SITEMAPS
FROM A MODEL DRIVEN PERSPECTIVE
A First Step for Bridging the Gap between Information Architecture
and Navigation Design
Antonia M. Reina-Quintero and Jes
´
us Torres-Valderrama
Department of Languages and Computer Systems, University of Seville, Avda. Reina Mercedes, s/n, Seville, Spain
Keywords:
Metamodelling, Model transformations, Information architecture, Navigation design, Sitemaps, Web design.
Abstract:
Researchers claim that there is a disconnection between information architecture and navigation design. One
way of approaching these two fields is to share deliverables. However, it is difficult to change the minds
of audiences to make them use deliverables they are not used to. Thus, we propose let audiences use those
deliverables they are more comfortable with, and then transform one deliverable into another, as far as possible.
To get this aim, firstly, we need to have a deep knowledge of deliverables, and secondly, a set of mappings have
to be defined in order to translate the information the source deliverable is covering into the target deliverable.
Our approach uses metamodelling as the technique to define the pieces that compose deliverables and their
relationships, and model transformations for mapping deliverables. In this context, the paper focuses on one
of the most widely used information architecture deliverables, sitemaps, and its main contributions are: (1) a
sitemap metamodel, which define the minimum set of elements that can be used for specifying sitemaps; and,
(2) a set of model to model transformations to obtain a XHTML skeleton of structural and utility navigation.
1 INTRODUCTION
The two disciplines that face up to some of the
greatest challenges when developing internet sites are
Information Architecture (IA) and Navigation De-
sign (ND). They try to answer non-technical questions
such as: What kind of information should be included
on a website? How should the site be organized? Or
how will users be able to find their way around? IA is
more related to organizing, structuring and labeling
information, while ND involves technical develop-
ment and visual design. However, researchers have
argued that there is a disconnection between these two
disciplines (Azam et al., 2005), and as a result, bad
designs are obtained because information technology
is too separated from the design process or too far in-
tegrated in it.
A good way to bring these two disciplines closer
is to share deliverables. But there are two main pro-
blems when different audiences work with delivera-
bles: on the one hand, consumers of documentation
do not read deliverables, but they look for nuggets
of information they can use to complete their own
work (Curtis, 2008); and, on the other hand, it is di-
fficult to change the mind of audiences to make them
utilize deliverables they are not used to. In this con-
text, we propose letting the members of the deve-
lopment team use those deliverables they are more
comfortable with, and then, transform one deliverable
into another, as far as possible. Thus, for example,
one common scenario can be the transformation of
sitemaps, one of the most widely used IA delivera-
bles. Sitemaps specify the structure of a website, but
also hold information related to structural navigation.
With our approach the navigational information ob-
tained from sitemaps will be mapped to other deli-
verables that deal with navigation, such as state ma-
chines or flow charts. This strategy makes easier the
navigation designer’s job, because he or she will not
read sitemaps, but deliverables he or she is comforta-
ble with.
Our approach is composed of two main groups of
activities: firstly, we need to have a deep knowledge
of deliverables, and secondly, a set of mappings have
to be defined in order to translate the source deliver-
able into the target deliverable. To address the first
group of activities, metamodelling is proposed, be-
cause metamodels capture the essential features and
properties of the deliverables that are being mod-
eled (Clark et al., 2008). The translation is obtained
by means of a set of model to model transformations.
Thus, this paper is focused on sitemaps, which
111
M. Reina-Quintero A. and Torres-Valderrama J.
SITEMAPS FROM A MODEL DRIVEN PERSPECTIVE - A First Step for Bridging the Gap between Information Architecture and Navigation Design.
DOI: 10.5220/0002804701110117
In Proceedings of the 6th International Conference on Web Information Systems and Technology (WEBIST 2010), page
ISBN: 978-989-674-025-2
Copyright
c
2010 by SCITEPRESS Science and Technology Publications, Lda. All r ights reserved
are not only one of the most widely used delivera-
bles in IA, but also one of the deliverables understood
by a great number of different stakeholders. Firstly,
a sitemap metamodel has been specified by means
of the Ecore Modeling Framework (EMF) (Stein-
berg et al., 2008). And, secondly, some transforma-
tions specified by means of QVT Operational Map-
ping (OMG, 2005) are defined. The result of the
transformations is a skeleton of a set of XHTML
pages. This skeleton is focused on structural and util-
ity navigation and it is used for validation purposes.
