HCI Architecture for Deaf Communities Cultural Inclusion
and Citizenship
Laura S. García, Cayley Guimarães, Diego R. Antunes and Sueli Fernandes
Informatics Department, UFPR, Curitiba, Brazil
Keywords: Human-Computer Interaction, Research Architecture, Deaf Culture, Social Inclusion, Computational Tools
for Citizenship.
Abstract: There is a lack of adequate support for the Deaf culture, and few researches to inform designers on how to
build computational tools to promote inclusion and citizenship for the Deaf. Deaf culture is an expression
applied to the social movement that holds deafness to be a difference in human experience – which includes
the right to use their natural language: Sign Language (SL) – rather than a disability. The present paper
describes an integrated approach through assumptions, methodological strategies and architecture which has
been proved so far to be adequate both computationally and socially.
1 INTRODUCTION
Deaf culture is an expression applied to the social
movement that holds deafness to be a difference in
human experience – which includes the right to use
their natural language: Sign Language (SL) – rather
than a disability. Available literature and
technological products show many limitations
towards promoting genuine inclusion and plain
citizenship of this population. The present paper
describes an integrated approach (assumptions,
methodological strategies and architecture) which
has been proved so far to be adequate both
computationally and socially.
The following sections show and overall of Deaf
issues and main problems encountered (2),
methodological issues (3), proposed architecture and
a few results (4) and conclusions and future work
(5).
2 DEAF ISSUES
Fernandes (2006, p.3) tells us the use of Sign
Language (SL) by the Deaf is a specific relationship
between her and her surrounding world; a different
way of being and learning. The traditional
oppression of the oral culture negatively affects the
development of the Deaf, who are treated as inferior
“[…] because, after all, they lack the essential
property for society, that is, the oral and hearing
language”. Oralism (the imposition, by society, of
the spoken language) colonized the Deaf by
forbidden them to use SL, the natural language for
the Deaf for intellectual development, information
acquisition, and knowledge creation. This systematic
exclusion from society has left in its wake dire
consequences (e.g. lack of intellectual development,
lack of SL knowledge, social exclusion, among
others). Boaventura de Souza Santos (2002) tells us
that science should promote inclusion. In that sense,
the Deaf communities’ specificities (i.e.
communication, language, way of understanding the
world) must be incorporated into tools to help the
Deaf in exercising citizenship.
The responsibilities to explore deeper into more
complex features of users (i.e. their culture) is a dear
notion to Human-Computer Interaction (HCI)
researchers in their task to inform design within
people’s lives. But there is a lack of research on how
these considerations should be incorporated in tools,
SL and Deaf issues in general. “[…] the need is
pressing […]”, especially when our work extends
into differing communities and constituencies other
than our own (Winchester, 2011, p.12). This section
presents a small picture of the Deaf issues, and some
(wrongful) attempts by the computer science field to
try to address such issues. Such inadequate attempts
prompted the proposed architecture, discussed on
Section 3.
126
S. García L., Guimarães C., R. Antunes D. and Fernandes S..
HCI Architecture for Deaf Communities Cultural Inclusion and Citizenship.
DOI: 10.5220/0004451201260133
In Proceedings of the 15th International Conference on Enterprise Information Systems (ICEIS-2013), pages 126-133
ISBN: 978-989-8565-61-7
Copyright
c
2013 SCITEPRESS (Science and Technology Publications, Lda.)
2.1 Daily Struggles in Deaf Lives
Guimarães et al., (2011b) conducted a series of
immersive research with Deaf students of an
undergraduate course of Linguistic/Libras (the
Brazilian Sign Language). There were 80 students,
35 men, 45 women, ages from 24 to 39 (average of
30 years of age). All of the students chose the
television as their source of information; they had no
real relationship with their families; they relied on
whatever little piece of information in SL they could
get; the teacher was their confident for personal
matter, among others. They all suffered prejudices,
lack of SL use and late acquisition of Libras:
Most of them were from non-deaf families that
imposed the oral culture. They were tossed from
school to school, and had no education in SL until
their late teens (at the age of 14, in average). The
following is a transcript of one of their statements:
“[…] my mother didn’t know I was Deaf until I
was two (2) years old. She forced me to oralize.
