target

audience, do not present repetitious or

irrelevant information and do not have

misleading links.

H1.2 The most important information is at the top of

the document.

H1.3 The information is grouped in sets that

facilitate the user its location, interpretation

and understanding in the easiest way.

H1.3.1 Groups are visually distinguished in a

clear way (in position, spacing and

codification).

H1.3.2 Groups of objects logically related are

internally marked within the document

so that they are identified by other ways

different from the visual (for users who

use screen readers or voice browsers).

H1.4 The name of each page is identified without

repetitions (to facilitate the organization in the

bookmarks).

H1.5 The elements (fields in forms, icons and

graphics) are distinguished and easily

discriminated and they can also be easily

understood. At all events, those whose

meaning is not obvious are clearly labeled in

order to facilitate their understanding.

H1.6 A clear and complete description of any

specific functionality of the document is

included (facilitating the use to the user).

H1.7 Relevant and equivalent alternatives to

auditory and visual contents are provided.

H2: Organization of the content.

H2.1 The content of the Web site follows a

hierarchic organization with more general

information appearing before more specific

detail (facilitating the user to reach the wished

detail level).

H2.2 Documents contain exactly one topic avoiding

the user to access multiple documents for a

same subject.

H2.3 The repeated elements maintain their

consistency throughout the Web site, always

using the same identifier (name or symbol).

H2.4 All the necessary aids and guides, giving the

user additional information to the normal

dialogue with the interface in order to facilitate

to the utmost its interaction with the system,

are provided.

H3: Consistency in the presentation.

H3.1 A presentation style common to all Web site

pages are kept (fonts -using two different, one

for the title and another for detail- and

backgrounds that don't affect the legibility of

the document) and homogeneous location of

all elements (buttons, sections, links and

navigation menu options).

H3.2 The typefaces´ size has an enough dimension

and their foreground colour keep an adequate

luminance contrast difference as for the

background colour, in such a way that the

document can be easily read. Moreover, there

is not any discriminatory use of the colours

that causes an overprint effect of the

presentation.

H3.3 Navigation menus with links to the different

sections of the Web site are designed in such a

way that they facilitate the user the control of

the navigation process.

H3.3.1 All the pages make an exclusive and

common use of top and left zones for

the navigation sections.

H3.3.2 The navigation menu provides the links

that lead, in an organized manner, to all

the main sections of the Web site.

H3.3.3 Navigation menus are grouped in a

level structure so that no menu has a

great amount of options (from eight

onwards) for its presentation in a

unique toolbar.

H3.3.4 The menu options are visually ordered

in categories, obvious for the user, or if

not possible, they appear in a coherent

way (alphabetically or by frequency of

use).

H3.3.5 The access to the menu options is made

both by the mouse and by shortcuts

(clearly identified for the user).

H3.3.6 The menu options are identified and

distinguished by brief and not

ambiguous names, familiar to the user´s

language

H3.4 The user knows where is in the Web site

structure, where he can go next and how return to

home.

H4: Presentation in the browser.

H4.1 In the final presentation standard languages

(HTML, XHTML), published by the World

Wide Web Consortium, are used (W3C,

1999c; W3C, 2002).

USABILITY HEURISTICS FOR XML-BASED WEB DEVELOPMENT

123

H4.2 The Web site is compatible with any browser,

in different platforms, screen resolutions and

bandwidth.

H4.3 All the system answers to user's petitions

include the own petition reminding him the same

one.

H4.4 The configuration of presentation of each user

is respected, avoiding impose him suitable

preferences for the Web site.

H4.5 Links are clearly marked and their target

should not be a new window (allowing the forward-

backward navigation).

H4.6 The pages are not upgraded automatically

in a periodic way.

H4.7 Technologies that the user cannot control are

not used (animated graphics, blinking text or

default background sound).

H4.8 All the multimedia elements are totally

controlled by the user (play, stop, ...).

H4.9 The pages are efficient since they avoid

unnecessary communications with the server

(checking all the form fields in the client, using

client-side image maps).

H4.10 The download is efficient avoiding

excessive page sizes, allowing optional download

for any heavy file.

H4.11 All the data entries are designed to

facilitate its use by the end user.

H4.11.1 Forms are designed in a way that the

user easily determines its general

structure and can insert data in a natural

way.

H4.11.2 All the fields are labeled in a clear and

not ambiguous manner, indicating the

purpose and the type of entries that are

linked.

H4.11.3 Information about the acceptable

values and the obligatory or optional

character of the entries is given.

