Observational Research Social Network
Interaction and Security
Rui Pedro Lopes
1,2
and Cristina Mesquita
1
1
Polytechnic Institute of Braganc¸a, Campus St. Apol
´
onia, Braganc¸a, Portugal
2
IEETA, University of Aveiro, Aveiro, Portugal
Keywords:
Preschool Education, Observational Research, Social Networks.
Abstract:
Quality in education depend heavily on the teachers’ professional development, as a mean for pedagogical
methodologies and practice improvement. In this sense, learning is enhanced by sharing and working in a
community of practice, a learning organization that generate knowledge and allow members to innovate. In
this context, children observation is fundamental, allowing a sound basis for reflection and action around
learning experiences and teaching environments. Specific guidelines and programs, such as the EEL/DQP, can
help reducing the inherent subjectivity, providing a common base for teachers and the preschool education
community. In this paper we propose an online, web based community to improve the observation process as
well as the communication between researchers. The social relations are identified and the security issues are
discussed.
1 INTRODUCTION
The literature about teachers’ professional develop-
ment shows that communities of practice constitute
centers for professional growth (Sheridan et al., 2009;
Evans et al., 2006). In this sense, communities of
practice are groups of people who gather from com-
mon professional interests and a desire to improve
their practice, sharing their knowledge, ideas and ob-
servations. A community of practice can be under-
stood as a social system for learning because, like
other social systems, provide: i) a structure where
they develop complex and dynamic relationships; ii)
ongoing negotiations between members; iii) shared
meanings and cultural identity (Wenger, 2010).
In this perspective, communities of practice can
constitute itself as an opportunity for learning about
what really matters, about epistemology, ontology
and methodology that can sustain the praxis. Com-
munities of practice are conceived as learning organi-
zations which investigate their situation and their rela-
tionships and generate praxeological knowledge that
allows teachers to innovate.
The Effective Early Learning (EEL) Project
(Bertram and Pascal, 2004), known in Portugal under
the designation Desenvolvendo Qualidade em Parce-
ria DQP (Bertram and Pascal, 2009) propose the
creation of communities that develop a collaborative
action, focused on shared purposes and in a sustained
process of pedagogical mediation by an external su-
pervisor. This intends to improve the quality of early
childhood education contexts, through an active in-
volvement of participants. It is a monitoring process
that uses several observations tools requiring that pro-
fessionals build knowledge and skills about the un-
derlying processes that allow them to reflect with the
peers about their action.
Observations in the EEL/DQP are used to assess
children physical, emotional, social, and intellectual
development, focusing on specific areas, such as so-
cial interaction, learning experiences, space manage-
ment and creations, and others. They can also help to
better understand how different areas of development
are interrelated, as well as helping recognizing what
behaviors are typical of various age groups. In turn,
this understanding will help the teacher to improve as
a person and as a professional.
Each observation process depends heavily on the
sensibility and experience of the teacher. It is very dif-
ficult to get consistent results if the observers diverge
in the way they interpret the setting. Although natural
and inherent to the process, it is important to mini-
mize this subjectivity. In order to do so, the observa-
tions are performed simultaneously by more than one
teacher in a democratic way. The external supervisor
also has a thorough perspective on the whole project
57
Pedro Lopes R. and Mesquita C..
Observational Research Social Network - Interaction and Security.
DOI: 10.5220/0004837800570064
In Proceedings of the 6th International Conference on Computer Supported Education (CSEDU-2014), pages 57-64
ISBN: 978-989-758-022-2
Copyright
c
2014 SCITEPRESS (Science and Technology Publications, Lda.)
and may participate in some or all observations. The
bottom line is that to be able to get meaningful in-
formation, the communication between all the teach-
ers and researchers is fundamental, as a way to foster
reasoning and action building.
In this paper we propose the design and develop-
ment of an online, web-based, regulated community,
providing:
A central and secure repository of observation
data, results and annotations;
A computer mediated social network.
This service allows each teacher to manage the
observation process, sharing information with autho-
rized colleagues and improving their experience and
knowledge.
2 EVALUATION PROCESS IN
EEL/DQP
The evaluation of quality of early learning in the
scope of EEL/DQP requires obtaining a considerable
amount of data through several techniques, includ-
ing detailed observations of children and adults, inter-
viewing parents, practitioners and children, documen-
tary analysis and others. This complex and somewhat
subjective process requires well-trained teachers and
researchers. In particular, the EEL/DQP initiative de-
fines a four phase/thirteen steps procedure (Figure 1).
