Open Resources as the Educational Basis for a Bachelor-level

Project-Based Course

Ville Isom

ott

onen and Tommi K

arkk

ainen

Department of Mathematical Information Technology, P.O.Box 35, FI-40014, University of Jyv

askyl

a, Jyv

askyl

a, Finland

Keywords:

Open Data, Open API, Project-Based, Innovation Ability, Creativity.

Abstract:

This article presents an innovation-based course concept for project-based learning. In this course, student

groups are asked to ideate and implement a software product based on Open Data and Open API releases. By

emphasizing students’ own product ideation, the course requires and enhances self-directed learning skills and

prompts the students to see the unlimited possibilities in becoming and being a practitioner of the computing

discipline. Relatedly, the course provides a tool to improve student self-efﬁcacy, as the students, coached

through challenges, come to know that they are able to produce software using various open interfaces.

1 INTRODUCTION

Along with the decades-long project education

in computer science and software engineering

(Tomayko, 1998), various options for project-based

courses have been conceptualized (Fincher et al.,

2001; Clear et al., 2001; Burge and Gannod, 2009).

These taxonomic works discuss aspects such as indi-

vidual vs group projects, the problem scale and cur-

ricular levels of projects, and levels of realism, just

to name a few. Recent course conceptualizations

also include team work in multi-cultural (Pears and

Daniels, 2010) and multi-disciplinary (Burnell et al.,

2003) settings. In general, project education reﬂects

educational thinking that ‘function drives the form’

(Fincher et al., 2001), i.e., the educational goals for a

project can be reached by selecting applicable course

attributes.

The present paper describes and reviews a project

course concept where students innovate a software

product. The theme of this bachelor level course is

‘Open Data and Open APIs (Application Program-

ming Interfaces)’, which are on the increase among

societal interfaces. This theme was initially regarded

as topical and thereby potentially interesting to the

students. Later, the course coordinators began to pay

attention to the innovation ability of the students, and

the course is now considered to be a means to il-

lustrate to the students their potential as practition-

ers of computer science and software engineering.

The course under study is located at the University

of Jyv

askyl

a, Finland.

Several scholars agree that there is a need to include

innovation ability into higher education curricula, and

research on this topic is ongoing. Yunfei and Qin

(2009) posit that innovation ability is a natural at-

tribute of every person and that it can be cultivated by

education. Fila et al. (2012) discovered that students

associated innovation mostly with creativity, while

their study also calls attention to other dimensions

such as desirability and feasibility. The study by Yang

and Cheng (2010) in turn suggests that interaction be-

tween individuals across project teams enhances cre-

ativity. In the present paper, the speciﬁc research aim

is to identify and discuss challenges that students,

who are expected to be inspired by the opportunity

for creative software development in groups, arrive at

a course fully based on the students’ own ideation in

a setting where the work is not done for the teacher or

an external client or sponsor.

The present work is part of the diagnostic phase of an

action research project. Action research, coined by

Lewin (1946), addresses local concerns to induce a

social change and improved understanding of the so-

cial situation. Educational action research is typically

conducted by or with authentic participants; through

such insider roles, action research is considered crit-

ical and practical (Carr and Kemmis, 1986). The

need to add a project course at the bachelor level was

identiﬁed by educators who observed a multitude of

learning challenges among students during a master’s

level authentic customer project in the local curricu-

lum. Adding the second/third-year practical course

was an attempt to smooth the students’ study path to

Isomöttönen V. and Kärkkäinen T..

Open Resources as the Educational Basis for a Bachelor-level Project-Based Course.

DOI: 10.5220/0005432100460056

In Proceedings of the 7th International Conference on Computer Supported Education (CSEDU-2015), pages 46-56

ISBN: 978-989-758-108-3

 2015 SCITEPRESS (Science and Technology Publications, Lda.)

the customer collaboration project. The Open Data

and APIs provided a realistic setting for this early

curriculum course, and thereby introduced a fruitful

educational environment. So far we have run the

course twice and are identifying the main educational

achievements and issues to be addressed. This pa-

per is presented in a lessons learned-style, much like

a self-ethnography wherein authentic participants in-

vestigate observations that have raised their research

interest (Kaihlavirta et al., 2015)—the action research

contextualization will be emphasized in future work.

2 BEING ‘OPEN’ IN THE

DIGITAL ERA

When a resource is open, it is available for others to

use. It is freely accessible, usable, modiﬁable, and re-

distributable as-is or as a derived work. Similarly, an

open activity or process can be followed and affected

by others during its execution. However, when the at-

tribute of openness is attached to different things, we

end up with multifaceted and varying conceptualiza-

tions, such as open data, open source software, open

service, open innovation, open education, open infor-

mation society, open government, open science and

research, etc. (see, e.g., (Jaakkola et al., 2014b)). To

be known as open, a resource or outcome of a pro-

cess is typically attached with an open license. For

a piece of software, this can mean complete freedom,

even to close the derivative works, as with the permis-

sive MIT license. Another well-known alternative is

to enforce the derivative works to be similarly avail-

able as the original source code, as with the copyleft

GPL licenses (see Tuunanen et al. (2009) and refer-

ences therein).

Open data is accessible and usable for further pro-

cessing and reﬁnement. According to Domingo et al.

(2013): ‘Open data is the concept that deﬁnes the pub-

lication of government or private company data with-

out copyright restrictions. The data should be format-

ted so that citizens can reuse it at their discretion to

create new, innovative services or applications.’ How-

ever, the overall rights for even a raw data resource

should be governed by a content license, such as one

from the creative commons family

. Moreover, open

(and sometimes big) data is often not enough, but

one should be able to turn it into interesting patterns,

models, and visualizations (Han et al., 2011) that pro-

vide answers to some speciﬁc questions of true value

(Hand, 2013). In the Finnish national context, the de-

velopment of the open data movement has been sum-

https://creativecommons.org/licenses/

marized by Jaakkola et al. (2014a).