The rest of the paper is structured as follows: Sec-
tion 2 introduces sitemaps as an information architec-
ture deliverable. Sections 3 and 4 describe the source
and the target metamodel, respectively. After that, the
model to model transformations are specified in Sec-
tion 5. Then, a general overview of related work is in-
troduced in Section 6. Finally, the paper is concluded
and some further lines of research are pointed out.
2 SITEMAPS
Sitemaps are one of the most widely used IA delive-
rables and define structural navigation. One of their
main values is that they allow designers to anticipate
information organization errors. As a consequence,
these errors can be corrected prior to time and money
investment. However, there is no a standard best prac-
tice for creating sitemaps. The initial sketches are
usually drawn by hand, and then, a neater version
could be drawn with some kind of visual editor. Some
information architects and designers use Visio to lay
out pages hierarchies and create connections between
them, but they could be also created in Word, Power
Point or any other tool the designer is comfortable
with.
The most basic sitemap (Brown, 2006) is similar
to an org-chart where boxes represent pages or diffe-
rent areas of the site (Figure 1 shows an example of
a simple sitemap). Boxes are connected by lines that
show the semantic relationships between the areas of
the site and may also represent navigation, but they
do not have to. For example, in some sitemaps lines
represent relationships between content without ex-
plicitly suggesting navigation. The place and connec-
tions imply that there is a hierarchy between pages, in
such a way that the site’s home page is usually at the
top of the chart. Although there is no standard best
practice, there are a set of common elements that usu-
ally are specified in many sitemaps (Kalbach, 2007):
nodes, representing pages in the site; connectors, that
show nodes relationships; numbering scheme, used
for identifying pages in an unique way; labels, which
Home
0
News
1
Contact
2
Teaching
3
Consulting
Hours
3.1
Courses
3.2
Ms. Thesis
3.3
Proposals
3.3.1
Finished
3.3.2
Research
4
Publications
5
Leisure
6
My
Papers
5.1
Collaborations
5.2
All
5.3
Citations
5.4
Main Navigation
Local Navigation
Ms. Thesis
3.3
Consulting
Hours
3.1
Home
0
Utility Navigation
Figure 1: A high-level sitemap for a researcher web page.
represent nodes titles; page attributes, showing other
page characteristics; notes and annotations; scope
and, finally, title and key, that are sitemap properties.
Sitemaps can be represented in a coarse-grained
or a fine-grained way. On the one hand, high-level
sitemaps (also called blueprints) show how the main
sections of a web site fit together. They do not show
all pages. On the other hand, detailed sitemaps docu-
ment the site finer-grained. As it is impractical to dia-
gram hundreds of pages, detailed sitemaps are written
in separated documents, one for each sitemap section.
3 SITEMAP METAMODEL
Sitemaps specify site structure and navigation, two re-
lated concerns. A detailed sitemap shows all pages
of a site, but the navigation is a limited view into
that structure. From any given page in the site,
navigation is a constrained window of all available
pages (Kalbach, 2007). The metamodel specified in
this section models coarse-grained sitemaps. In ad-
dition to this constraint, three additional assumptions
have been made: (1) not all pages in a site are go-
ing to be modeled. (2) The pages represented in the
sitemap imply navigation. That is, if there is a rela-
tionship between two nodes, then there will be a link
in the web site. And, (3) only a minimum set of the
common sitemap elements introduced in the previous
section are going to be modeled, in order to simplify
the metamodel and to be the least restrictive as possi-
ble.
In order to guarantee these assumptions, the
sitemap metamodel shown in Figure 2 has been pro-
posed
1
. The Sitemap metaclass is the root of the
metamodel. One sitemap is composed of one or more
sitemap elements. This is represented by means of
the reference named elements. SitemapElement is the
abstract metaclass that contains the properties that are
1
The ecore file that holds the metamodel can be down-
loaded from: http:// www.lsi.us.es/ reinaqu/ org.mwacsl/
siteMap/metamodel/SiteMap.ecore
WEBIST 2010 - 6th International Conference on Web Information Systems and Technologies
112
Figure 2: Sitemap metamodel.
common to all the sitemap elements. That is, every
sitemap element has an id and a name. There are two
kinds of elements: Node and Area. One node repre-
sents a page or set of pages in a web site, while an
Area models a set of nodes that have a similar na-
vigation function. Node and Area are subclasses of
SitemapElement. The id attribute represents a unique
identifier, and, in the case of nodes, it is usually a hier-
archical numbering scheme. The root of the sitemap
has been modeled by means of the home relationship
that specify which relates Sitemap and Node.
