<!DOCTYPE article PUBLIC "-//NLM//DTD JATS (Z39.96) Journal Archiving and Interchange DTD v1.0 20120330//EN" "JATS-archivearticle1.dtd">
<article xmlns:xlink="http://www.w3.org/1999/xlink">
  <front>
    <journal-meta />
    <article-meta>
      <title-group>
        <article-title>An Attempt to Fathom the Role of Annotations in User- Centered Design Process</article-title>
      </title-group>
      <contrib-group>
        <contrib contrib-type="author">
          <string-name>Jean Luc Hak</string-name>
          <email>jean-luc.hak@softeam.fr</email>
          <xref ref-type="aff" rid="aff0">0</xref>
          <xref ref-type="aff" rid="aff1">1</xref>
        </contrib>
        <contrib contrib-type="author">
          <string-name>Olivier Nicolas</string-name>
          <email>olivier.nicolas@softeam.fr</email>
          <xref ref-type="aff" rid="aff1">1</xref>
        </contrib>
        <contrib contrib-type="author">
          <string-name>Philippe Palanque</string-name>
          <email>palanque@irit.fr</email>
          <xref ref-type="aff" rid="aff0">0</xref>
        </contrib>
        <contrib contrib-type="author">
          <string-name>Marco Winckler</string-name>
          <xref ref-type="aff" rid="aff0">0</xref>
          <xref ref-type="aff" rid="aff2">2</xref>
        </contrib>
        <aff id="aff0">
          <label>0</label>
          <institution>Institut de Recherche en Informatique de Toulouse</institution>
          ,
          <country country="FR">France</country>
        </aff>
        <aff id="aff1">
          <label>1</label>
          <institution>Softeam</institution>
          ,
          <addr-line>Toulouse</addr-line>
          ,
          <country country="FR">France</country>
        </aff>
        <aff id="aff2">
          <label>2</label>
          <institution>Université Nice Sophia Antipolis</institution>
          ,
          <country country="FR">France</country>
        </aff>
      </contrib-group>
      <abstract>
        <p>This paper investigate the role played by annotation along the development process of interactive systems. Empirical observations have demonstrated that development teams often make an extensive use of annotations, mainly as a communication support. Whilst the use of annotation is a fact (also supported by many prototyping environment, IDE and model editors), very few studies have investigated the use of the annotations for building interactive systems. In this paper, we propose a process to explain this co-evolution of annotations and artefacts along the development process of interactive systems. The ultimate goal is provide mechanisms that could help the development team to follow design decisions using annotations as a support.</p>
      </abstract>
      <kwd-group>
        <kwd>Annotations</kwd>
        <kwd>interactive system</kwd>
        <kwd>design process</kwd>
      </kwd-group>
    </article-meta>
  </front>
  <body>
    <sec id="sec-1">
      <title>-</title>
      <p>
        Design is a problem-solving process whose objective is find a way to implement
requirements, respecting constraints, and ensure good quality. According to the ISO
standard 9241-210 (2008) [
        <xref ref-type="bibr" rid="ref1">1</xref>
        ], the design process of an interactive systems is iterative:
design solutions are created, tested, revised, and improved until the development team
produces a proper version of the fully fledge system. This process produces two types
of results: a specification of the design solution to be implemented (the interactive
system) and a set of design decisions that drive the evolution of the design along the
iteration cycles.
      </p>
      <p>
        User interface prototypes are the most common type of artifacts used to specify
design solutions for interactive systems. In early phases of the development process,
drawings are acceptable as prototypes to support ideation of the product but as the
process advances, drawings are replaced by interactive specifications and then by
executable prototypes. It is interesting to notice that prototypes are useful and necessary but
they are not sufficient to fully specify an interactive systems. On one hand, prototypes
are not self-explicative, which is illustrated by the fact that annotations have to be used
to explain for instance the use of icons in a design. On the other hand, prototypes cannot
directly inform other important aspects of the interactive system, for that other artefacts
such as task models [
        <xref ref-type="bibr" rid="ref3">3</xref>
        ], dialog models and interaction models [
        <xref ref-type="bibr" rid="ref4">4</xref>
        ] must be used.