Recently, many tools and methods for utilizing

open data have been proposed: visual exploration of

open data (Otjacques et al., 2012), a ﬂexible envi-

ronment for Web data integration (Castanier et al.,

2013), a domain-speciﬁc language (DSL) for open

data visualizations over the Web (Morales-Chaparro

et al., 2014), and a high level library to help de-

velopers build safe mashups over APIs in HTML5

(Telikicherla and Choppella, 2014), to mention a

few. Open data can also be utilized in existing or

novel games, as suggested by Friberger and Togelius

(2012). Concerning the Semantic Web tools, the inte-

grated knowledge base of open data is referred to as

the Linked Open Data (LOD) (see Auer et al. (2014)

and articles therein). Arguably, such massive data

sources increase complexity of applications but also

their potential beneﬁts.

Activities to create, publish, and utilize open data

are becoming more and more popular. In many cases,

availability of open data can be linked to an open in-

novation process fostering collaboration between pri-

vate companies (especially in the creative industry),

governmental or public actors, citizens, and academia

(scholars and students), once again to create new

products and services through purposive inﬂows and

outﬂows of existing or newly created knowledge (e.g.,

(Huizingh, 2011; Conradie et al., 2012)). The exis-

tence of a local co-creative platform can be an impor-

tant precondition for publishing some existing data,

even if to reach the actual innovative outcomes re-

quires continuous efforts and allocation of time from

all the participants.

It should be noted that to be able to be innova-

tive may require a basic understanding of the actual

term innovation, which, for undergraduate students,

could be challenging (Fila et al., 2012). Similarly,

as argued by Dahlander and Gann (2010), scientiﬁc

inquiries related to open innovation reveal a lack of

clarity regarding the term’s meaning, especially in the

business context. However, the paradigm of open in-

novation has been proposed by Chesbrough (2003),

with the encouragement of knowledge inﬂow and out-

ﬂow to ensure that tasks are completed efﬁciently and

effectively. He sees open innovation as the key to de-

veloping novel products and services in the modern

business world. He proposes six principles to achieve

open innovation, two of which are relevant here: 1)

work with smart people inside and outside the com-

pany (university students of CS should qualify), and

2) a ﬁrm (or a student team) does not have to originate

the research (or data and software) to proﬁt from it.

The innovation process and outcome can be closed

or open. For instance, an open source software com-

OpenResourcesastheEducationalBasisforaBachelor-levelProject-BasedCourse

ponent can come from a private software company

taking part in the open source development. Simi-

larly, the user-centric design process of, for example,

texture for a new curtain model, might end up as a

new product sold under a (closed) trademark. As edu-

cators in the digital era, we should encourage students

to make the outcomes of their Open Data/API projects

open (Huizingh, 2011). So far, the main focus of the

course studied in this paper has been inbound open

innovation (i.e., utilization of existing open resources

to produce an own software deliverable).

The present section contextualized our study in

reference to the use of open resources, while we are

aware of the other kind of literature describing var-

ious options for implementing a project course (e.g.

(Fincher et al., 2001; Clear et al., 2001)). The pecu-

liarity of the course studied here is the high degree of

student responsibility expected in the early curricu-

lum. Comparison between this aspect and early cur-

riculum project models in the literature merits a sepa-

rate study.

3 THE COURSE

3.1 Workload, Learning Objectives, and

Facilities

The project course spans 12 weeks and students are

rewarded with 4 to 6 ECTS credits: the intention is

that each student will earn 5 credits. The course is

intensive and challenging; the software product pro-

totypes illustrating at least a proof of concept are

ideated and implemented in the given time frame

through newly formed groups.

The initial learning objectives were to introduce

and conceptualize group processes and software pro-

cess issues to the students through realistic project

work at the bachelor level. After two course itera-

tions, these objectives have been complemented with

the ones emerging from the innovation-based course

concept; that is, prompting self-directed study pro-

cesses among the students and improving student self-

efﬁcacy. These latter kinds of learning objectives are

the main focus in the present paper.

Related to the possibility of improving self-

efﬁcacy, ideating and producing software from

scratch sets a technical learning goal for the students.

Pre-requisite courses include CS1 and CS2, and all

the students are thus expected to have basic program-

ming skills. In our CS1 course, students program a

graphical game using a particular library. A few stu-

dents have taken a web programming course and our

CS2 has an elective continuation part brieﬂy introduc-

ing web programming. The introductory program-

ming courses (CS1 and CS2) are the minimum pro-

gramming background for the project, however. From

this premise, technical support by which students’

problem situations are reviewed and resolved, often

by directing students to useful learning resources and

software component resources, is an important facili-

tating element of the course (see Section 3.2).

Each group is provided with a lockable work room

equipped with personal computers for each student

to support students’ realistic and autonomous work.

The course thus differs from studio-based learning en-

vironments where students’ work is guided through

ﬁxed practice sessions (see, e.g. Suri (2007)). The

faculty’s PC support is available to the students, and

they are granted local administrative rights to install

and conﬁgure software independently. Typically, pro-

gramming code is managed in a version control sys-

tem, and each student pulling from the remote reposi-

tory can run the needed server applications and so on

in the local computer (localhost). The intended group

size is four students, though, due to course popula-

tion, a few groups have also comprised 3 or 5 stu-

dents.

3.2 Teaching Resources

The course is taught by two supervisors. A depart-

mental teacher (the ﬁrst author) supports students

with group work issues and software process issues,

while a senior student works as a technical supervisor.