When I started school, they beat me in the
mouth: ‘You can’t sign’. I didn’t understand
anything. I got beaten, she pulled my ears, and I
was there, helpless […]”
Such situation was common among the Deaf
students, and show the problems faced by the child,
left to their own devices, suffering prejudices, not
developing proper social, affective and intellectual
skills; and not being exposed to an environment
permissive of natural language acquisition.
Additionally, it is no wonder that they
demonstrated poor Libras skills due, mostly, to
lack of Standard and dissemination. The lack of a
standard vector for the Deaf implies in several
inadequate skills in SL (e.g. errors, lack of
knowledge of grammar, several gestures, instead of
signs, among others). Tools in Libras (educational or
of another nature) to which the students had access
were rare, and proved to be detrimental (rather than
helpful) to the development of Libras. For example,
Capovilla et al., 2009 and Rybená, 2011 are
dictionaries, but only present a one-to-one mapping
of the oral language to the SL (thus, making it of
little use to the Deaf who doesn’t know the oral
language.
2.2 Lack of Natural Language
Acquisition
For Chomsky (1986), the ability to understand,
create and transform culture is a human trait that is
language-dependent. Kyle (2005) tells that the gaps
between the Deaf and her family, due to the lack of
communication, are the cause of high levels of
mental diseases later in the Deaf’s life, in direct
relation to life and survival of the Deaf.
Brito, (1993) tells that without SL acquisition,
the Deaf has a diminished ability to perform tasks
for the development of intelligent action: the Deaf
does not learn how to plan and how to overcome
impulsive action; the Deaf does not become
independent of the visual, concrete situation and the
Deaf has difficulties to control herself and to
socialize. Consequently, members of the Deaf
community are more likely to suffer from: the lack
of meaning and knowledge creation; the lack of
identity and cultural diversity; the lack of intellectual
development, among others, says Finau, (2006). All
of these predicaments are dire consequences caused
by the lack of affective ties of the human being with
language. Due to this language barrier, the non-Deaf
parents encounter difficulties to teach their Deaf
children even the basics, such as personal hygiene
etc.
2.3 Literacy and Sign Language
For Sánchez, (1991), language is more than a way of
communication, and it includes a regulation function
of thought, according to Vigotsky, (1974).
Bilingualism, considered to be more adequate for
Deaf education, is the movement that claims the use
of, at least, two languages: SL, as a first language,
and a second language in its written form – in our
case, Portuguese, the oral language of Brazil.
For Lévy, (1999, p.10), society must go beyond
the mere use of computers for games and leisure,
thus limiting its use. This is a clear call for an
innovative, intellectual, interactive use of
technologies: to make sense of the world, the child
must be able to construct her own mental models of
interactions, and, for such, e-learning tools (that
scaffold this process) are needed: “new intellectual
technologies” that are to be used for Literacy.
Literacy is the resulting process of social
practices of the use of the written form of the oral
language as a symbolic system and as a technology,
in specific contexts, for specific goals (in our
context, to be acquired by the Deaf by a functional
use of the language, where the language assumes a
character of real meaning). “Therefore, Literacy as
effective appropriation is pleasurable, is leisure, is
access to information, is communication, is a way to
exercise citizenship in different social practices”
(Fernandes 2012, p.131).
Since Stokoe, (2000) we have that SL are
HCIArchitectureforDeafCommunitiesCulturalInclusionandCitizenship
127
considered a legitimate, complete linguistic system,
of gestural spatial-visual manner. Languages are
social constructs that reflect in the identity of a
certain group, and serve many functions and
purposes, other than communication (e.g. modelling
the world). SL are fully conventionalized, with rules
and structure capable of providing the Deaf with an
adequate means to realize all of their linguistic
potentialities (Fernandes, 2012). They are not
universal (i.e. each country has its own SL). They
are not mere mimics. They are not gestures. They
are independent from the oral language.