Data is gathered and systematically organized in
research portfolios, that will be used in a cyclic pro-
cess of thinking-do-thinking to research and create
change (Mesquita-Pires, 2012). This process is en-
hanced by the utilization of observation techniques
which measure the effectiveness of the learning and
teaching processes, such as the Child Tracking Ob-
servation Schedule, to gain a snapshot of the childs
day and providing information of learning experi-
ences (Bertram and Pascal, 2006), the Child Involve-
ment Scale, an observation technique which mea-
sures the level of a childs involvement in an activity,
the Adult Engagement Scale, to evaluate the interac-
tion between the practitioner and the child (Laevers,
1994).
The application of the procedure has a broad set
of difficulties and challenges. Initially, it is neces-
sary that the research group learn about participatory
pedagogies as theoretical foundations about EEL re-
search techniques and the practicalities of their use.
The application of interviews come soon after, which
lead the participants to reflect on the ethical issues in-
volving its use. Learning to observe is another chal-
lenge, because the signs are not always evident and
Evaluation
Initial Preparation
Initial Data Gathering
Interviews
Child Tracking Observation Schedule
Child Involvement Scale
Adult Engagement Scale
Development of Evaluation Report
Action Planing
Development of the Action Plan
Development
Document and implementation of the action plan
Child Tracking Observation Schedule
Child Involvement Scale
Adult Engagement Scale
Reflection
Reflection on the impact of the plan in the future
Final Report
Figure 1: EEL/DQP phases and steps.
observers must be well trained to identify them. Be-
sides, the observation process should be systematic
requiring that kindergarten teachers find time in their
daily routine to observe the children and understand
how they are learning.
2.1 EEL/DQP Procedure
The EEL/DQP overall procedure follows the four
phases described above. It starts by an initial orien-
tation of the work to be performed, in which all the
process is prepared and all the participants informed
in detail. Initial data gathering follows, where the in-
stitution is characterized, including the interior and
exterior spaces, its education philosophy, the differ-
ent learning activities, and others.
The third step includes performing interviews to
the dean, staff (about 50 %), children (20%), and par-
ents (20%). It is very important that all the stake-
holders are well informed and understand the process.
The teacher records the interviews and take notes of
key phrases, to support the written report. In the end,
access to interview transcripts must be given to par-
ticipants.
CSEDU2014-6thInternationalConferenceonComputerSupportedEducation
58
The fourth step requires an observation process,
using the Child Tracking Observation Schedule, with
the main purpose of understanding the child’s daily
routine. This technique gives information about the
learning experiences, the level of choice, his involve-
ment, the group organization and interaction with
adults.
The Child Involvement Scale seeks to measure
not only the learning outcomes but also the under-
lying processes. Essentially, it gathers information
about the participation in activities and projects, thus
giving indicators of concentration and motivation as
well as of satisfaction (Laevers, 2005). When lack-
ing, chances are that the children development will
stagnate, and all the actors in the education process
should do everything in order to create an environ-
ment in which children can engage in a wide variety
of activities. The details are registered in a specific
form.
The Adult Engagement Scale evaluates the inter-
action between the practitioner and the child (Laev-
ers, 1994). It targets the effectiveness of the teaching-
learning process through observation of adult-child
interaction. The quality of the adult’s intervention is a
critical factor for the child’s knowledge building. Up
to a maximum of 5 adults should be observed, paying
special attention to the sensibility, stimulation and au-
tonomy categories.
2.2 Ethical Behavior
Ethical behavior is fundamental in the whole process,
since it tackles professional behavior, privacy, and
confidentiality concerns (always keep in mind that the
observer is representing his school as well as himself).
Children, parents and professionals should be treated
with courtesy, always respecting the privacy rights of
children and family.
All the participants in the project, including chil-
dren, should be informed about all the details of the
project and their role in the whole process, giving
their consent. This will ensure that all of them are
comfortable and willingly, contributing to better re-
sults.
During observations, teachers may gather sensi-
tive information, such as details about child’s devel-
opment and behavior, as well as videos or pictures.
Children and their parents must know that this data is
restricted and will not be used in other contexts. Even
with adequate permission to observe and record these
details, the information must be stored appropriately,
to avoid misuses and eavesdropping.
