Introducing the fundamentals of software systems to early undergraduates is a tedious journey for the instructors. It is crucial for students as it lays the foundation and establishes a perception of the upcoming computing courses. With rapid advancements in technology and considering the changes in workforce practices, it is required to upgrade the course curriculum to up-skill the students in contrast to prevailing times. This paper discusses our journey towards upgrading a decade-old introductory course on Software Systems for early undergraduate students. We present our approach towards updating the course curriculum and its delivery modes to ease learnability among early undergraduate students. We implemented mixed pedagogy methods over the years and captured student feedback to evaluate our approach.
Technology is changing at an ever-increasing pace, says
the leading research and advisory firm Gartner in its
report [
• Unable to think inside-out in terms of problemsolving as they never experienced competitive programming • Lack confidence on upskilling at later part of the
career due to academic setbacks • Switch careers into non-technical roles to escape technology productivity • Fear of programming due to lack of clarity in
introductory concepts • Fail to crack programming interviews as they are
not exposed to such evaluation conditions
course formulated in 2018 by merging and updating traditional programming and software systems courses. This course is ofered to first-year students at IIIT Hyderabad to learn tools and processes to build simple software systems.
two 4-credit workshop courses called IT-Workshop1 and IT-Workshop2. These two workshop courses spanned across two semesters during the first year of the undergraduate engineering program. These courses cover conventional knowledge on assembling and disassembling the computer hardware, installing operating systems, technical aspects of hardware-softwareifrmware integration, SHELL programming, and web technology concepts focusing on HTML, CSS, and Javascript. This course was designed in the early 2000s to provide broader exposure to freshman undergraduate students on hardware and web technology. These courses are ofered for about two decades. Instructors followed collaborative learning and hands-on-based pedagogical methods. With time, the scope and scale of the technology revolutionized the face of software systems in practice. The curriculum and pedagogy approach of traditional courses became outdated. Thus there was a need to upgrade existing course curricula to meet the expectation of current trends of software systems in practice.
Curricula Changes: Considering the trends of technological innovation and software adoption in the IT market, the ‘Introduction to Software Systems’ course is designed. This course aims to replace the 4-credit IT Workshop-I and IT Workshop-II with a regular 2-credit course to provide foundations on programming and technology. Our approach to the course is ‘Practice-Theory-Practice’ to build software systems while teaching concepts. The scope and extent of the course concepts are discussed below: • Software Engineering: Software engineering concepts, flavors of programming languages, concepts on object oriented programming, networking (OSI layers, Network appliances, Wired and Wireless communication) and cyber security (OWASP Top 10, Static testing and penetration testing).
The contents of this introductory software systems are finalized after a thorough review. These curriculum contents are gauged based on their relevance and impact on upcoming graduation courses. The foundation concepts of an operating system and Linux commands are practiced first as it becomes the basis for programming. Gradually, we introduced web technologies as it helps students to select desired elective courses on the web and be ready for internship. Considering the increased interest of AI and machine learning, python programming concepts are introduced to students.
Course Delivery: This course is covered for about
eleven weeks during an academic semester. There
are two one-and-half hours instructor-led classroom
sessions per week and three hours of lab sessions per
week led by teaching assistants. Each instructor-led
class has a live-programming hands-on session with
a small class activity. This class activity is conducted
to understand the learning levels and engagement of
students during the classroom session. There are various
graded activities like bi-weekly code assignments to
test and explore the concepts through self-learning,
Homework to investigate and knowledge discovery from
existing resources, Surprise quiz to test their learning
consistency. We also conducted mid-term and final
term papers, which tests students’ programming skills
through an automated test evaluation system.
• OS Concepts: Foundation concepts on operating
system, scheduling, memory management and Pedagogy: Pedagogy is a method or a practice of
Hardware-Software integration. teaching, especially as an academic subject or theoretical
• SHELL: SHELL command-line interface and intro- concept. As the curriculum contains varied concepts, we
duce basic Linux commands to interact with oper- adopted diferent pedagogical approaches to teaching
ating system. We also cover SHELL programming various concepts [
modes - classroom-based assessments, online-code assessments, and Self-learning assignments.
This section discusses our attempts to track and record
student feedback about the upgraded course content.