H4.11.4 Errors management clearly identifies

the wrong fields and give detail of the

cause of the mistake and permits its

correction with no need to introduce all

the data again.

H4.11.5 The System provides the user a

confirmation of the sending of the data

and the updating of the Data Base

(when necessary).

5 CONCLUSIONS

The degree of usability of a Web site developed in

XML can be validated through an heuristic

evaluation organized in a layer model (associated

with the levels of internal structure of the XML

technology) by a small group of experts who check

the degree of fulfilment of the list of the usability

heuristics.

The heuristic evaluation is far better than the rest

of techniques and methods of usability evaluation in

cost, as it only needs a small number of evaluators.

Another additional advantage is that it does not need

an external interpretation of the user´s actions, as the

evaluators make the tests and elaborate the detailed

reports of the problems found. Moreover, in the

heuristic evaluation, there is the possibility of

interacting and helping the evaluators in order to

optimize the process of errors detection, in contrast

to the user´s tests, in which these must interact with

the Web site with no additional information, so that

it may be possible that any serious problem at the

beginning of the tests impedes them to continue with

the evaluation, making impossible to detect the rest

of the mistakes.

For all these advantages, it is advisory to make a

heuristic evaluation of any developed Web site, so

that most of the problems are found and solved with

a reduced cost, previous to any type of user´s test

planned on this Web site.

REFERENCES

ANSI/HFES 200. Human Factors Engineering of

Software User Interface. Parts 1-4. Human Factors

and Ergonomics Society.

ISO 9241. Ergonomic requirements for office work with

visual display terminals (VDTs). Parts 10 - 17.

International Organization for Standardization,

Geneva, Switzerland.

ISO 13407, 1999. Human-centred design processes for

interactive systems. International Organization for

Standardization, Geneva, Switzerland.

Nielsen, J., 1994. How to Conduct a Heuristic Evaluation.

( Online: http://www.useit.com/papers/heuristic/

heuristic_evaluation.html)

Nielsen, J., 1995. Severity Ratings for Usability Problems.

(Online: http://www.useit.com/papers/heuristic/

severityrating.html)

Nielsen, J. and Mack, R.L., 1994. Usability Inspection

Methods. John Wiley & Sons, Inc.

Nielsen, J. and Molich, R., 1990. Heuristic evaluation of

user interfaces, Proceedings ACM CHI'90 Conference,

pp. 249-256. ACM Press .

ICEIS 2004 - HUMAN-COMPUTER INTERACTION

124

W3C, 1998. Cascading Style Sheets, level 2. CSS2

Specification, World Wide Web Consortium,

Recommendation REC-CSS2-19980512; Bos, B.,

Wium Lie, H., Lilley, C. and Jacobs, I. (Online:

http://www.w3.org/TR/REC-CSS2/)

W3C, 1999a. Web Content Accessibility Guidelines 1.0,

World Wide Web Consortium, Recommendation

REC-19990505; Chisholm, W., Vanderheiden, G. and

Jacobs, I. (Online: http://www.w3.org/TR/WAI-

WEBCONTENT/)

W3C, 1999b. Cascading Style Sheets, level 1, World Wide

Web Consortium, Recommendation revised 11 Jan

1999, REC-CSS1-19990111; Wium Lie, H. and Bos.

B. (Online: http://www.w3.org/TR/REC-CSS1)

W3C, 1999c. HTML 4.01 Specification, World Wide Web

Consortium, Recommendation REC-html401-

19991224; Raggett, D., Le Hors, A. and Jacobs, I.

(Online: http://www.w3.org/TR/html401/)

W3C, 2000. Extensible Markup Language (XML) 1.0

(Second Edition), World Wide Web Consortium,

Recommendation REC-xml-20001006; Bray, T.,

Paoli, J., Sperberg-McQueen, C.M. and Maler, E.

(Online: http://www.w3.org/TR/REC-xml)

W3C, 2001. Extensible Stylesheet Language (XSL)

Version 1.0, World Wide Web Consortium,

Recommendation REC-xsl-20011015; Adler, S. et al.

(Online: http://www.w3.org/TR/xsl)

W3C, 2002. XHTML 1.0: The Extensible HyperText

Markup Language (Second Edition ). A Reformulation

of HTML 4 in XML 1.0, World Wide Web

Consortium, Recommendation revised 1 August 2002,

REC-xhtml1-20020801; Pemberton, S. et al. (Online:

http://www.w3.org/TR/xhtml1)

USABILITY HEURISTICS FOR XML-BASED WEB DEVELOPMENT

125