The senior student recruitment is based on a strong

personal interest in the course topic (creative software

development based on the open theme).

Absorbing the overall architectural idea of how to

work with APIs and integrate data and software com-

ponents into new products is important for the stu-

dents. It is this competence that the technical su-

pervisor of the course needs to possess. Many API

releases are supported with an open source wrapper

code. Then, what remains as a task of the program-

mer, in order to receive the data needed, is studying,

integrating, and potentially modifying the wrapper.

Similarly, parsing various data formats used in open

data and API context, such as JSON, is often an ef-

fortless task, while open source components also tend

to exist for parsing less popular data formats such as

PC-Axis. In light of these examples, students are to

be informed of the conventions and existing techno-

logical possibilities of small-scale open source (web)

development. The technical supervisor, attending all

the per group supervision sessions, importantly facil-

itates the students’ technology adoption process.

CSEDU2015-7thInternationalConferenceonComputerSupportedEducation

The aim is to develop a course climate where the

personnel reﬂect strong interests in their speciﬁc ex-

pertise areas, to inspire students to develop profes-

sional interest in the course topics. In our experience,

the course has been manageable for two persons with

up to 7–8 groups and a maximum of four students in

each.

3.3 Joint Course Events

The course structure is displayed in Figure 1. The

top line describes events concerning all the students.

The course begins with a start-up meeting during

which the course idea, course events, and documen-

tation are explained to the students. This meeting

also provides students an introduction to Open Data

and APIs, by giving examples of the related appli-

cations found on the Web and a comprehensive list

open resources. Both national and international (EU)

open data and API links are listed, and known ser-

vices with API releases, such as Spotify, Twitter, Bib-

sonomy, and Trello, are noted. To enable students to

picture about the scope of their project, exemplary re-

sources are discussed through speculative examples

(“Trello could be complemented with a component

that ﬁrst exports users’ work hours markings and then

imports a sum of them with visualization”). Students

are also guided to independently seek other fresh data

releases based on their own interests. They are also

told that HTML scraping could be needed to receive

data within the interesting topic. Students are grouped

by the teacher at the start-up meeting. Because the

course pedagogically focuses on group work issues,

the aim is to group student who have not worked to-

gether previously. This way each group’s dynamic

has a fresh and equal start.

The next shared event is the project lecture

where group work concepts, including norms, sta-

tuses, roles (Brown, 1988), and justice in group work

(Isom

ott

onen, 2014), are discussed. The topic of soft-

ware processes is also included. Here, discipline in

the software development is carefully explained and

linked to project safety. A particular emphasis is also

given to iterative software process, which has been

found to ﬁt realistic student projects (Brown, 2000;

Isom

ott

onen, 2011). On the third week, a shared ses-

sion is arranged where the groups describe the project

topics they have ideated. This session is intended

to reinforce the innovation theme across groups and

share considerations on the use of data resources and

the designs of software products. Near mid-course,

an expert lecture on open source software licenses is

provided to increase student awareness on the licens-

ing issues and enable them to agree on the licensing

of the group products.

At the end of the course, groups present their soft-

ware products during an ‘open day’ for the depart-

ment’s personnel and other students. In place of tra-

ditional presentations, students present in their project

rooms and allow the audience to try the software prod-

ucts and discuss design and implementation. This

way of presenting fosters dialog among the attendees.

3.4 Minimal but Important

Documentation

The mid-line in Figure 1 describes the documenta-

tion required during the course, which is minimal. At

the beginning of the course, student groups prepare

short synopses for 1) what they have agreed as their

project topic and 2) how they are going to manage

their work. These are intended to help the students

be aware and take responsibility. In the synopsis on

the project topic, student groups provide an abstract

for the topic but also address the important feasibility

questions below:

• further ideas and visions,

• target group (utility aspect),

• licenses and terms of conditions of the intended

data usage,

• data formats,

• technical environment (programming languages,

platforms, libraries, etc.),

• licensing of the product (under what open source

license the product is released), and

• boundaries of the project, i.e., what parts of

the ideated project are most relevant and to be

achieved during the project.

These questions are reviewed in per group supervi-

sion sessions. A potential gray zone in data usage

is thus avoided since the students are guided to ﬁnd

out and comment on the conditions of their data us-

age. Should there be any open questions, these will

be naturally raised since the synopsis questions are

reviewed. Because student work is by default owned

by the students, they naturally decide on the licens-

ing of the products. In the synopsis, groups make an

initial agreement on an open source license for their

product. The licensing question is later supported by

an expert lecture (see Figure 1), after which the stu-

dents are prepared to make an informed decision, tak-

ing into account the licenses of the used components.

In the course based on the students’ own ideation, the

university does not require any transfer of rights.

Given the emphasis on the students’ own prod-

uct ideation, the projects are based on the students’

OpenResourcesastheEducationalBasisforaBachelor-levelProject-BasedCourse

Figure 1: Course events and tasks during 12-week creative software development effort.

proposals with no initial inclusion/exclusion criteria

on APIs, data, or application type. For instance, the

product can be a web service or a game, as well as a

desktop application, and we have also suggested the

use of Arduinos and Raspberry Pis in the context of

open resources. Students must nevertheless consider

a target group, which implies that the projects become

realistic as compared to programming exercises with

no user interest. The open theme enables small-scale

products starting from ones making single data source

utilizable through data manipulation and visualization

(cf. small tools for data journalism), while often the

ideated products are larger. For this reason, student

groups prioritize and deﬁne boundaries for what they

aim to achieve during the project (see synopsis items

above). We want to prompt students to be creative,

while the synopsis document and weekly supervision

discussions help groups set their goals realistically

during the project.