2.3 Community Use Case
The procedure described in the previous sections is
traditionally paper-based, requiring a lot of written
material. In a typical set of 24 children, of which only
50% are observed, as much as 48 to 72 pages of forms
are filled. In a kindergarten, this procedure is repeated
in all the rooms, totaling more than 200 pages of gath-
ered data. Moreover, all the visual and audio details
are lost.
The subjectivity inherent to this process also re-
quires that all the observers, usually the kindergarten
teachers, receive an uniform training, allowing them
to achieve similar interpretations of similar situations.
This is only possible if the communication between
them is open and regular, requiring several in person
meetings.
Moving this information to an online service, such
as a social network, will allow the observers to store
and organize all the observation data in a single pro-
file, combining video, audio and text data in the same
observation procedure. This also makes it possible
to communicate asynchronously with other observers,
providing a valuable tool for subjectivity reduction
and better overall learning.
3 THE TEOBS SOCIAL
NETWORK
The interactions between observers (kindergarten
teachers) is fundamental to provide a stimulating en-
vironment for reflection and discussion, essential to
ensure low levels of subjectivity and to improve ob-
servers’ skills. The EEL/DQP process expects sev-
eral, face-to-face, meetings to discuss about the data
gathered in all of its steps.
The social relationships established between the
participants in this process (friendship, co-working,
information exchange, . . . ) can be mediated by com-
puters through the TeObs social network. Computer-
mediated communication (CMC) gives the possibil-
ities for asynchronous exchange of information, re-
gardless of where participants are. The community is
no longer defined as a physical place, but as a set of
relationships where people interact socially for mu-
tual benefit (Garton et al., 2006). However, this does
not preclude face-to-face meetings, should the com-
munity decide accordingly.
MySpace, Facebook or Twitter are remarkable ex-
amples of social networks, connecting millions of
people around the world. TeObs intends to be an audi-
ence specific social network, connecting kindergarten
teachers and providing a constantly updated memory
ObservationalResearchSocialNetwork-InteractionandSecurity
59
of experiences and previous knowledge. This can re-
sult in connections between individuals that would
not be made otherwise and that can prove valuable
for the overall process as well as for each participant
training (Boyd and Ellison, 2007).
3.1 Social Interactions
TeObs is a web-based service that allow individuals
to construct a public or semi-public profile within a
community. Connections among participants are ar-
ticulated through a list of other users, keeping in mind
the ethical behavior and security concerns. Under this
restrictions, each user is allowed to traverse their list
of connections and those made by others within the
system.
Each user can be in two broad status:
Idle: the user is not currently participating in an
observation project, having access to specific con-
nections and to previous notes and contents;
Active: the user is participating in an observation
project, with mandatory connections to peers and
to the external advisor (the supervisor).
Each role the user can have is associated to his
experience level (Figure 2). The bottom layer is
populated with members unable to perform observa-
tions without undergoing a training process dot-
ted circles. After acquiring basic skills, the member
gets the observer status, able to perform observations
autonomously (as long as following the predefined
guidelines).
Observers
Local supervisor
Supervisor
Figure 2: Community.
With increasing experience, the status may be el-
evated to experienced observer and, later on, to local
supervisor. Finally, with complete domain of the ob-
servation process and with several successful projects,
the teacher can become a full supervisor (Table 1).
An observation process starts with a supervisor
creating a community and adding a number of kinder-
garten teachers. Usually, the community corresponds
Table 1: Observer level.
4 Supervisor
3 Local Supervisor
2 Experienced
1 Basic training
0 No training
to an institution, although it is not mandatory. Next,
the supervisor establishes the relationship between
the community elements, granting or removing access
to the data collected by each observer (Figure 3).
S
A
B
C
Figure 3: Social relations in an observation process.
Each observer, identified by a circle, collects and
organizes information, stored in his personal area (the
large rectangle). The observation data is structured as
a tree, enclosed in a round rectangle. Each observer
has full access to his area and all the data elements
within. To reduce subjectivity, it is important that the
data is shared with a peer and with a supervisor, to
ensure similar criteria among observers. Sharing is
configured by the supervisor, granting access to some
information. In figure 3, the supervisor ‘S’ has access
to the current procedure (gray round rectangles), the
observer A has access to the data collected by ‘B’
and vice-versa.
Each observer can also have data from previous
procedures (white round rectangles). This is private
information and is only shared with the supervisor
and the observers of the same community at that time
– ‘B’ and ‘S’ may not have access to it.