Areas and nodes can have a parent (parent associ-
ation) or not. One node can access to its children by
means of the children reference (the opposite to par-
ent). Furthermore, one Area can contain one or more
nodes (nodes reference). Figure 1 shows three areas
that have been labeled as Main Navigation, Local Na-
vigation and Utility Navigation, respectively. The navi-
gational aim of an area has been modeled by means of
the enumeration named AreaType. These navigatio-
nal aims are inspired on (Kalbach, 2007), where three
main categories are defined: structural navigation, as-
sociative navigation and utility navigation. Usually,
sitemaps represent structural navigation, that is, na-
vigation that links pages that belong to the site hier-
archy. We have made an extension and our sitemap
metamodel also addresses utility navigation. Utility
navigation connects pages that help users utilize the
site and it supplements main navigation. Usually, it
contains links to pages that you would like to have on
every page, but do not have the same visual weight as
main navigation. Utility navigation is out of the site
hierarchy. As a consequence, utility navigation is de-
picted as an area that is drawn out of the site hierarchy
(Figure 1).
The AreaType enumeration has three different va-
lues: MAIN NAVIGATION, LOCAL NAVIGATION and
UTILITY NAVIGATION. Main navigation and local na-
vigation are two kinds of structural navigation. Main
navigation is the menu or set of links that appears on
every page of the site and links to all of the main sec-
tions. Local navigation guides a user to certain sec-
tions in a long page. As it can be seen, local and main
navigation have a behavior a bit different; as a conse-
quence, they have been modeled separately. Finally,
it has to be highlighted that associative navigation has
not been modeled because this kind of navigation is
out of the scope of sitemaps.
Annotations are the mechanism used to attach ad-
ditional information to a sitemap element (Annotation
metaclass). Annotations are composed of proper-
ties (Property metaclass) that represent a pair name-
value which is used to express additional node prop-
erties. The annotations association represents that one
sitemap element can contain zero or more annota-
tions. As this association is bidirectional, an Annota-
tion can access to its SitemapElement via its opposite
reference, annotatedElement.
Finally, two enumerations have been defined:
PagePurpose and NodeType. PagePurpose specifies
the main aim of a node. According to this criterion,
pages (and, as a consequence, nodes) can be classified
in three main categories (Kalbach, 2007): navigatio-
nal pages, content pages and functional pages. The
aim of navigational pages is to show the user the path
to its final destination. Examples of this kind of pages
are the home page or some gallery pages. Content
pages are those ones that users are looking for, while
functional pages are the ones that allow users to finish
certain tasks, such as searching for some information
or sending an e-mail. Node Type indicates if the node
represents a simple page (SIMPLE) or a similar set of
pages (STACK).
In order to maintain the metamodel as simple as
possible, some design decisions have been specified
by means of OCL constraints. As the metamodel
has been specified by means of EMF, the OCL con-
straints have been added to the metamodel with the
implementation of the Object Constraint Language
(OCL) for EMF-based models (EMP, 2007). These
constraints express the following ideas: (I) as utility
navigation is out of the sitemap hierarchy, utility na-
vigation areas do not have a parent. (II) The main na-
vigation area must have a parent. (III) A node cannot
be parent of itself. (IV) The node that is the sitemap
home cannot be the children of any other node. (V)
A node that belongs to an area has the same parent
as the area in which is included, only if the area is
not a utility navigation area. (VI) Nodes that be-
long to the main navigation area cannot be included
in other areas. Links that belong to the main naviga-
tion area have to appear in every page, so it has no
sense that nodes in the main navigation area are in
other areas, because they will appear twice. (VII) If
SITEMAPS FROM A MODEL DRIVEN PERSPECTIVE - A First Step for Bridging the Gap between Information
Architecture and Navigation Design
113
website xhtml
XMLType
XMLNamespace
«import»
«import»
«import»
Figure 3: Metamodels Relationship.
the node is an external site, then its source property
must have a value (the URL that points to that ex-
ternal site). (VIII) The root of the sitemap does not
have a parent; its pageType property should be set
to FUNCTIONAL PAGE and its type to SIMPLE. (IX)
There only can be one main navigation area in the
whole site. (X) Nodes and areas are disjointed ele-
ments.