At mid-course, self-evaluations on group work

and software process are conducted, each student

completing a survey form. The teacher inspects the

evaluations and raises their main points during a

group discussion session; the group situation revealed

by the evaluations is openly discussed. When the

projects are complete, each student prepares a per-

sonal learning report, reﬂecting on group work and

software process issues in light of both lectured the-

ory and conceptualizations that emerged during the

project. The teacher gives a written response to each

learning report to enhance student learning.

The inclusion of very little documentation means

that all tasks are substance tasks that advance the ac-

tual ideation and software product development. This

makes it almost impossible for students to limit their

involvement to completing some secondary tasks; po-

tential ‘passenger’ roles become visible and can be

raised as group issues. For project management, the

students use VCSs (with no exceptions, this has been

Git) and project management software, such as Trello.

It should be noted that limited documentation does

not indicate low teacher workload, as sensitive group

discussions (Section 3.5) are very challenging for su-

pervisors and require preparation and continuous re-

ﬂection.

3.5 Emphasis on Dialog Through

Group Discussions

The bottom line in Figure 1 illustrates that teach-

ing this course means coaching through group dis-

cussions. Thus, a discussion session attended by the

course teacher, the technical supervisor, and the stu-

dent group, is arranged each week for all the groups.

The group situation, software process, and various is-

sues in product ideation and implementation are dis-

cussed in these sessions. Informal discussions were

considered suitable for the creativity-based course,

and are a tool to introduce theory in the presence of

authentic practical work; emergent problems are con-

templated in terms of theory. During a particular ses-

sion, the written self-evaluations provide the basis of

discussion, which aims to guarantee that all the stu-

dents are heard through personal writings during the

project.

In this short course, the main learning objectives

are introduced at the beginning through lectures and

are then intensively addressed in the group discussion

sessions throughout the course in the context of ac-

tual individual projects. The main pedagogic princi-

ple is based on the realist epistemology (see Bhaskar,

1978; Moore, 2000). Thus, it is based on an assump-

tion that there are important objectiﬁcations that can

explain to the students their project successes and

challenges, and that these objectiﬁcations must be

raised during the project to foster conceptual under-

standings among the students. It is important to note

that the innovation-based open-themed projects have

provided a good forum to introduce truly authentic

project work where group issues, for instance, emerge

naturally and can be conceptualized to the students re-

alistically. Taking group work as the example case,

the pedagogy of the course was described in detail in

another study (Isom

ott

onen, 2014). As mentioned, in-

CSEDU2015-7thInternationalConferenceonComputerSupportedEducation

stead of group processes and software process issues,

this paper focuses on educational achievements and

challenges arising from the course concept.

3.6 Pass/Fail gGrading

Considering the sensitive issues (e.g., justice in group

work) discussed during the course and the aim to

promote the students’ innovation ability, grading is

pass/fail. Without competitive or external pressure

of numeric grading, students are prompted to over-

come their difﬁculties in adopting self-directed, cre-

ative, learning processes and to fully focus on concep-

tualizations that explain their group experiences. Pos-

itive experiences with a project-based course without

numeric grading have been reported by Daniels et al.

(2004). Promoting student interest in course content

instead of ‘just passing the course’ is also in line with

the Klug’s (1976) work. He linked (numeric) grading

with a degree system that does not necessarily match

with the learner’s personal intellectual development.

In our course, students track their work hours.

However, passing the course is not based on quanti-

tative inspection of student work hours or amount of

programming code. These attributes approximate stu-

dent role in the group, but they do not explain the ef-

fect of group situation on the student’s possibility to

participate. The passing is based on active participa-

tion, which is fairly easy for the teacher to interpret

with the selected course arrangements: the frequent

meetings between the supervisors and groups and the

related discussions on group processes. During these

discussional sessions, reasons for low participation

are objectively addressed in terms of group work pro-

cesses, following that the students themselves become

aware if their difﬁculties result from group work or if

they are truly not participating in the course. In the

ﬁrst case, various solutions for improving the group

situation are sought for. For instance, by improv-

ing intra-group communication the division of work

could be improved to match the skill levels of group

members.

On the basis of the exposition above, ‘failing’ and

‘dropping out’ can be said to mean same thing in this

course. Accordingly, two dropped out students of the

course agreed to drop out for low participation which

was due to reasons external to the course. It should

also be noted that group difﬁculties do not directly

mean dropping out; rather, through active participa-

tion the difﬁculties undergone can be reﬂected on for

the sake of conceptual learning.

4 LESSONS LEARNED

Based on the experiences collected, Open Data and

Open APIs enable both a unitary course theme and

full innovation by students. In the context of en-

trepreneurial learning, Kyr

o (2005) has stated that

room for possibilities and creativity must be provided

by the teacher; in our experience, these are easy to

implement with the open theme where software de-

velopment and creativity easily elide.

4.1 Main Success: Ability to Create a

Software Product in a Student

Group

All the groups have been able to implement software

products that are proofs of concept for the ideated

uses. Some products have been mature and ready for

deployment. Preliminary analyses of students’ learn-

ing reports indicate that the students are able to re-

ﬂect their experiences on the group work concepts

and become aware of the meaning and purpose of

the software process in software development work.

After two course iterations, of the 43 students who

have started the course, two have dropped out, as

they noticed their low participation. Besides the two

dropouts, no-one has failed. As noted in Section 3.6,

in this course where active participation is the main

criterion for passing, failing and dropping out mean

same thing. Other two students were given a small

additional assignment to give them a genuine chance

to improve on their skills in the presence of different

skill levels in the group, the issue that led to uneven

participation and could not be satisfactorily resolved

in the course of the short project. In our experience,

the open theme provides a feasible contextualization

for the 12-week project-based course.