Guimarães et al., (2013) surveyed some related
work on existing tools for literacy: existing
technologies are inadequate to the Deaf specificities
(i.e. they are not in SL); they lack usability for the
target audience; they do not allow for multiple and
full collaboration; they are not designed for Literacy
as per the needs of the Deaf. The surveyed works
present a lexical approach, without context; simple
one-to-one mapping of Portuguese to SL; use of
only the hand configuration of signs (out of at least
five parameters); the use of restrictive technology
(e.g. sensors, gloves etc.). Above all, they do not
present stakeholders with a process for
computational development of tools and Artifacts,
the focus of this paper.
2.4 Natural Language Processing
Most of the related works in Natural Language
Processing (NLP), such as Dasgupta et al., (2008)
and Marshall and Sáfár (2003) fail to build tools in
SL. Huenerfauth, (2004) proposes a translator that
goes from written text (of the oral language) to
American Sign Language (ASL) animation, but does
so by adding complexity creating an interlingua with
which to work. El Ghoul and Jemin, (2009) propose
a screen reader, from text to SL animation based on
avatars; but, as per the publication, their proposal is
highly device-dependent, they did not incorporate
SL and Deaf users in their work, and it is a one-to-
one mapping (i.e. if the word from the text doesn’t
have a correspondent sign in the system, the system
will fail). Such limitations also appear in Buttussi et
al., (2007). San-Segundo et al., (2008) incorporate
gestures into semantic analysis. But SL are not mere
gestures.
Fernades, (2012) proposes some approaches that
are paramount to the proposed architecture: center
on teaching of SL; the oral language is to be seen
and presented in its written form as a set of visual
elements; the written system should be used for
literacy, with the word as the basic unit (akin to the
teaching of Chinese ideograms). Therefore, the word
is the minimal element: the oral language for the
Deaf is what they see and there is an ideographic
relation between the sign and its representation.
2.5 Image Processing/Computer Vision
Antunes et al., (2011) compiled dozens of work
from the literature of Image Processing (IP) and
Computer Vision (CV) that claimed to be directed at
the Deaf issues. The reviewed studies failed to
supply real end-use technology. Most studies had
some (or combination of some) sort of approach
inadequacy: for example, some of them use a
restricted set of data (i.e. static images; videos that
represent only the alphabet; or only a small set of
hand shapes only; or just isolated signs, without a
context; or randomly selected signs, among others)
that render them inadequate for generalization, for
instance. Other approaches use specific,
sophisticated devices (e.g. gloves, sensor,
accelerometers etc.) that exclude the context of
natural signalling by the Deaf.
The authors were able to identify four (4) main
categories of problems:
Inadequate Framing of the Research Object:
use of restrictive devices; non-natural
environments that are difficult to replicate
outside lab conditions; the Deaf are not a part of
the design process; inadequate use of SL (e.g.
one-to-one mapping from the oral language,
small sets, gestures, mere alphabets).
Inadequate Methodological Approaches: the
studies use two models: a) whole word (isolated
signs described in the oral language – one-to-
one, no use of the parameters of SL, limited
results) and b) the phoneme-base approach as
sub-units, but with no computational model.
Inadequate Treatment of the SL: disregard for
essential elements of SL (e.g. temporal aspects,
quality); data sets limited to static images;
limited use of the phonological parameters (e.g.
only a few hand shapes or the alphabet, small
sets, isolated/random data sets, use of just one
hand, no non-manual expressions).
Inadequate Use of Technologies: Two main
categories: a) use of devices (e.g. glove, sensors):
reduce movement, movements are not natural,
limited identification of signs, high costs,
availability, technical problems) and b) direct
measurement: technical problems.
ICEIS2013-15thInternationalConferenceonEnterpriseInformationSystems
128
2.6 Image Synthesis (3D Avatar)
Most of the avatar 3D analysed were mere one-to-
one reproduction of a lexical element from the oral
language. The avatars are capable of signing only
the set of signs that are pre-programed in their
database. Most of the avatars lack the SL natural
qualities, such as smooth movements, transitions
from one sign to the other, speed (the signs are
chopped, too fast or too slow, thus making it
difficult for the Deaf to understand them).