4 WEBSITE METAMODEL
In order to avoid building the whole website meta-
model from scratch, three other metamodels have
been reused. Thus, the only package defined from
scratch has been website. Figure 3 shows the rela-
tionships among the different packages which con-
tain the metamodels. The ecore file that holds the
website package has been defined manually
2
, while
the other three metamodels have been obtained au-
tomatically. The xhtml, XMLType and XMLNames-
pace packages have been obtained by means of the
XSD model importer component included in the EMF
Eclipse plug-in, as it is mentioned in (Gronback,
2009). The importer only needs a XML Schema
Definition file which defines the structure and con-
straints of XHTML files. One of these files is the
one specified by the W3C consortium and it can be
found at http://www.w3.org/2002/08/xhtml/xhtml1-
strict.xsd. As a result of the importation an
xhtml.
¬
ecore
file is obtained.
Figure 4 depicts the metaclasses included in the
website metamodel. The WebSite metaclass is the
root of the metamodel. A website is composed of a
set of HTMLFile. An XHTML file holds some content.
The content is modeled by means of the relationship
named content. The DocumentRoot metaclass is at
the other end of this relation. This metaclass belongs
to the xhtml metamodel, which has been depicted by
means of the
<<from xhtml>>
stereotype. The Doc-
umentRoot metaclass is also related to the HTMLType
metaclass, but as these two metaclasses belong to the
xhtml, the relation has not been depicted in Figure 4.
2
This file can be downloaded from: http://
www.lsi.us.es/reinaqu/ org.mwacsl/webSite/metamodel/
website.ecore
In addition to this, a set of minor modifications
have been done to the original
xhtml.ecore
file ob-
tained from the importation. The AContent, Flow, In-
line and TitleType metaclasses are affected by these
modifications as it is stated in (Gronback, 2009). This
set of changes is needed because there is no way to
declare text elements between XHTML labels such as
<li>
or
<p>
. Finally, there is also another set of mo-
difications whose goal is to make easier the model to
model transformations that will be introduced in the
next section. The AType metaclass is affected by this
changes, concretely, there is an attribute named class
whose type has been modified from NMTOKENS to
NMTOKEN.
5 M2M TRANSFORMATION
The transformation has as input a model that con-
forms to the sitemap metamodel explained in Sec-
tion 3 and returns a model that conforms to the web-
site metamodel explained in Section 4. The main idea
behind the transformation is that, objectively, utility
and structural navigation are a list of links to vari-
ous pages on the website (Grannell, 2007). As the
focus of the transformation itself is to obtain a draft
of utility and structural navigation, the web pages
that are going to be generated are only a skeleton
containing the structural and utility navigation links
without any other content. Thus, the generated pages
are structured into different areas that have been se-
parated by means of XHTML
<div>
tags (Figure 5).
These tags are part of the
<body>
section and each
one will have a class attribute whose value will be
used later to apply some visual style, by means of
Cascade Stylesheets (CSS) files. Figure 5 depicts the
values of the class attributes corresponding to the di-
fferent
<div>
sections. Thus, container is the class
of the outer
<div>
, while the class of the inner ones
are: header, utilityNavigation, mainNavigation, local-
Navigation, mainContent and footer.
Figure 6 shows how a node is transformed into
an xhtml file. The left hand of the figure shows the
Figure 4: Website package in the website metamodel.
WEBIST 2010 - 6th International Conference on Web Information Systems and Technologies
114
#container
#utilityNavigation
#header
#mainNavigation
#localNavigation
#mainContent
#footer
Figure 5: Structure of the generated web pages.
:Node :HTMLFile
:DocumentRoot
:HTMLType
:HeadType
:BodyType
:TitleType
id = container
: DivType
id = header
: DivType
id = utilityNavigation
: DivType
id = mainNavigation
: DivType
id = localNavigation
: DivType
id = mainContent
: DivType
id : String = footer
: DivType
:UlType
:LiType :LiType
:AType
:AType
...
:UlType
:LiType :LiType
:AType
:AType
...
:UlType
:LiType :LiType
:AType
:AType
...