      </p>
      <p>
        Empirical observations have demonstrated that development teams often make an
extensive use of annotations as a communication support [
        <xref ref-type="bibr" rid="ref2">2</xref>
        ]. It is a basic assumption
that decisions made by the development team will create iteration along the process and
affect the way artefacts must evolve. Indeed, prototypes and artefacts evolve along the
development process and many of the design decisions might be described in the form
of annotations. The dangling question is: What happens with annotations in a UCD
development process? This paper presents a micro-development process that is aimed
to describe the life cycle of annotations and how annotations co-evolve with artifacts
along the development process of interactive system. Moreover, we discuss how
stakeholders might make better use of annotations for improving the communication in a
UCD process.
2
      </p>
    </sec>
    <sec id="sec-2">
      <title>Related Work</title>
      <p>
        The first studies about annotations started with the identification of common practices
by university students on their paper textbooks [
        <xref ref-type="bibr" rid="ref5">5</xref>
        ]. Paper-based annotations work as a
conceptual model for electronic documents. Therefore, common definitions of
annotation often refer to text documents [
        <xref ref-type="bibr" rid="ref6">6</xref>
        ]. Kahn &amp; Koivunen [
        <xref ref-type="bibr" rid="ref7">7</xref>
        ] define annotations as “user
made statements”, consisting in a body (i.e. text note or graphical content), a link (the
so-called anchor) to the target which include a location within the document as well as
other metadata. As we shall see, these three elements (body, target and link) are core
concepts not only for paper-based or electronic documents but they are essential to
understand how annotations applies to the many artefacts used to build interactive system
as well.
      </p>
      <p>
        In [
        <xref ref-type="bibr" rid="ref8">8</xref>
        ], Li et al. have defined a classification of annotation approach for
Computeraided Design. This classification identify the following categories of attributes that
complete the specification of annotation: targeted media, audience, rendering system,
usage and function, representation, and storage location. This classification of
annotations bring another complementary view of annotations. Based on the literature on text
and electronic documents [
        <xref ref-type="bibr" rid="ref10 ref11 ref12 ref5 ref6 ref7 ref8 ref9">5-12</xref>
        ], we can summarize three main functions played by
annotations: to enrich a document, to support communication and to support an
intention/activity carried out by the author of the annotation. Whilst most of the literature in
the matter refers to text documents, that classification is relevant for the development
of interactive systems.
      </p>
      <p>
        More and more prototyping and IDE environments at least some basic mechanisms
for annotating artifacts [
        <xref ref-type="bibr" rid="ref13">13</xref>
        ]. Very recently, the W3C proposed a standard called Web
Annotation Data Model which was created for specifying a model and a format to
ensure the sharing and reuse of annotations across different hardware and platform. All
these tools testimony of the increasing importance of annotation for building interactive
systems.
      </p>
    </sec>
    <sec id="sec-3">
      <title>Use of Annotations by Stakeholders</title>
      <p>
        Very few studies have investigated how annotations could affect the development of
interactive systems in a UCD process. The study performed by Gutierrez et al. [
        <xref ref-type="bibr" rid="ref2">2</xref>
        ]
pointed out that annotations are used by members of development teams to: record the
results of discussion including decisions and upcoming tasks, communicate and inform
other team members of the work done, gather internal and external feedback on
artefacts stored in the workspace, conduct usability evaluations by documenting
information and by recording conversation between design teams and UX experts, justify
design choices, and document the design choices by describing them retrospectively.
      </p>
      <p>
        The work of Gutierrez et al. [
        <xref ref-type="bibr" rid="ref2">2</xref>
        ] does not make any distinction between types of
stakeholders. It is worthy of recalling the classification of Winckler et al. [
        <xref ref-type="bibr" rid="ref16">16</xref>
        ] who
identify two groups of stakeholders taking part in a UCD process: the development
team (which encompass roles having responsibility with respect to the production of
artefacts) and external members (such as clients and end-users) who provide opinions,
requirements, and/or constraints for the design. Interesting to notice that these two
groups of stakeholders collaborate along the development process in iterative cycle.