A correct approach in this process has been
found in (Lombardo et al., 2011). Their paper
describes an avatar synthesis process embodied in a
large research and industrial consortium project,
Atlas, which determines a completely different
working conditions form our own effort. Never-the-
less, we can see that their work adopt the mark-up
language strategy to decompose and represent the
Italian Sign Language (LIS), often found in the few
correct pieces of work available in literature.
2.7 Writing System (Signwriting)
Most of the editors for the writing system are in the
oral language. They require an extensive work from
the user to compose a written sign. They lack
usability: the SignWriting primitives are hidden in
several layers of menu choices, overloading the
working memory of the user. The positioning of the
primitives to form the sign presents several
difficulties: precision, speed, and rotation among
others. Some editors are a one-to-one mapping from
the oral language: you type in the word, and it brings
from the data base the completed written form of the
sign – highly dependent on the knowledge of the
oral language, and whether the word is on the data
base.
3 METHODOLOGICAL ISSUES
The genesis of our project can be described as
follows. Troubled by the fact that, even if our work
was geared towards producing for public education
and health, we still produced for the majority of
people in Brazil. This research arena was, therefore,
excluding of many, which made us turn our research
focus to minorities. Our first hypotheses about
technology for Deaf was that they did not demand
critical changes regarded to non Deaf people; we
considered the visual nature of most information in
technological artifacts and the presence of non visual
information in Portuguese language written form.
After reading related literature based on the “clinical
view of deaf” (which sees deafness as an illness) for
almost a year, we stayed within this first hypothesis.
Our work with an expert with large experience
with the Deaf Culture and and deaf communities (a
colleague who later became an integral member of
our research group), taught us that very few
Brazilian Deaf in fact could read and write in
Portuguese. The reality was even worse than we had
expected. Due to decades of social, political and
educational exclusion, which happened throughout
the World and also in Brazil up to the last decade,
and even considering that Libras, the Brazilian Sign
Language was, by law, decreed to be an official
language in Brazil, neither Libras nor any “second
language” was granted at school. Additionally to the
lack of their natural language, which acts as the
mediation language for any second language
acquisition (in our case Portuguese), the Deaf even
suffered from the illness of having to “learn”
Portuguese by the oral method. The situation is so
severe now that literate and even under-graduating
Deaf students cannot write Portuguese following the
grammar rules, since their natural language
articulates just meaning-full signs.
Secondly, we reached the conscience of the need
to work in a genuine multi and interdisciplinary
manner, this meaning having as normal interlocutors
(from the beginning and during the whole process)
experts in Computer Science, Education and
Linguistics. Additionally, our posture made
Computer Science other subareas from HCI join our
view and effort, extending the initial research group
to subareas like Algorithms and Formal Structures
and Robotics. It is worth mentioning that this
approach is not, in fact, a frequent practice so far as
Computer Science researchers available work refers.
Finally, our work is characterized also by the
join consideration of theoretical and practical
knowledge (in this case school practices) and by the
crucial strategy of involving the deaf community all
over the Project: at beginning, to define correct
hypotheses, during design to establish their real
principal needs and during developing to test
whether our artifacts were appropriate for them. This
partnership is to be maintained indefinitely, as it will
allow our project to follow the technology cycle
(real needs, product developing, limitations and
need, new product developing, and so on).
4 PROPOSED ARCHITECTURE
The proposed approach can be described through the
HCIArchitectureforDeafCommunitiesCulturalInclusionandCitizenship
129
architecture properly considered and by a few pair
approved internationally published results.
4.1 The Proper Architecture
The proposed architecture can be described trough
two main elements: the abstract four general layers
hierarchy (with special relevance of the basic
interface structure built) and the integration model,
which shows all the modules and their identified
interrelations that give support for the execution of
the abstract architecture.
Figure 1 presents an abstract of the proposed
HCI Architecture, in its four levels.
Figure 1: Abstraction of the proposed HCI Architecture to
promote inclusion and citizenship for Deaf communities.