Figure 6: Result of the mapping applied to a node.
piece of the source model that is going to be trans-
formed, while the right hand shows the result of the
mapping. As it can be seen, each node generates an
instance of the HTMLFile metaclass. A HTML file has
a content, whose type is DocumentRoot. The root of
the HTML document is related to an instance of the
HTMLType metaclass, which represents the
<html
¬
>
tag. The figure also shows how different instances
of the DivType metaclass are generated, one for each
<div>
tag mentioned in the previous paragraph. It is
also worth to be noticed that the only instances of Di-
vType related to instances of UlType are the ones that
represent utility, main and local navigation. In this
case, the UlType metaclass models
<ul>
tags. Finally,
each UlType instance is related to a number of LyType
instances. The number of instances will depend on
the number of nodes included in each navigation area.
Thus, for example, if the utility navigation area in-
cludes three nodes, three instances of LyType will be
generated; each one will represent a
<li>
tag contain-
ing a link (modeled by means of AType). If we look
at the node 5.3 (Figure 1), the XHTML file generated
should looks like the one depicted in Figure 7.
The transformation has been implemented with
the Operational QVT (The Eclipse Foundation, 2010)
included in the M2M subproject of the Eclipse Mod-
Figure 7: File 5.3-all.html obtained from the node whose id
is 5.3.
eling Project. The transformation
3
is composed of
nine mappings, two constructors and twelve queries.
Finally, it can be noticed that until now there are
no details about presentation. They are specified
by means of different CascadeStyleSheet (CSS) files.
Thus, a set of CSS files have been defined for diffe-
rent presentations and layouts of navigation. This in-
formation is included in the transformation by means
of four configuration properties, named cssContainer,
cssUtilityNav, cssMainNav, cssLocalNav. Each pro-
perty should hold the URI of the CSS file that will
define the visual aspect of the item. For example, the
property cssMainNav will have the URI of the CSS
file that defines the visual aspect of the main naviga-
tion.
6 RELATED WORK
As one of the aims of the approach is to bring closer
two areas of research, information architecture and
navigation design, different areas of knowledge have
been surveyed in order to found related work. As
far as we know, information architecture researchers
do not formalize sitemaps, as there is no a standard
best practice. The only attempt of formalizing infor-
mation architecture deliverables can be found in (Bo-
gaards, 2004), where the building blocks of many IA
deliverables are specified using Dublin core attributes.
Therefore, it can be stated that the focus of most of the
IA publications is on describing how sitemaps can be
used. Thus, for example, in (Brown, 2006), sitemaps
are explained in a communication design context. In
this book, and also in (Kalbach, 2007), the Visual Vo-
cabulary defined by J.J. Garret (Garrett, 2002) is pro-
posed as a modeling notation. This notation is broader
than the one we have proposed in this paper, because
in addition to sitemaps it deals with web flows. We
prefer to use the separation of concerns principle and
define a metamodel whose concerns are related only
to sitemaps.
Concerning to Web Engineering, there are seve-
ral approaches that bet for a model-driven develop-
ment process such as WebML (Ceri et al., 2003),
3
The source code can be downloaded from: http://
www.lsi.us.es/ reinaqu/org.mwacsl/siteMap/transformations/
sitemap2website.qvto.
SITEMAPS FROM A MODEL DRIVEN PERSPECTIVE - A First Step for Bridging the Gap between Information
Architecture and Navigation Design
115
UWE (Koch, 2001), OOWS (Pastor et al., 2005),
OOHDM (Schwabe and Rossi, 1998).
WebML authors have not defined an explicit meta-
model. The concepts managed in this approach have
been defined in the XML technical space by means of
a Document Type Definition (DTD). However, other
authors have defined these concepts by means of an
explicit metamodel (Schauerhuber et al., 2006) or a
UML 2.0 profile (Moreno et al., 2007).
UWE also has been specified by means of a meta-
model implemented as a UML profile (Kroib and
Koch, 2008). Furthermore, a set of model to model
transformations has been proposed (Koch, 2006).
The transformations map Web requirement models to
UWE functional models. They also define mappings
between requirement models and architectural mod-
els. Finally, they provide mappings to platform spe-
cific models.
OOHDM has been extended to follow an approach
based on MDA (Schmid and Donnerhak, 2005).
Thus, in (Schmid, 2004) a behavioral semantics defi-
nition of the OOHDM core features and business pro-
cess is made. The models proposed in this approach
are used as PIM in a MDA-based development pro-
cess.