      </p>
      <p>
        Thus, regarding the use of annotations, it is possible to identify two main roles: the
writer of the annotation and the readers. More generally, annotations is a mean to
convey a large variate of intentions to the reader. Naghash et al [
        <xref ref-type="bibr" rid="ref12">12</xref>
        ] suggest 6 different
usages of annotations: i) clarifying and explaining the design; ii) verifying and
requesting a verification from other designers or users; iii) exploring by asking questions to
obtain more details on end users’ needs; iv) altering or requesting an alteration proposed
by the end users; v) confirming and giving feedback on a design; and vi) asking
questions to the designers.
      </p>
      <p>Whilst the development team are responsible for creating artefacts and make use of
annotation to coordinate their activities, external member might use annotations to
express opinions and comment on what is being developed. In the rest of this paper, we
assume that annotations are suitable communication tools that must be available to the
diverse stakeholder (readers/writers) that take part in the development process of
interactive systems.
4</p>
    </sec>
    <sec id="sec-4">
      <title>Life Cycle of Design Artefacts</title>
      <p>The Figure 1 represent the lifecycle of a design artefact within a UCD process. This
process acknowledges that the creation artefacts are a starting point for the work.
Nonetheless, it does not impose any artefact to be created, which might be dependent on
project needs. After creation, the design team should be able to perform following tasks:
• Edit the artefact, either for enriching it, for correcting it, or for making to match
new requirements.</p>
      <p>• Archive the artefact within a workspace for future use.</p>
      <p>• Submit the artefact for evaluation, which leads to the creation of a new artefacts
such as an “evaluation report”. The results of an evaluation have their own lifecycle
within the design process and might lead to the creation/updating of other artefact (ex.
new user interface design is created following recommendations of a usability
evaluation). These outcomes are represented by the red arrow labelled “External influence”.
• Dispose the artefact, when it is no longer useful.</p>
      <p>Artefacts might pre-exist, for that, the process includes an “existing” state
represented by the grey rectangle in Figure 1. Depending on the collaborative tools used by
the design team, a distinction can be made between a local copy and a shared copy of
the artefact. This duplication of artefacts require an effort of synchronization for the
design team who have to manage the consistency between the local copies of each
contributor with the shared copy.
It is interesting to notice that annotations are, at some extension, a special case of
artefact. Annotations depends on the artefact they are attached to, but they possess their
own lifecycle which can be evolving independently from the design artefact. One
particular aspect of the annotations in a UCD process is that annotations can be related to
certain versions of the artefact but not each of its version (e.g. an annotation indicating
to fix an error).</p>
      <p>The life cycle of an annotation shown by Figure 2 starts with a decision. The creation
might be motivated by a variety of reasons and influenced by external influences (e.g.
in reaction of other annotations, of the content of an artefact). This creation can occurs
when the artefact is being consulted, edited or evaluated. After its creation, the
annotation is in a private state and only visible to its author. In this state, the annotation can
be updated and reviewed anytime by its author. Depending on the annotation, its author
can decide to publish it to make it visible to other members of the design team.</p>
      <p>Published annotations are presented to the different actors of the design process who
can argue with the information contained in the annotation (which can lead to the
creation of an annotation as a response) or who can validate the annotation to ensure its
relevance toward the artefact and to assess its content.</p>
      <p>When validated by stakeholders, annotations are kept as a reference about activities
taking place along the design process (e.g. information on the requirements,
appreciation marks of the design, highlight of problems). These information can have an impact
on other design artefacts which is represented by the red arrow. For instance, a problem
on a prototype identified with an annotation can motivate and justify a decision to edit
the prototype in order to fix it. After that, annotations can be managed by indicating
that it has been processed.</p>
      <p>If the annotation is not validated, the annotation will not have an impact on other
artefacts or for future uses in its current state. Similarly to artefacts, annotations can be
archived for keeping the annotation in its current state or disposed when it is no longer
useful.</p>
    </sec>
    <sec id="sec-5">
      <title>Reciprocal Influence of Annotations and Artefacts</title>
      <p>Annotations can also affect the evolution but are also be affected by the evolution of
the artefact itself. Thus, while annotation have their own lifecycle, this lifecycle is
interweaved by the life cycle of annotated artefacts.</p>
      <p>An annotation is created or updated on an artefact in reaction to the content of the
artefact as illustrated by the red arrow “Induce the creation of annotations” in the Figure
3. After the creation of the annotation, the annotation can be attached or detached to
any artefact or fragment of artefact to include it to its target list represented with orange
arrows.