Various project topics have been ideated and im-

plemented. One group developed a zombie game us-

ing Open Street Map information as the game ﬁeld

(as suggested by Friberger and Togelius (2012)). An-

other group developed a service where listings for

cheap apartments for rent (suitable for a student bud-

get) were collected from several service providers and

shown on a map of the local area with links to fur-

ther information. Another of the location-based appli-

cation placed elk accidents on the map with various

UI controls for ﬁltering the information. One group

developed an application that calculated and visual-

ized expenses related to car, train, or bus travel. An-

other project enabled the user to place various United

Nations data sets in a three-dimensional world map

where they could visualize differences between coun-

tries using height and color.

OpenResourcesastheEducationalBasisforaBachelor-levelProject-BasedCourse

Of the eleven projects ﬁnished so far, nine are web

services, one is a desktop application, and the remain-

ing one a game started as a desktop application. Data

usage scenarios include scraping, having data locally,

and using data through API calls. As for the scope

of projects, students typically (in minimum) aim at

ﬁnishing the main use case where the integration of

components that request, parse, manipulate, and show

about data is already in place, and by which the prod-

uct idea can be demonstrated to an audience.

Students overcome considerable technological

learning challenges on top of starting the project with

a newly formed group. Many of the students are unfa-

miliar with Linux, which is a straightforward operat-

ing system for many frameworks used in small scale

Web application development (the typical technologi-

cal context for the student ideated projects). Students

often learn a new programming language, such as

Python, while they use C# and Java during their intro-

ductory programming courses. They typically write

code for running a server and build UIs with HTML5

and JavaScript, with a particular CSS library, such

as Bootstrap. Some groups choose to adopt modern

frameworks, such as AngularJS. The students learn to

seek and select recent tools particularly applicable for

their projects. This process is supported by the techni-

cal supervisor. Overall, students learn to build useful

software products by integrating various data sources

and technologies, meaning that many of the products

reﬂect the ‘mashup’ character in recent Web develop-

ment.

4.2 Challenges and Suggested

Improvements

When students begin this course, which requires both

creativity and implementation of a ‘proof of concept’

product in a short time frame, they are undertaking a

learning process where issues are bound to emerge.

These are reviewed below.

4.2.1 The Challenge of Self-directed Learning

Students tend to struggle with relating to a course re-

quiring self-direction and individual responsibility in

a group. After being exposed to the highly structured

introductory university courses during their prerequi-

site studies, they face a course in which they do not

work for the teacher at all. Therefore, we must con-

sider system level in terms of schooling background,

and, in particular, the ﬁrst, usually highly structured,

university courses. In the present computer science

context, for instance, we wish to answer the question

What is the effect of evenly educating and

highly structured introductory programming

courses on students’ perceptions of self-

direction required in the actual profession?

Beginner’s difﬁculties with programming are ex-

tensively documented. If highly structured and

evenly educating introductory courses are needed,

then courses such as the one presented here could

help prompt self-direction early after the introductory

courses. Students’ ﬁrst experiences with university

courses are in any event likely to reinforce a ‘depart-

mental view’ of their positions as learners (Parlett,

1977).

We also put forward the question above on the

grounds that, during the project course, students are

heavily informed of the active learner position re-

quired and the related pedagogic purpose, yet we have

observed that a great deal of coaching resources are

expended on this concern of self-direction. We are

also fully aware of and using the coaching practice

that prompts students to adopt concrete project rou-

tines early to decrease the known confusion stage at

the beginning of projects.

To improve students’ awareness of the require-

ment of self-direction, we plan to better describe the

roles of all stakeholders at the beginning of the course.

This means that we will increasingly inform students

about the role of the teacher, which is not that of a

customer (but a facilitator). In this connection, we

will also emphasize how the course setting is different

from other courses that require coursework through

weekly lectures and exercises and provide example

programs that can be followed as references for the

course assignments.

4.2.2 The Challenge of New Technology

Students may feel overwhelmed by the requirement

to ideate, implement, and showcase a software prod-

uct from scratch in 12 weeks. They often seem un-

clear on the effort needed to implement a useful soft-

ware by integrating data resources using technolo-

gies that are new to them. We argue that this re-

lates to students’ experiences of self-efﬁcacy, such

that low self-efﬁcacy hinders students’ ability to adopt

work in the innovation-based environment. In prin-

cipal, students are usually not troubled by the task

of ideating, whereas their in-progress professional

self-efﬁcacy related to adopting new technologies can

present severe obstacles in project initiation. We ob-

serve that students who are further in their bachelor

studies struggle less with this issue.

So far our attention to technical issues during the

starting session has been a review of couple of exam-

ple programs in the sense of promoting student cre-

CSEDU2015-7thInternationalConferenceonComputerSupportedEducation

ativity, showing examples of open resource usage, and

providing initial tip list of web development and other

frameworks and libraries.

We now plan to concretize more the software de-

velopment effort for the students at the beginning of

the course. This could mean adding a separate crash

course lecture that shows example applications to-

gether with programming code, and carefully explains

the design and the code to the students. This would

allow us to provide a correct picture of the course

challenge among the students immediately when the

course starts, and illustrate the course as a doable ef-

fort. At the same time, we would be able to demon-

strate scraping and typical uses of APIs early for all

the students, and potentially reduce less experienced

students dependence on more experienced ones. Fur-

ther, potential emotional reactions to independent

technological learning requirement should also be ex-

plicitly discussed at the beginning of the course to

help students understand that these feelings are nat-

ural.