OOWS has also been extended by means of a
model-driven extension in order to support business
processes integration (Torres and Pelechano, 2006).
The business process model is the start point for trans-
formations. This model is mapped into a default
navigation model. Later, the designer has to refine
this model in order to be transformed into a concrete
web technology. Transformations have been specified
using QVT Operational Mappings.
Other approaches that are applying a model-
driven approach for web development are: Hy-
perDE (Nunes and Schwabe, 2006), MIDAS (C
´
aceres
et al., 2004), Moreno et al. (Moreno and Valle-
cillo, 2005), Muller et al. (Muller et al., 2005),
W2000 (Baresi et al., 2006) and WebSA (Meli
´
a and
Cachero, 2004).
As it can be seen in the previous paragraphs, there
is a great number of web engineering approaches.
In order to bring closer many of these approaches,
a project named MDWEnet has come up (Vallecillo
et al., 2007). One of the main aims of this project
is to achieve interoperability of the different model-
driven web engineering methods. We also want to
achieve interoperability between web engineering de-
liverables and information architecture deliverables.
Thus, we can state that the main difference between
our approach and the previous ones is that they do not
have into account information architecture delivera-
bles and concepts.
7 CONCLUSIONS AND FURTHER
WORK
The main contribution of this paper is to take a step
forward to bring closer information architecture and
navigation design in the context of a model driven ap-
proach. The general approach consists in metamodel-
ing and transforming by means of model transforma-
tions the deliverables used by the different audiences
in a project, specially information architects and de-
signers. The paper is centered on sitemaps, one of the
most widely used IA deliverables, and it proposes a
metamodel focused on structural and utility naviga-
tion, in such a way that thanks to a set of model to
model transformations, a skeleton of XHTML pages
is obtained. Although the aim of the transformations
proposed in the paper is to obtain XHTML, this ap-
proach is being applied to other deliverables, such as
statecharts or webflows.
ACKNOWLEDGEMENTS
This work has been partially supported by the Span-
ish Ministery of Science and Technology and FEDER
funds: TIN-2007-67843-C06-03 and TIN2007-64119.
REFERENCES
Azam, F., Li, Z., and Ahmad, R. (2005). Introducing UML
profile for modelling information architecture of web
applications. In Proceedings of the 11th Joint Interna-
tional Computer Conference - JICC 2005, pages 245–
250.
Baresi, L., Colazzo, S., Mainetti, L., and Morasca, S.
(2006). Web Engineering, chapter W2000: A Mod-
eling Notation for Complex Web Applications, pages
335–408. Springer.
Bogaards, P. J. (2004). Info design/arch deliver-
able schemas. http://www.bogieland.com/ infode-
sign/resources/misc/iadelschemas.htm.
Brown, D. M. (2006). Communicating Design: Develo-
ping Web Site Documentation for Design and Plan-
ning, chapter 8, Site Maps. New Riders.
C
´
aceres, P., de Castro, V., and Marcos, E. (2004). Naviga-
tion modelling from a user services oriented approach.
In Proceedings of the Conference on Advances in In-
formation Systems (CAISE04), pages 150–160.
Ceri, S., Fraternali, P., Bongio, A., Brambilla, M., Comai,
S., and Matera, M. (2003). Designing Data-Intensive
Web Applications. Elsevier Science.
Clark, T., Sammut, P., and Willans, J. (2008). Applied Meta-
modelling. A Foundation For Language Driven Deve-
lopment. 2nd Edition. Ceteva.
WEBIST 2010 - 6th International Conference on Web Information Systems and Technologies
116
Curtis, N. (2008). Audiences & artifacts. Bulletin of the
American Society for Information Science and Tech-
nology, 34(6):20–26.
EMP (2007). EMF Technology OCL Project. Avalaible at:
http://www.eclipse.org/emft/projects/ocl]ocl.
Garrett, J. J. (2002). Visual vocabulary for infor-
mation architecture. version 1.1b. Avalaible at:
www.jjg.net/ia/visvocab.
Grannell, C. (2007). The Essential Guide to CSS and HTML
Web Design, chapter 5. Using Links and Creating Na-
vigation, pages 147–232. friendsof. APress.
Gronback, R. C. (2009). Eclipse Modeling Project: A
Domain-Specific Language (DSL) Toolkit. Addison-
Wesley Professional.