The general user environment, as the research
emerged initially, is to support Portuguese as well as
Libras, the Brazilian Sign Language. In this way,
user interfaces must allow user-user dialogues either
in Portuguese or in Libras and also mixed ones
either between Deaf and non Deaf persons or
interactions between Deaf persons and system. In
these last cases, interpreters in real time should
mediate the interaction. This proposal demanded, of
course, a mandatory user profile for the Libras
interpreter. The middle layer is to allow for Deaf
communities’ accessibility to information, to
communication and to knowledge creation, through
the prism of an inclusive perspective.
The surface layer is responsible for providing
adequate applications, mainly in the axes of i) giving
adequate support to Deaf natural language (i.e. Sign
Language) acquisition and registration (as occurs
traditionally in the written codes of any oral
language); ii) supporting teaching-learning processes
of Libras itself and of written Portuguese as their
second language (i.e. literacy); and, what is self-
explaining of the situation of real severity level, iii)
supporting knowledge acquisition of every other
area, considered the hypothesis of the Sign
Language acting as the mother language for the Deaf
in any interpretation process.
The services needed referred specially to those
associated to language itself, starting by dictionaries,
thesaurus and translators. Though being themselves
applications, they can be seen and even more they
are critic as tools to allow for plain applications for
Deaf communities.
The interface between the services and the
internal APIs has as its principal function providing
correct frameworks for both tools layers being built
scientifically sound. Rescuing oral natural languages
analytical hierarchy for the traditionally so called
“natural languages” processing when Sign
Languages were still not recognized in this way, we
could quote the phonological level (surface level:
phonemes), the lexical one (words), the syntactic
level (grammar rules), the semantic level
(underlying meaning), and, last but not least, the
pragmatic level (language in context).
As for Sign Languages, the surface level is the
signal production one, the first structure of our
approach applied for the Libras phonological level:
the level that describes sign production
(articulation). Though not being of an audio nature,
the sign’s elementary components level of the Sign
Languages is called phonological by analogy to oral
languages. The choice of this structure as the basic
one in the whole architecture has proved productive
after four or five years’ work, as the following
section shows. Our hypothesis for this to happen is
that it captures the correct approach of Natural
Language Processing, together with the correct
understanding of the Libras itself, which, for
communication purposes, takes advantage of the
Portuguese Language alphabet mapping signs only
to denote words with no corresponding sign in
Libras and to refer to Brazilian proper names.
Finally, the internal level is responsible for the
Computer Science subareas knowledge and
technology necessary for the several tools and
applications.
After presenting the overall architecture, we
proceed to describe the integrating model. This
representation makes explicit all the modules and
bases involved, and their necessary interrelations.
ICEIS2013-15thInternationalConferenceonEnterpriseInformationSystems
130
Figure 2: Proposed HCI Architecture with developed and in progress modules and bases.
Figure 2 shows it, together with the modules
instantiated (developed/in progress) up to the present
moment.
Libras processing, considered as Natural
Language Processing, requires several modules
shown in a general way in Figure 3. The precise
logic of the modules execution is not yet defined.
Figure 3: Libras processing steps (in a general way).
In order to complete the overall description of
our work, we show the applications already
addressed (Figure 4). It is pertinent to note that these
applications priority was determined by the deaf
community always present (both as a design actor
and as potential user) during our research.
Figure 4: External applications (take advantage of internal
and tools architecture).
4.2 Results so Far
As the foundations stone of the architecture, the
Libras Phonological Model was first designed and
developed. The Model describes the sign production
components by a hierarchical structure whose most
abstract levels are shown in Figure 5. This model
was described in some international events
published papers, namely: Antunes et al., (2011),
Guimarães et al., (2010a,b), Guimarães et al.,
(2011a, b, c), Guimarães et al., (2012a,b),
Guimarães et al., (2013), Trindade et al., (2012).
Figure 5: Abstraction of the proposed HCI Architecture to
inform design of effective otols for Deaf inclusion.
As previously shown in Figure 2, several other
component modules and bases had been addressed
so far.
Regarding the internal and tools levels, we can
quote the Libras dictionary and grammar and the
Libras model - this one by (Gonçalves, 2013). All
this work is in progress and has not been published.