6</p>
      <p>In return, the annotation can have an impact on the artefacts it is attached to. Indeed,
annotation can be used a medium of communication for discussing, for contributing to
the elaboration of an artefact, to point out modifications to make on the artefact. The
content annotations can be varied from the topic discussed, the intentions of the persons
involved in the annotations, the precision of the information, and the quantity of
information contained. Depending on this content, several type of impact can be identified:
no modification required (e.g. for informative annotations), localized modification
restricted to the artefact (e.g. correcting a typo, adding a precision) or global modification
that can impact other artefacts (e.g. appearance of new requirements or adjustment of
existing requirements). While annotations may have an impact on design artefacts, they
are not always factual and can reflect opinions that should be nuanced and
cross-referenced with others opinions or concrete facts prior to taking decisions. Thus, annotations
can be used to motivate or support a decision regarding the artefact as illustrated by the
red arrow “Induce or support a decision”.</p>
      <p>Regarding the impacts of an annotation to the update of an artefact, their weight can
depends on several factors. Indeed, annotations can point out problems directly, reflect
an opinion or unverified data from different sources and thus, the information conveyed
needs to be validated. This can be done by several means such as checking the person
involved in the discussion, analyzing the relevance or the trustworthiness of the
information. After the validation, another aspect can be taken into account that can influence
the impact on targeted artefacts. Indeed, a decision process can be integrated prior to
the editing of the artefact. This decision process can assess the cost of the editing and
its planning if the editing has been adopted by the design team.</p>
      <p>Another interaction between the artefact and the annotation is the mutual update they
can trigger. When updating an artefact, the content of each annotations attached to the
artefact may be questioned or the state of the annotation can be updated to match it with
the new state of the artefact.</p>
    </sec>
    <sec id="sec-6">
      <title>Discussion and future work</title>
      <p>Annotations are a versatile tool for documenting the design by associating
documents or by explaining. They can be used for communication, for planning tasks, for
reviewing the design and by allowing stakeholders to highlight problems, to question
the design or to give opinion.</p>
      <p>This presented the connections between annotations and artefacts along the
development process of interactive systems. This work is an attempt to promote annotations
as a first class artefact that could be used for tracking design decisions along the
development process of interactive.</p>
      <p>Currently, we have implemented tools that allows to connect annotations to multiple
artefacts. Our ultimate goal is to develop tools that could help the development tool to
make a better use of annotations to communicate, trace design decisions and follow the
evolution of artefacts along the development process. These tools are suitable for a
demonstration and future work will encompass the evaluation of them with real users.
In a long run, we expect that our tools would be able to collect design decisions along
many projects. The analyses of design decisions and their association with the evolution
of artefacts, might provide useful data for have a better understanding on the real
practice of UCD process.</p>
    </sec>
  </body>
  <back>
    <ref-list>
      <ref id="ref1">
        <mixed-citation>
          1. ISO 9241-210
          <year>2008</year>
          .
          <article-title>Ergonomics of human system interaction-Part 210: Human-centred design for interactive systems</article-title>
          . Standard. International Organization for Standardization, Geneva, CH.
        </mixed-citation>
      </ref>
      <ref id="ref2">
        <mixed-citation>
          2.