When starting a new community, the supervisor
can invite a non-trained observer (dotted shapes in the
figure). He is not able to contribute with observation
CSEDU2014-6thInternationalConferenceonComputerSupportedEducation
60
data without acquiring basic skills so, the first step is
to go through a training process. Other than “brick-
and-mortar” training, observer ‘C’ may have access
to specific data from other observations, granted by
the supervisor, as well as other materials. He must
be familiar with the team members, their perspectives
and pedagogical practices so that among peers could
be developed positive interactions.
The data, the social relations as well as the ex-
perience level is instantiated in a web-based, online
application, which we call TeObs.
4 STRUCTURE AND
IMPLEMENTATION DETAILS
TeObs is built in Java Enterprise Edition, following
an multitier architecture structured in three functional
layers (Figure 4):
Data Model – responsible for data persistence and
structure;
Business Layer library of actions available to
upper layers and external applications, such as
web browser, smartphone or desktop user inter-
face;
Web Layer – Web browser interface with the user.
Data Access
JDBCJDBC
BusinessEJB EJB EJB
Web Services JSF JSF JSF
iOS
Android
Web Browser
JSON
HTML
Figure 4: Multitier architecture.
Each layer is self contained, allowing the encap-
sulation of functionality and the distribution of re-
sources to better cope with peak usage patterns or the
increase of the number of users. This approach al-
lows to build large-scale, scalable, reliable, and secure
applications as well as simplifying the development
of complex, distributed applications. The existence
of a representational state transfer (REST) communi-
cations API allows integrating external applications,
running in smartphones, for example (Mesquita-Pires
and Lopes, 2014).
4.1 Data Management
As mentioned above, the observation process in the
EEL/DQP project is implemented in three proce-
dures: the Child Tracking Observation Schedule, the
Child Involvement Scale and the Adult Engagement
Scale. The observation details are registered in forms,
structured according to the type and nature of each
field. The EEL/DQP defines several fields, such as
the institution and the observer name, the date, time,
and the child’s name, sex and age. In addition it also
records the number of children and adults present dur-
ing the session, the child’s level of initiative (1 to
4), learning experiences, involvement (1 to 5) and in-
teraction, among others (Mesquita-Pires and Lopes,
2014).
At a macro level, the procedure is structured hi-
erarchically, in which an observation has several ses-
sions and each session has many activities. Consid-
ering the structure and the nature of each data field
as well as the associations between data entities, the
structure of information in this situation is represented
as five entities (Figure 5). The AES prefix indicates
Adult Engagement Scale specific data.
AESSession
AESObservation
Community Observation
*
1
SharedKey
User
1
1
Session
*
1
*
1
Activity
*
1
ACL
*
1
*
1
Figure 5: Entity diagram.
4.2 User Management
All the actions and data in the TeObs Social Network
are associated to and belong to a specific user. In this
context, users represent human agents that use a net-
work service. He has to be identified, through a user-
name, and authenticated using a password. This in-
formation is organized in specific entities (Figure 6).
The password is stored in the database in the form
of a unidirectional hash, so that even if the password
table is compromised, the attacker will not be able
to decipher it. The entity Certificate stores the
user’s certificates, containing the public key necessary
to protect the information he gathers.
ObservationalResearchSocialNetwork-InteractionandSecurity
61
String username
String email
String passwordHash
String name
int experienceLevel
User
String group
String username
Group
byte[] certificate
boolean invalid
Date invalidDate
Certificate
*
1
*
1
Figure 6: User elated entities.
5 SECURITY ISSUES
The ethical behavior inherent to all observation pro-
cedures demand rigorous security measures. In par-
ticular:
Users must be authenticated;
Data must be kept private;
Users must be authorized to perform an action.
Each of this points depend on different mech-
anisms. Authentication is performed through the
identification of a user and verification. Privacy is
ensured through cryptography, both symmetric, to
cypher data, and asymmetric, to deal with the key
distribution. Finally, authorization depends on access
control, through an Access Control List.
5.1 Authentication
Before any user is allowed to perform any action, he
has to be authenticated or, in other words, his identity
has to be confirmed. This is performed in three pos-
sible ways: using something the user has, something
he knows or something he is or does. We chose to
authenticate the user through something he knows – a
password.
To make password cracking more difficult, we
add a salt to each password (Wagner and Goldberg,
2000). The salt is a sequence of characters, generated
through a Cryptographically Secure Pseudo-Random
Number Generator (CSPRNG), that causes the hash
to be different even is situations where the password
is the same. The salt is stored in the user table along-
side the hash (Figure 6).