Still within the main architecture and as far as
the addressed communities are concerned, we can
mention the SignWriting generator, the SignWriting
interpreter – themselves complementar compilers,
the Libras manager - who aims to allow for Libras
dictionary queries by written Libras signs, the
SignWriting editor – that focuses on usability and
communicability issues in order to facilitate task
completion by deaf and non deaf.
HCIArchitectureforDeafCommunitiesCulturalInclusionandCitizenship
131
Considered as the most relevant applications by
the deaf community itself, the three applications
built have been already approved by peer review
when presented at international events.
The interactive artifact to support bilingual
culture (Libras and written Portuguese) is an
innovator conceptual environment which, in fact,
supports the teaching-learning binomial in virtually
any knowledge discipline. Guimarães et al., (2012;
2013). It is a doctoral theses work to be presented
soon.
The Interactive artifact to support written
Portuguese acquisition by deaf children has been
designed having as a premise the literacy concept
and as its methodological strategy a totally
interdisciplinary work (Computer Science,
Education – theories and practices, and Linguistics).
It is also a theses work, to be presented within a year
effort. The results reached so far have not been
published.
Last but not least, the Collaborative artifact for
knowledge building was designed having in mind
the critical deaf necessity of the Libras interpreter in
order to grant not only communication processes but
also knowledge building ones.
5 CONCLUSIONS AND FUTURE
WORK
Available theoretical results and technological
products for deaf communities have several
limitations, as far as our studies have revealed. The
present paper described a HCI architecture – or,
better, an approach that aims at treating deaf
communities in an adequate manner, regarding the
computing field interests. It implies, in the authors
view, differential premises and strategies
synthesized as follows: 1) deaf considered as
culturally bilingual citizens (in opposition to not
normal clinical ones); 2) multi and interdisciplinary
work both with external areas (Education and
Linguistics) and with internal ones (in our case, so
far, Algorithms and Formal Structures and
Robotics); 3) join consideration of theoretical and
practical knowledge (School practices); and 4)
involving the deaf community all over the Project: at
beginning, to define correct hypotheses, during
design to establish their real needs and during
developing to test whether our artifacts were
appropriate for them.
The differential factors just mentioned lead
naturally to our Libras Computational Phonological
Model and, from it, to all the bases and modules
themselves now in progress. The present paper
showed the architecture that integrates them, as well
as described briefly (since they are unpublished
results) the group main partial results.
At close future work, the group will give priority
to the Libras dictionary, together with the Natural
Language Processing tools, necessary to the
Translator, which is another critically desired
product.
The authors thank Fundação Araucária and
CAPES for the financial support.
REFERENCES
Antunes, D.R. et al., 2011. A framework to support
development of Sign Language Human-Computer
Interaction: Building Tools for Effective Information
Access and Inclusion of the Deaf. V IEEE RCIS’11.
Guadaloupe. P. 126—137.
Brito, L.F., 1993. Social integration and Deaf education.
Rio de Janeiro: Babel.
Buttussi, et al., 2007. Using Web3D technologies for
visualization and search of signs in an international
sign language dictionary. In: Proceedings of the 12
th
I.
Conf. on 3D Web tech. Perugia, Italy. 61—70.
Capovilla, F.C., Raphael, W.D., Maurício, A.C.L., 2009.
Novo Deit-Libras: Dicionário Enciclopédico Ilustrado
Trilíngue. São Paulo: Edusp.
Chomsky, N., 1986. Knowledge of language: its nature,
origin and use. New York: Praeger Publishers.
Dasgupta et al., 2008. A multilingual multimedia Indian
Sign Language dictionary tool. 6
th
workshop on Asian
Language Resources. Indian School of Business.
Hyderabad, 57—64.
Fernandes, S., 2006. Avaliação em Língua Portuguesa
para Alunos Surdos. SEED/SUED/DEE, CURITIBA.
Fernandes, S. 2012. Educação dos Surdos. Curitiba:
IBPEX.
Finau, R., 2006. Possible encounters: Deaf culture,
education and linguistics. In: R. Quadros (Org.) Deaf
Studies I. Petrópolis: Arara Azul.
el Ghoul, O., Jemni, M. 2009. A sign language screen
reader for Deaf. USAB’09. A. Hozinger, K.