          <string-name>
            <given-names>Marisela</given-names>
            <surname>Gutierrez</surname>
          </string-name>
          <string-name>
            <surname>Lopez</surname>
          </string-name>
          , Gustavo Rovelo, Mieke Haesen, Kris Luyten,
          <string-name>
            <given-names>Karin</given-names>
            <surname>Coninx</surname>
          </string-name>
          .
          <article-title>Capturing Design Decision Rationale with Decision Cards</article-title>
          .
          <source>INTERACT (1)</source>
          <year>2017</year>
          :
          <fpage>463</fpage>
          -
          <lpage>482</lpage>
        </mixed-citation>
      </ref>
      <ref id="ref3">
        <mixed-citation>
          3.
          <string-name>
            <given-names>Célia</given-names>
            <surname>Martinie</surname>
          </string-name>
          , David Navarre,
          <string-name>
            <given-names>Philippe A.</given-names>
            <surname>Palanque</surname>
          </string-name>
          ,
          <string-name>
            <given-names>Camille</given-names>
            <surname>Fayollas</surname>
          </string-name>
          .
          <article-title>A generic tool-supported framework for coupling task models and interactive applications</article-title>
          .
          <source>EICS</source>
          <year>2015</year>
          :
          <fpage>244</fpage>
          -
          <lpage>253</lpage>
          .
        </mixed-citation>
      </ref>
      <ref id="ref4">
        <mixed-citation>
          4.
          <string-name>
            <surname>Winckler</surname>
            ,
            <given-names>M.</given-names>
          </string-name>
          ,
          <string-name>
            <surname>Vanderdonckt</surname>
            ,
            <given-names>J.</given-names>
          </string-name>
          ,
          <string-name>
            <surname>Trindade</surname>
            ,
            <given-names>F.</given-names>
          </string-name>
          ,
          <string-name>
            <surname>Stanciulescu</surname>
            ,
            <given-names>A.</given-names>
          </string-name>
          <article-title>Cascading Dialog Modeling with UsiXML</article-title>
          .
          <source>International Workshop on the Design</source>
          ,
          <article-title>Verification and Specification of Interactive Systems (DSVIS'</article-title>
          <year>2008</year>
          ). Kingston, Ontario, Canada, July
          <volume>16</volume>
          -18
          <year>2008</year>
          . Springer LNCS 5136. pp.
          <fpage>121</fpage>
          -
          <lpage>135</lpage>
          .
        </mixed-citation>
      </ref>
      <ref id="ref5">
        <mixed-citation>
          5.
          <string-name>
            <surname>Catherine</surname>
            <given-names>C.</given-names>
          </string-name>
          <string-name>
            <surname>Marshall</surname>
          </string-name>
          .
          <year>1997</year>
          .
          <article-title>Annotation: from paper books to the digital library</article-title>
          .
          <source>In Proceedings of the second ACM international conference on Digital libraries (DL '97)</source>
          . ACM, New York, NY, USA,
          <fpage>131</fpage>
          -
          <lpage>140</lpage>
          . DOI=
          <volume>10</volume>
          .1145/263690.263806 http://doi.acm.
          <source>org/10</source>
          .1145/263690.263806
        </mixed-citation>
      </ref>
      <ref id="ref6">
        <mixed-citation>
          6.
          <string-name>
            <surname>Bringay</surname>
            <given-names>S.</given-names>
          </string-name>
          ,
          <string-name>
            <surname>Barry</surname>
            <given-names>C.</given-names>
          </string-name>
          ,
          <string-name>
            <surname>Charley</surname>
            <given-names>J.</given-names>
          </string-name>
          ,
          <string-name>
            <surname>Annotations</surname>
            <given-names>:</given-names>
          </string-name>
          <article-title>A new type of document in the Electronic Health Record</article-title>
          .
          <source>Paper presented at the 2nd International Conference on Document Research and Development in Sciences, arts and business: DOCAM</source>
          <year>2004</year>
          , University of California, Berkeley, Etats-Unis, octobre
          <year>2004</year>
          .