The procedure to store a password requires that a
random salt is generated, using a CSPRGN. Next, the
salt is concatenated to the password characters and
the hash is computed, using the SHA256 algorithm.
Finally, both the salt and the resulting hash are stored
in the database.
When accessing the social network, the user is re-
quested a password. Since only the hash is stored in
the database, password validation require:
1. Retriving the user’s salt and hash from the
database;
2. Prepending the given password with the salt;
3. Computing the SHA256 hash;
4. Comparing the resulting hash with the one from
the database.
If both hashes match, the user is authenticated.
Otherwise, the password is incorrect. To ensure pri-
vacy, users also store their certificates in the system,
making the public keys available to other users and to
the system.
5.2 Privacy
Data has to remain confidential, regardless of where
and how it is stored. The nature of the TeObs So-
cial Network requires the existence of a rigid privacy
policy as well as a verifiable consent from a parent to-
wards the protection of children’s privacy and safety
online. In this situation, privacy is achieved through
cryptography.
All the observation material is cyphered before
storing, so that it remains protected against disclosure.
However, data should not be prevented to be shared
between authorized members, although it should be
completely protected to a third party.
Traditional cryptography falls under symmetric,
where the same key is used to cipher and to decipher,
and asymmetric, also knows as public key cryptogra-
phy. This uses two keys: a private and a public. One
of the keys is used to cipher and the other to decipher.
In TeObs we use symmetric cryptography to cipher
data and asymmetric to distribute the shared key (Fig-
ure 7).
When starting a project, the supervisor (the gray
circle with a ‘C’) creates a new Community. The pro-
cess starts by generating a key, which will be ciphered
to the supervisor’s public key (black key from the
Certificates repository), and stored in the KeyStore.
Only the supervisor (or another element through del-
egation) has the possibility to add members to the
community, automatically creating a relationship with
him.
When a new observer is added to the community,
the shared key is deciphered (using the supervisor pri-
vate key), ciphered to the public key of the new ob-
server and stored in the KeyStore. Prior to data gath-
ering, the observer retrieves and deciphers the shared
key from the KeyStore. All the data is then ciphered
CSEDU2014-6thInternationalConferenceonComputerSupportedEducation
62
Figure 7: Key management.
and stored in the Community, ensuring that only the
members have access to the content.
In other words, each community will have a sin-
gle, specific, shared key. Although simple, this re-
duces the risk of compromise the keys, since no key
exist before the start of the procedure and it changes
with the community. At the same time, the data is se-
cure from eavesdropping, even if all the database and
all the stores are attacked.
To protect data against unauthorized or improper
modifications, each community member digitally
signs the piece of information submitted to the Com-
munity with his private key. This allows confirming
the ownership as well as the integrity of the data.
5.3 Authorization
Cryptography protects privacy and integrity of infor-
mation, as described above. However, enforcing pro-
tection also requires that every access to a system and
its resources be controlled and that all and only autho-
rized accesses can take place (Samarati and de Vimer-
cati, 2001).
Access control models are generally concerned
with whether subjects (any entity that can manipu-
late information), can access objects (entities through
which information flows through the actions of a sub-
ject), and how this access can occur.
In the TeObs Social Network, the concept of Com-
munity defines the members with privileged access to
information. However, further, fine grained, access
control policies are necessary. Considering the pre-
vious social interactions, we have several access pro-
files (Table 2).
Table 2: Access profiles.
N No access
R Can read
W Can write
R/W Can read and write
D Can delegate access
Considering a community member as the subject
and the data as the object, the possibilities of access
include reading, writing or delegating control, as sum-
marized above. The relation between the subject, the
object and the associated actions define the access
control policies, which will have to be met by the se-
curity mechanism.
This approach is designated by Discretionary Ac-
cess Control (DAC), based on the identity of the re-
quester and on access rules stating what requesters
are (or are not) allowed to do. This is instantiated
in an Access Control Matrix, a three dimensions ma-
trix relating subjects, objects and actions (Lampson,
1974). However, since this matrix is very sparse, it is
not very efficient to be stored directly. An alternative
implementation allows storing the matrix by column,
defining Access Control Lists (ACL).
ACL are associated with the objects, registering
who has access to it and how. In TeObs this is stored
as an entity associated with the Activity and the
User, in a many-to-many association (Figure 5). The
ACL will have an entry for each user with access as
well as the authorized action (Table 3).