4.2.3 The Challenge of Variations in Student

Characteristics

After two course iterations, we have noticed the need

to address differences in skill level in forming the

groups. We acknowledge here that differences in skill

level can lead to enriched learning experiences dur-

ing group work (Oakley et al., 2007), which is why

we have not given a lot of attention to this aspect in

grouping so far. However, the project’s goal setting

on the basis of similar skill levels in a group would

be important in light of how students may be over-

whelmed by the requirement of technological learn-

ing (cf. Section 4.2.2). With great differences in skill

level, the skillful students’ high goals may force less

skillful students to be dependent on others through-

out the project. We want to avoid such phenomena

stealing attention from the task of innovation itself.

We plan to roughly group students according to study

year and hobbyism in the future.

Another challenge relating to student characteris-

tics is how they relate to (in this case, creative) soft-

ware development work. Regardless of the empha-

sis on creativity and innovation, the course is one of

many required in the degree system. For us, the im-

portant question is to what extent we should push stu-

dents toward passionate work. This concern emerges

from students reporting discomfort with a highly pas-

sionate group member and a desire to take a more

moderate approach to work. Practical courses re-

ﬂect working life, and the question of how to emo-

tionally and motivationally relate to the work in the

group context could be raised during the project lec-

tures among other group work topics.

4.2.4 Lack of a Proper Release Challenge

An unfavorable pattern can emerge when students

overcome their group work and self-direction chal-

lenges and show abilities to ideate and implement a

product, only to see the completed project as ‘just

a course assignment’ without big-picture relevance.

This perspective can leave the course without pur-

posive outﬂow of newly created products (Huizingh,

2011). Different from our master’s level customer

project, these is no ﬁnal release of project deliver-

ables. In addition, in the context of the local CS cur-

riculum, there is no larger business study component

underpinning the course that would inspire students

to continue with their projects as the course ends.

Here, our plan is to consult with students about es-

tablishing a showcasing site with links to GitHub or

other platforms used by students for their product de-

velopment. Such a site could be connected with a lo-

cal open data community, meaning that student work

would become ‘involved’. The open data contests in

Finland (see (Etel

aaho, 2014)) could be exploited for

this purpose as well. The important question here is

how to deal with differences in the maturity of the

products (i.e., how will students perceive these differ-

ences if the products are publicly available).

4.2.5 The Intellectual Property Challenge

When exploring various resources for their work

(data, APIs), students encounter a ‘jungle of terms of

conditions (TOCs)’. This particularly occurs when a

student group uses data sources that are ‘semi-open’

in nature, with some speciﬁc and often ambiguous

clauses stating the conditions of use. The intention is

to act legally. Unclear situations have been currently

resolved with the teacher requesting expert comments

from the department personnel known to possess the

competences needed. Alternatively, where students

have scraped web content, the teacher has guided

groups to contact the service providers to request per-

missions for using the data, which has been a suc-

cessful procedure. The groups have either received

permissions or an answer that has helped them to re-

vise their project goals. Altogether the IP issues have

introduced a great challenge, kind of a delay element

from the teaching perspective, as the interpretation of

TOCs and various small clauses can require juridic

expertise.

Preferably, those groups who deal with complex

license and TOC texts could be provided with a ‘con-

sultation hour’ from the university jurist. The project

OpenResourcesastheEducationalBasisforaBachelor-levelProject-BasedCourse

course is one alternative for the mandatory practi-

cal study component in the students’ bachelor stud-

ies, and the responsibility for answering juridic ques-

tions that arise during the course naturally rests on the

course provider. Having an explicit course arrange-

ment for this purpose would make the course begin-

ning more ﬂuent and underline IP issues as an impor-

tant learning topic. Another option would be to guide

the students to limit their project ideation to resources

whose conditions are phrased without any ambigui-

ties. However, in our view, this latter option would

limit how the course concept can foster student inno-

vation ability.

In our experience, the use of various data re-

sources and APIs increases both students’ and su-

pervisors’ useful knowledge on intellectual property

rights and licenses, for what reason we recommend

that also the course teachers attend the expert con-

sultation hours outlined above. These topics are

also similar to recommended topics in CS by the

ACM/IEEE Computing Curricula (‘Social Issues and

Professional Practice/Intellectual Property’) (Sahami

et al., 2013, p. 24).

4.2.6 Summary

Many of the challenges above reﬂect our overall expe-

rience that the course concept described is an efﬁcient

tool for prompting particular transformation and self-

related honesty early in the curriculum. That is, stu-

dents can be coached to overcome their self-efﬁcacy-

related challenges and learn how interested he or she

is in the computing discipline. In our experience, this

transformation does occur, and at different paces for

different individuals. The teacher’s patience and con-

tinuous coaching are keys to the transformation to-

ward self-directed creative work.

5 CONCLUSIONS

This paper described a course concept that builds on

student innovation ability, including attributes such as

creativity and co-creation. The course is based on

open data and programming application interfaces.

Apiola et al. (2012) have recently proposed that

the CS higher education in Finland should better sup-

port creativity and innovation-friendliness, especially

creative problem solving. The purpose of this pa-

per was to describe, as the early stage of an ac-

tion research project, one approach toward this di-

rection. More precisely, Apiola et al. proposed that

components of a creativity-supporting learning envi-

ronment include competence (i.e., solving challeng-

ing problems), autonomy (i.e., self-direction), relat-

edness (i.e., social interaction and self-adaptive work-

ing methods), domain relevant skills (i.e., good com-

puting skills), and cognitive processes and working

styles (i.e., creativity-enhancing methods in course

sessions). We conclude that, according to such a def-

inition, a creativity-supporting learning environment

in the undergraduate project course was established.

The open theme provided the framework for imple-

menting a course concept of this kind.