Miesenberger (Eds.). Springer, Berlin Heideber,
476—483.
Gonçalves, D. A., 2013. Avatar 3d para Síntese
Automática de Sinais da Língua de Sinais Brasileira.
M.Sc. Dissertation. Informatics Dept. UFPR.
Guimarães et al., 2010a. Structure of the Brazilian Sign
Language (Libras) for computational tools: citizenship
and social inclusion. WSKS’2010, v.111, p. 365—370.
Guimarães et al., 2010b. Technological artifacts for social
inclusion: structure of the Brazilian Sign Language
(Libras), Gestures for Citizenship. IADIS
WWW/Internet, v.1, p.267—271.
ICEIS2013-15thInternationalConferenceonEnterpriseInformationSystems
132
Guimarães et al., 2011a. Evaluation of a computational
description model of Libras (Brazilian sign Language).
Guadaloupe: V IEEE RCIS’11. P.485—494.
Guimarães et al., 2011b. Collaborative Consensus and
Knowledge Creation: Computer-Mediated
Methodology for Sign Language Studies. Mykonos:
WSKS’11. P. 278—292.
Guimarães et al., 2011c. Empowering collaboration
among the Deaf: Internet-based knowledge creation
system. IADIS WWW/Internet, p. 137—144.
Guimarães et al., 2012a. Conceptual meta-environment for
Deaf children Literacy challenge: How to design
effective Artifacts for bilingualism. VI IEEE RCIS.
CFP1240D-ART, v.6, p.1—12.
Guimarães et al., 2012b. Deaf Literacy: A computational
process to design Sign Language/Portuguese Artifacts
for Internet. IADIS WWW/Internet, Madrid, p. 250—
259.
Guimarães et al., 2013. Pedagogical Architecture –
Internet Artifacts for Bilingualism of the Deaf (Sign
Language Portuguese). 46th HICSS. IEEE CPS, v. 46.
40—49.
Huenerfauth, M., 2004. A multi-path architecture for
machine translation of English text into ASL
animation. ACL, Morristown, 25—30.
Kyle, J.G., 2005. Beginning Bilingualism. Ibero-American
Congress on Bilingual Education. Lisbon.
Lévy, P., 1999. Cyberculture. Paris: Odile Jacob.
Lombardo, V., Battaglino, C., Damiano, R., Nunnari, F.,
"An Avatar-based Interface for the Italian Sign
Language", International Conference on Complex,
Intelligent and Software Intensive Systems (CISIS),
pp.589, 594, 2011.
Marshall, I., Sáfár, E., 2003. A prototype text to BSL
translation system. ACL’03, v.2, 113-116.
Rybená, 2011. Available at http://rybena.com.br accessed
in 05/05/11.
Sánchez, C., 1991. La educación de los sordos en um
modelo bilíngue. Mérida: Diakonia.
Santos, Boaventura de Souza, 2002. Um discurso sobre as
ciências. Porto: Apontamentos.
San-Segundo et al., 2008. Proposing a speech to gesture
translation architecture for Spanish Deaf people.
Journal of Vis. Lang. Comput., 19, 5. 523—538.
Stokoe, W.C., 2000. Models, signs and universal rules.
SLS, v.1, n.1, p. 10—16.
Trindade et al., 2011. Communication and cooperation: an
analysis of a community of practice by non-Deaf and
Deaf to study sign language. WSKS’2011. P. 191—
205.
Trindade et al., 2012. Challenges of knowledge
management and creation in communities of practice
organizations of Deaf and non-Deaf members:
requirements for a Web plataform. Behaviour &
Information Technology, v,31, p.1—12.
Vigotsky, L.S., 1974. Mind in society: the development of
higher psychological processes. MA: Harvard
University Press.
Winchester, W.W., 2011. III REALizing our messy future:
Towards culturally responsive design tools in
engaging our deeper dives. ACM Interactions, V.XVII
(6). New York: MIT Press. 12—19.
HCIArchitectureforDeafCommunitiesCulturalInclusionandCitizenship
133