Table 3: Access Control List.
Activity 1
User A R
User B R/W
User C R/W/D
The ACL builds on top of the privacy mechanism
to further define the security policy. This allows re-
stricting the possibilities within the community, al-
lowing or denying specific members specific actions.
For example, an user may read, write or delete infor-
mation although another may only read it.
Each user is responsible for defining each activ-
ity’s ACL. However, he can also delegate this function
to other user, allowing the supervisor, for example, to
update it.
ObservationalResearchSocialNetwork-InteractionandSecurity
63
6 CONCLUSIONS
Quality in education depends on the teachers’ pro-
fessional development, enhanced by observational re-
search in preschool environment and sharing and
working in a community of practice. Traditionally,
this community is based on face-to-face discussion
and exchange of experiences. This learning organi-
zation can be instantiated in an online social network,
where communication is mediated by computers.
A web based community allows enhancing com-
munication processes between members, particularly
out of regular teaching periods. Moreover, it allows
the integration of professionally isolated teachers.
In this paper we propose an observational research
social network for preschool education, where the in-
teractions between members are maintained and ex-
tended with the possibility for asynchronous commu-
nication. Moreover, a database of previous knowl-
edge is also available, allowing better training and fur-
ther studies.
Ethical behavior is fundamental, particularly pri-
vacy and access control. Cryptography is used for ci-
phering information and for key distribution and dis-
cretionary access control is used to specify the infor-
mation each user can access and associated actions.
This social network is complemented with smart-
phone application for gathering and exporting data, as
well as web pages for overall process management.
REFERENCES
Bertram, T. and Pascal, C. (2004). Effective Early Learning
({EEL):} A handbook for evaluating, assuring and
improving quality in settings for Three to Five Year
Olds. Amber Publishing, Birmingham.
Bertram, T. and Pascal, C. (2006). The baby effective
early learning programme: Improving quality in early
childhood settings for children from birth to three
years. Birmingham: Centre for Research in Early
Childhood.
Bertram, T. and Pascal, C. (2009). Manual {DQP} - de-
senvolvendo a qualidade em parceria. Minist
´
erio da
Educac¸
˜
ao, Lisboa.
Boyd, D. M. and Ellison, N. B. (2007). Social Network
Sites: Definition, History, and Scholarship. Journal of
Computer-Mediated Communication, 13(1):210–230.
Evans, K., Hodkinson, P., Rainbird, H., and Unwin, L.
(2006). Improving Workplace Learning (Improving
Learning). Routledge.
Garton, L., Haythornthwaite, C., and Wellman, B. (2006).
Studying Online Social Networks. Journal of
Computer-Mediated Communication, 3(1).
Laevers, F. (1994). Adult Style Observation Schedule for
Early Childhood Education ({ASOS-ECE)}. Centre
for Experiential Education, Lovaina.
Laevers, F., editor (2005). {Sics/Zicko.} Well-being and In-
volvement in Care Settings. A Process-oriented Self-
evaluation Instrument. Kind & Gezin e Research Cen-
tre for Experientel Education, Lovaina.
Lampson, B. W. (1974). Protection. ACM SIGOPS Operat-
ing Systems Review, 8(1):18–24.
Mesquita-Pires, C. (2012). Children and professionals
rights to participation: a case study. European Early
Childhood Education Research Journal, 20(4):565–
576.
Mesquita-Pires, C. and Lopes, R. (2014). Data Model
and Smartphone App in an Observational Research
Social Network. In 6th International Conference
on Computer Supported Education - CSEDU 2014,
Barcelona.
Samarati, P. and de Vimercati, S. (2001). Access control:
Policies, models, and mechanisms. Foundations of Se-
curity Analysis and Design, 2171:137–196.
Sheridan, S. M., Edwards, C. P., Marvin, C. A., and Knoche,
L. L. (2009). Professional Development in Early
Childhood Programs: Process Issues and Research
Needs. Early education and development, 20(3):377–
401.
Wagner, D. and Goldberg, I. (2000). Proofs of security for
the Unix password hashing algorithm. Advances in
CryptologyASIACRYPT 2000.
Wenger, E. (2010). Communities of practice and social
learning systems: the career of a concept. Social
learning systems and communities of practice.
CSEDU2014-6thInternationalConferenceonComputerSupportedEducation
64