        </mixed-citation>
      </ref>
      <ref id="ref7">
        <mixed-citation>
          7.
          <string-name>
            <given-names>José</given-names>
            <surname>Kahan</surname>
          </string-name>
          and
          <string-name>
            <surname>Marja-Ritta Koivunen</surname>
          </string-name>
          .
          <year>2001</year>
          .
          <article-title>Annotea: an open RDF infrastructure for shared Web annotations</article-title>
          .
          <source>In Proceedings of the 10th international conference on World Wide Web (WWW '01)</source>
          . ACM, New York, NY, USA,
          <fpage>623</fpage>
          -
          <lpage>632</lpage>
          . DOI=http://dx.doi.org/10.1145/371920.372166
        </mixed-citation>
      </ref>
      <ref id="ref8">
        <mixed-citation>
          8.
          <string-name>
            <surname>Li</surname>
            ,
            <given-names>C</given-names>
          </string-name>
          &amp;
          <string-name>
            <surname>Mcmahon</surname>
            ,
            <given-names>Chris</given-names>
          </string-name>
          &amp; Newnes,
          <string-name>
            <surname>Linda.</surname>
          </string-name>
          (
          <year>2009</year>
          ).
          <article-title>Annotation in design processes: Classification of approaches</article-title>
          .
          <source>DS 58-8: Proceedings of ICED 09, the 17th International Conference on Engineering Design</source>
          .
          <fpage>251</fpage>
          -
          <lpage>262</lpage>
          .
        </mixed-citation>
      </ref>
      <ref id="ref9">
        <mixed-citation>
          9.
          <string-name>
            <given-names>Manuel</given-names>
            <surname>Zacklad</surname>
          </string-name>
          . Annotation : attention, association, contribution.
          <source>Annotations dans les Documents pour l'Action, Hermes science publications</source>
          , pp.
          <fpage>29</fpage>
          -
          <lpage>46</lpage>
          ,
          <year>2007</year>
          .
        </mixed-citation>
      </ref>
      <ref id="ref10">
        <mixed-citation>
          10.
          <string-name>
            <surname>Lortal</surname>
            <given-names>G.</given-names>
          </string-name>
          ,
          <string-name>
            <surname>Lewkowicz</surname>
            <given-names>M.</given-names>
          </string-name>
          ,
          <string-name>
            <surname>Todirascu-Courtier</surname>
            <given-names>A.</given-names>
          </string-name>
          ,
          <year>2005</year>
          , Annotation: Textual Media for Cooperation,
          <source>in Proceedings of Annotation for Cooperation Workshop November 24-25th</source>
          (p.
          <fpage>41</fpage>
          -
          <lpage>50</lpage>
          )
        </mixed-citation>
      </ref>
      <ref id="ref11">
        <mixed-citation>
          11.
          <string-name>
            <surname>Maristella</surname>
            <given-names>Agosti</given-names>
          </string-name>
          , Giorgetta Bonfiglio-Dosio, and
          <string-name>
            <given-names>Nicola</given-names>
            <surname>Ferro</surname>
          </string-name>
          .
          <year>2007</year>
          .
          <article-title>A historical and contemporary study on annotations to derive key features for systems design</article-title>
          .
          <source>Int. J. Digit. Libr. 8</source>
          ,
          <issue>1</issue>
          (
          <year>October 2007</year>
          ),
          <fpage>1</fpage>
          -
          <lpage>19</lpage>
          . DOI=http://dx.doi.org/10.1007/s00799-007-0010-0
        </mixed-citation>
      </ref>
      <ref id="ref12">
        <mixed-citation>
          12.
          <string-name>
            <surname>Amir M. Naghsh</surname>
          </string-name>
          , Andy Dearden, and
          <string-name>
            <surname>Mehmet</surname>
            <given-names>B.</given-names>
          </string-name>
          <string-name>
            <surname>Özcan</surname>
          </string-name>
          .