Kalbach, J. (2007). Designing Web Navigation. O’Reilly
Media, Inc.
Koch, N. (2001). Software Engineering for Adaptive
Hypermedia System: Reference Model, Modeling
Techniques and Development Process. PhD thesis,
Ludwig-Maximilian-Universitat Munchen.
Koch, N. (2006). Transformations techniques in the model-
driven development process of uwe. In In Proc. Of 2nd
Model-Driven Web Engineering Workshop (MDWE
2006), volume 155.
Kroib, C. and Koch, N. (2008). UWE metamodel and pro-
file: User guide and reference. Technical Report 0802,
Ludwig-Maximilians-Universitt Mnchen (LMU).
Meli
´
a, S. and Cachero, C. (2004). An MDA approach for
the development of web applications. In Proceedings
of the International Conference on Web Engineering
(ICWE04), pages 300–305.
Moreno, N., Fraternali, P., and Vallecillo, A. (2007).
WebML modelling in UML. IET Software, 1(3):67–
80.
Moreno, N. and Vallecillo, A. (2005). A model-based ap-
proach for integrating third party systems with web
applications. In Proceedings of International Confer-
ence on Web Engineering (ICWE’05), number 3579,
pages 441–452.
Muller, P. A., Studer, P., Fondement, F., and B
´
ezivin, J.
(2005). Platform independent web application mod-
eling and development with Netsilon. Software and
System Modeling, 00:1–19.
Nunes, D. A. and Schwabe, D. (2006). Rapid prototyp-
ing of web applications combining domain specific
languages and model driven design. In Proceed-
ings of International Conference on Web Engineering
(ICWE’06).
OMG (2005). Meta Object Facility (MOF) 2.0
Query/View/Transformation Specification. Technical
report, OMG.
Pastor, O., Pelechano, V., Fons, J., and Abrahao, S.
(2005). Conceptual Modelling of Web Applications:
the OOWS Approach. In Web Engineering. Theory
and Practice of Metrics and Measurement for Web De-
velopment, pages 277–302.
Schauerhuber, A., Wimmer, M., and Kapsammer., E.
(2006). Bridging existing web modeling languages to
model-driven engineering: A metamodel for webml.
In Proceedings of the 2nd Int. Workshop on Model-
Driven Web Engineering (MDWE’06).
Schmid, H. A. (2004). Model Driven Architecture with
OOHDM. In Matera, M. and Comai, S., editors,
ICWE Workshops, pages 12–25. Rinton Press.
Schmid, H. A. and Donnerhak, O. (2005). OOHDMDA. an
MDA approach for OOHDM. In Proceedings of Inter-
national Conference on Web Engineering (ICWE’05),
number 3579, pages 569–574.
Schwabe, D. and Rossi, G. (1998). Developing Hypermedia
Applications using OOHDM. In Proceedings of Work-
shop on Hypermedia Development Processes, Meth-
ods and Models, Hypertext’98.
Steinberg, D., Budinsky, F., Paternostro, M., and Merks, E.
(2008). EMF: Eclipse Modelling Framework, 2nd Ed.
Addison-Wesley.
The Eclipse Foundation (2010). Operational QVT. Avail-
able at: http://www.eclipse.org/m2m/qvto/doc/.
Torres, V. and Pelechano, V. (2006). Building business
process driven web applications. In Dustdar, S., Fi-
adeiro, J., and Sheth, A., editors, Proceedings of the
BPM2006, volume 4102 of Lecture Notes in Com-
puter Science.
Vallecillo, A., Koch, N., Cachero, C., Comai, S., Fraternali,
P., Garrigs, I., Gmez, J., Kappel, G., Knapp, A., Mat-
era, M., Meli, S., Moreno, N., Pr
¨
oll, B., Reiter, T.,
Retschitzegger, W., Rivera, J. E., Schauerhuber, A.,
Schwinger, W., Wimmer, M., and Zhang, G. (2007).
MDWEnet: A practical approach to achieving inter-
operability of model-driven web engineering meth-
ods. In In Proc. of 3rd Model-Driven Web Engineering
Workshop (MDWE 2007), volume 261. CEUR-WS.
SITEMAPS FROM A MODEL DRIVEN PERSPECTIVE - A First Step for Bridging the Gap between Information
Architecture and Navigation Design
117