Our speciﬁc interest was to report challenges that

could inform our subsequent action research and that

of other educators designing similar courses. In our

view, many of the challenges reviewed relate to self-

direction and the curricular position of the course. Af-

ter highly structured introductory university courses,

students may be overwhelmed by the challenge of a

creativity-based course. Generally, this concerns how

students are able to undertake self-directed study pro-

cesses, which here signiﬁes the responsibility to ad-

vance the project without an external customer. A

related challenge concerns the students’ self-efﬁcacy,

which emerges from the challenge of independent

technological learning required by the course. Stu-

dents may not know that useful, proof of concept open

data applications can be created with reasonable ef-

fort, and demonstrate low self-efﬁcacy not knowing

that this is unnecessary.

We have altogether observed that student groups

are able to overcome their challenges to produce

an outcome. As reported by Lin and Chen (2012),

commitment and self-efﬁcacy are important mediat-

ing factors for a student group with diverse compe-

tences to grow into a team capable of open innova-

tion. Based on the collected experiences, continuous

coaching and teacher patience contribute to transfor-

mation toward condition that is fruitful for innova-

tive work in the early curriculum. In this regard, our

subsequent action research is likely to beneﬁt from

the theory of transformative learning (Mezirow, 1981)

as a framework for a ‘hidden curriculum’ designed

to improve students’ self-efﬁcacy in the computing

discipline. More precisely, we need to conceptual-

ize a particular ‘perspective transformation’ (a term

used by Mezirow) where a learner realizes that self-

directed work emphasized through creative effort re-

ﬂects the computing profession and is therefore valu-

able.

REFERENCES

Apiola, M., Lattu, M., and Pasanen, T. A. (2012).

Creativity-supporting learning environment—CSLE.

CSEDU2015-7thInternationalConferenceonComputerSupportedEducation

Trans. Comput. Educ., 12(3):11:1–11:25.

Auer, S., Bryl, V., and Tramp, S., editors (2014). Linked

open data – creating knowledge out of interlinked

data: results of the LOD2 Project. Springer.

Bhaskar, R. (1978). A Realist Theory of Science. Harvester

Press, Sussex, UK, second edition.

Brown, J. (2000). Bloodshot eyes: Workload issues in com-

puter science project courses. In Software Engineer-

ing Conference. APSEC 2000. Proceedings. Seventh

Asia-Paciﬁc, pages 46–52. IEEE Computer Society.

Brown, R. (1988). Dynamics within and between Groups.

Basil Blackwell, Oxford, UK.

Burge, J. and Gannod, G. (2009). Dimensions for catego-

rizing capstone projects. In Software Engineering Ed-

ucation and Training. Proceedings. 22nd Conference

on, pages 166–173, Los Alamitos, CA. IEEE Com-

puter Society.

Burnell, L. J., Priest, J. W., and Durrett, J. R. (2003). As-

sessment of a resource limited process for multidisci-

plinary projects. SIGCSE Bull., 35(4):68–71.

Carr, W. and Kemmis, S. (1986). Becomming Critical: Ed-

ucation, Knowledge and Action Research. The Falmer

Press, London.

Castanier, E., Coletta, R., Valduriez, P., and Frisch, C.

(2013). Websmatch: A tool for open data. In Pro-

ceedings of the 2Nd International Workshop on Open

Data, WOD ’13, pages 10:1–10:2, New York, NY.

ACM.

Chesbrough, H. W., editor (2003). Open Innovation: The

New Imperative for Creating and Proﬁting from Tech-

nology. Harvard Business School Press, Boston, MA.

Clear, T., Goldweber, M., Young, F. H., Leidig, P. M., and

Scott, K. (2001). Resources for instructors of capstone

courses in computing. In ITiCSE-WGR ’01: Working

group reports from ITiCSE on Innovation and tech-

nology in computer science education, pages 93–113,

New York, NY. ACM.

Conradie, P., Mulder, I., and Choenni, S. (2012). Rotter-

dam open data: Exploring the release of public sec-

tor information through co-creation. In Engineering,

Technology and Innovation (ICE), 18th International

ICE Conference on, pages 1–10.

Dahlander, L. and Gann, D. M. (2010). How open is inno-

vation? Research Policy, 39(6):699–709.

Daniels, M., Berglund, A., Pears, A., and Fincher, S.

(2004). Five myths of assessment. In Proceedings

of the sixth conference on Australasian computing ed-

ucation - Volume 30, ACE ’04, pages 57–61, Dar-

linghurst, Australia. Australian Computer Society.

Domingo, A., Bellalta, B., Palacin, M., Oliver, M., and

Almirall, E. (2013). Public open sensor data: Revolu-

tionizing smart cities. Technology and Society Maga-

zine, IEEE, 32(4):50–56.

Etel

aaho, A. (2014). Analysis of the received applica-

tions in the open data contests in Finland 2010–2013

(draft.). Technical report, Tampere University of

Technology, Department of Pori.

Fila, N., Myers, W., and Purzer, S. (2012). Work in

progress: How engineering students deﬁne innova-

tion. In Frontiers in Education Conference (FIE),

2012, pages 1–6.

Fincher, S., Petre, M., and Clark, M., editors (2001). Com-

puter Science Project Work: Principles and Pragmat-

ics. Springer-Verlag, London.

Friberger, M. G. and Togelius, J. (2012). Generating game

content from open data. In Proceedings of the In-

ternational Conference on the Foundations of Digi-

tal Games, FDG ’12, pages 290–291, New York, NY.

ACM.

Han, J., Kamber, M., and Pei, J. (2011). Data Mining: Con-

cepts and Techniques. Morgan Kaufmann Publishers,

San Francisco, CA, USA, 3rd edition.