          <year>2005</year>
          .
          <article-title>Investigating annotation in electronic paper-prototypes</article-title>
          .
          <source>In Proceedings of the 12th international conference on Interactive Systems: design, specification, and verification (DSVIS'05)</source>
          , Stephen W. Gilroy and
          <string-name>
            <given-names>Michael D.</given-names>
            <surname>Harrison</surname>
          </string-name>
          (Eds.).
        </mixed-citation>
      </ref>
      <ref id="ref13">
        <mixed-citation>
          13.
          <string-name>
            <surname>Silva</surname>
            ,
            <given-names>T. R.</given-names>
          </string-name>
          ,
          <string-name>
            <surname>Hak</surname>
            ,
            <given-names>J-L.</given-names>
          </string-name>
          ,
          <string-name>
            <surname>Winckler</surname>
            ,
            <given-names>M.</given-names>
          </string-name>
          <string-name>
            <surname>Nicolas</surname>
            ,
            <given-names>O.</given-names>
          </string-name>
          <article-title>A Comparative Study of Milestones for Featuring GUI Prototyping Tools</article-title>
          .
          <source>Journal of Software Engineering and Applications (JSEA)</source>
          , Vol.
          <volume>10</volume>
          No.
          <issue>6</issue>
          ,
          <string-name>
            <surname>June</surname>
            <given-names>23</given-names>
          </string-name>
          ,
          <year>2017</year>
          ,
          <string-name>
            <given-names>ISSN</given-names>
            <surname>Online</surname>
          </string-name>
          :
          <fpage>1945</fpage>
          -
          <lpage>3124</lpage>
          , ISSN Print:
          <fpage>1945</fpage>
          -
          <lpage>3116</lpage>
          , PP.
          <fpage>564</fpage>
          -
          <lpage>589</lpage>
          , DOI: 10.4236/jsea.
          <year>2017</year>
          .106031
        </mixed-citation>
      </ref>
      <ref id="ref14">
        <mixed-citation>
          14. W3C.
          <article-title>Web Annotation Data Model</article-title>
          . Available at: https://www.w3.org/TR/annotationmodel (April 30,
          <year>2019</year>
          , last visit).
        </mixed-citation>
      </ref>
      <ref id="ref15">
        <mixed-citation>
          15.
          <string-name>
            <surname>Jean-Luc</surname>
            <given-names>Hak</given-names>
          </string-name>
          , Marco Antonio Winckler, David Navarre.
          <article-title>PANDA: prototyping using annotation and decision analysis</article-title>
          .
          <source>8th ACM SIGCHI conference Engineering Interactive Computing Systems (EICS2016)</source>
          ,
          <source>Jun</source>
          <year>2016</year>
          , Brussels, Belgium.
          <source>EICS '16: Proceedings of the 8th ACM SIGCHI Symposium on Engineering Interactive Computing Systems</source>
          , pp.
          <fpage>171</fpage>
          -
          <lpage>176</lpage>
          ,
          <year>2016</year>
          . &lt;hal-01712526&gt;
        </mixed-citation>
      </ref>
      <ref id="ref16">
        <mixed-citation>
          16.
          <string-name>
            <surname>Winckler</surname>
            ,
            <given-names>M.</given-names>
          </string-name>
          ,
          <string-name>
            <surname>Palanque</surname>
            ,
            <given-names>P.</given-names>
          </string-name>
          ,
          <string-name>
            <surname>Farenc</surname>
            ,
            <given-names>C.</given-names>
          </string-name>
          ,
          <string-name>
            <surname>Pimenta</surname>
            ,
            <given-names>M.</given-names>
          </string-name>
          <article-title>Who does what with whom in Web Development?</article-title>
          2nd International Workshop on Task Models and
          <article-title>Diagrams for User Interface Design -</article-title>
          TAMODIA'
          <year>2003</year>
          , Heraklion, Greece,
          <year>June 2003</year>
          .
        </mixed-citation>
      </ref>
    </ref-list>
  </back>
</article>