Hand, D. (2013). Data, not dogma: Big data, open data,

and the opportunities ahead. In Tucker, A., H

oppner,

F., Siebes, A., and Swift, S., editors, Advances in

Intelligent Data Analysis XII, volume 8207 of Lec-

ture Notes in Computer Science, pages 1–12. Springer

Berlin Heidelberg.

Huizingh, E. K. (2011). Open innovation: State of the

art and future perspectives. Technovation, 31(1):2 –

9. Open Innovation - {ISPIM} Selected Papers.

Isom

ott

onen, V. (2011). Theorizing a one-semester real cus-

tomer student software project course. In Jyv

askyl

Studies in Computing, volume 140. University of

Jyv

askyl

a. PhD Thesis.

Isom

ott

onen, V. (2014). Making group processes explicit to

student: A case of justice. In Proceedings of the 2014

Conference on Innovation and Technology in Com-

puter Science Education, ITiCSE ’14, pages 195–200,

New York, NY. ACM.

Jaakkola, H., M

akinen, T., and Etel

aaho, A. (2014a). Open

data: Opportunities and challenges. In Proceedings of

the 15th International Conference on Computer Sys-

tems and Technologies, CompSysTech ’14, pages 25–

39, New York, NY. ACM.

Jaakkola, H., M

akinen, T., Henno, J., and M

akel

a, J.

(2014b). Open

. In Information and Communi-

cation Technology, Electronics and Microelectronics

(MIPRO), 2014 37th International Convention on,

pages 608–615.

Kaihlavirta, A., Isom

ott

onen, V., and K

arkk

ainen, T.

(2015). A self-ethnographic investigation of contin-

uing education program in engineering arising from

economic structural change. Studies in Continuing

Education, 37(1):99–114.

Klug, B. (1976). To grade, or not to grade: A dramatic dis-

cussion in eleven parts. Studies in Higher Education,

1(2):197–207.

Kyr

o, P. (2005). Entrepreneurial learning in a cross-cultural

context challenges previous learning paradigms? In

Kyr

o, P. and Carrier, C., editors, The dynamics of

Learning Entrepreneurship in a Cross-Cultural Uni-

veristy Context, pages 68–103. University of Tampere

Research Centre for Vocational and Professional Edu-

cation.

Lewin, K. (1946). Action research and minority problems.

Journal of social Issues, 2(4):34–46.

Lin, C. P. and Chen, S.-H. (2012). The role of integra-

tion mechanism in open innovation team: An ex-

OpenResourcesastheEducationalBasisforaBachelor-levelProject-BasedCourse

ploratory study on cross-ﬁeld student team contests.

In Technology Management for Emerging Technolo-

gies (PICMET), Proceedings of PICMET ’12, pages

1953–1960.

Mezirow, J. (1981). A critical theory of adult learning and

education. Adult Education, 32(1):3–24.

Moore, R. (2000). For knowledge: Tradition, progres-

sivism and progress in education—reconstructing the

curriculum debate. Cambridge Journal of Education,

30(1):17–36.

Morales-Chaparro, R., S

anchez-Figueroa, F., and Preciado,

J. (2014). Engineering open data visualizations over

the web. In Luo, Y., editor, Cooperative Design, Vi-

sualization, and Engineering, volume 8683 of Lecture

Notes in Computer Science, pages 51–59. Springer In-

ternational Publishing.

Oakley, B., Hanna, D., Kuzmyn, Z., and Felder, R.

(2007). Best practices involving teamwork in the

classroom: Results from a survey of 6435 engineering

student respondents. Education, IEEE Transactions

on, 50(3):266–272.

Otjacques, B., Stefas, M., Cornil, M., and Feltz, F. (2012).

Open data visualization: Keeping traces of the explo-

ration process. In Proceedings of the First Interna-

tional Workshop on Open Data, WOD ’12, pages 53–

60, New York, NY. ACM.

Parlett, M. (1977). The department as a learing milieu.

Studies in Higher Education, 2(2).

Pears, A. and Daniels, M. (2010). Developing global team-

work skills: The Runestone project. In Education En-

gineering (EDUCON), pages 1051–1056. IEEE.

Sahami, M., Danyluk, A., Fincher, S., Fisher, K., Gross-

man, D., Hawthrone, E., Katz, R., LeBlanc, R., Reed,

D., Roach, S., Cuadros-Vargas, E., Dodge, R., Ku-

mar, A., Robinson, B., Seker, R., and Thompson, A.

(2013). Computer science curricula 2013.

Suri, D. (2007). Providing “real-world” software en-

gineering experience in an academic setting. In

ASEE/IEEE Frontiers in Education Conference, 37th

annual, pages S4E–15–S4E–20. IEEE.

Telikicherla, K. C. and Choppella, V. (2014). Enabling

the development of safer mashups for open data. In

Proceedings of the 1st International Workshop on In-

clusive Web Programming - Programming on the Web

with Open Data for Societal Applications, IWP 2014,

pages 8–15, New York, NY. ACM.

Tomayko, J. E. (1998). Forging a discipline: An outline

history of software engineering education. Annals of

Software Engineering, 6(1):3–18.

Tuunanen, T., Koskinen, J., and K

arkk

ainen, T. (2009). Au-

tomated software license analysis. Automated Soft-

ware Engineering, 16(3-4):455–490.

Yang, H.-L. and Cheng, H.-H. (2010). Creativity of stu-

dent information system projects: From the perspec-

tive of network embeddedness. Computers & Educa-

tion, 54(1):209–221.

Yunfei, P. and Qin, D. (2009). Cultivating the innova-

tion ability of college students in course teaching. In

Education Technology and Computer Science, 2009.

ETCS ’09. First International Workshop on, volume 1,

pages 828–831.

CSEDU2015-7thInternationalConferenceonComputerSupportedEducation