1https://www.codechef.com/
2https://www.hackerearth.com/
• All classroom-based assessments are debugging
assignments where we provide defective code Survey Setup: We intended to investigate how
snippets, and we ask students to suggest correc- freshman computer science students feel when exposed
tions to address the defects through visual inspec- to software systems and develop software through this
tion. These assessments include SHELL, HTML, survey. We are interested in their feedback regardless of
CSS, Javascript, SQL, and Python program snip- their skillset and expertise. We captured data through a
pets. Some other classroom-based assessments survey called - ‘Questionnaire for Freshman Under-Grad
include multiple-choice questionnaires with au- Students’ Experience towards Programming.’ [
line code assessment platforms like CodeChef1 and HackerEarth2. We publish the list of questions on this platform and provide certain test cases to address the motive of the provided code question. These assessments are tailored to the individual students as the test cases for evaluation are personalized. In some cases, the questionnaire is generic with specific personalized inputoutput test cases for evaluation to avoid copy cases. In some cases, we created multiple diferent • First section of the survey requests participant to self-declare their current skillset and levels of participation in programming before taking up this course. • In the second section, we asked participants about the dificulty, complexity, understandability, and intellectual efort they spent solving the tasks assigned to them during this course. The captured responses are based on a Likert scale ranging from ‘Strongly Agree’ to ‘Strongly Disagree’ • In the next section, we used the Scale of Positive
and Negative Experience (SPANE) [
Fig 2 illustrates the % time spent by the users during the participation in this course. We observed that most students spent more than 50% of the time during their participation in this course. However, certain students tend to spend only 5% and 10% of the time. On average, we observed that students are proactive and are focused on participating in the course. We also asked students Participant Demography: We reached out to students who opted for this course to seek their feedback and challenges. We obtained feedback post-completion of the course through a survey questionnaire. We recorded Figure 2: Count of students spending % time on coursework this data for four semesters, i.e., Spring Semester 2018, Spring Semester 2019, Spring Semester 2020, and Spring Semester 2021. During Spring Semester 2018 - 221 about their experiences while performing tasks assigned students took this course, and 179 participated in the during this course with the Likert scale for agreement survey. In the Spring semester of 2019, 202 students took ranging from Strongly Agree to Strongly Disagree. Fig 3 this course, and 198 students participated in the study. illustrates the student responses aggregated across years, In spring semester 2020, 222 students have taken this i.e., between 2018-2021. Following are the questions and course, and 188 students participated in the survey. In details about the student observations. the spring semester of 2021, 211 students took this course, and 159 students participated in the survey. Overall, we Q1: The tasks were dificult to answer - 39% of reached out to 856 students in four semesters, and 724 of the participants were neutral about the dificulty of them provided their feedback. Out of all respondents, answering the tasks. On the other hand, 33% of the 36% of them are female, and the rest are male participants. participants found it was not easy to answer.
years and illustrated it in this section. We asked students to self-declare their programming skillset before starting this software system course as part of the survey. Fig 1 illustrates the count of participants and their skillset before joining this course grouped as Basic users - with
46% of the participants were neutral about task complexity. However, 25% of them found the tasks to be complicated.
Q3:
found the tasks to be challenging. However, the rest participants are not confident about working on SQL. of them are neutral and disagree about the tasks being After interviewing the participants, we observed that challenging. students are not satisfied with working with python and Q4: The tasks were easy to work on - 49% of the SQL despite thorough lab activities for practice. Also, participants disagree about the tasks being easy. in terms of python and SQL - our goal was to provide a preliminary understanding of these programming Q5: The contents of the tasks were easy to un- concepts. Students will eventually learn SQL in detail derstand - 35% of the participants agree that the tasks as part of their future semester full-course on database were easy to understand and comprehend. systems and python programming as part of their future semester full-course on machine learning. This course Q6: The tasks were easy to solve - 46% of them will create a learning rigor on making the acquaintance disagree that the tasks are easy to solve. about these programming languages and eventually help them excel in their future studies and career.
tasks - 34% of the participants disagree that they spend less amount of time answering the tasks.
rectly - 49% of the participants have claimed that they have not tried hard to answer the tasks correctly. They have spent a significant amount of time while solving the tasks.
of the participants have claimed that they tried hard to answer the tasks correctly.
swering tasks - 45% of participants strongly agree, and 38% of them claimed that they made an intellectual efort while answering tasks. They were serious about tasks and solved them with a thorough review.
tasks - 40% of them strongly disagree, and 39% of them disagree that they have not responded to tasks with focus.
42% of strongly agree, and 45% of them agree that they did their best to solve the tasks.
Curriculum Design: Designing the course contents was challenging for instructors as the concepts involved in this course are elaborate. It was tedious for instructors to plan concluding one concept and switching to another. It required a lot of preparation and more examples to position a concept and introduce another. As it was a 2-credit course, we tried to make the concept short and simple for students to consume.
than 80% of the score during our first-course delivery.
With improvements in our pedagogy approaches, we observed that 78% of students had achieved more than 80% in our second-course delivery. During our third edition of this course, 88% of the students have scored more than 80% of the score. We observed that the students could consume good content if delivered with a better pedagogy organization.
Students’ Observations: We received positive Fig 3 illustrates the data captured across years. feedback from students on course content and execution. We observe participant responses group by Likert scale We learned that students faced practical challenges while on the y-axis and question identifier on the x-axis. participating in surprise tests and In-class activities due The circle in the gfiure indicates the response rate of to technical issues. However, we could address them in a the participant’s responses with percentages labeled case-by-case manner. While practicing for final exams, in the center. We also asked participants about their the majority of students have approached us for practice confidence levels while learning respective technologies programming exercises. We have provided them with post-completion of this course. Fig 4 illustrates the many practice programming problems ranging from confidence levels of working on technologies like easy to dificult to test their programming ability and SHELL, HTML, CSS, Javascript, SQL, and Python. If we prepared them for their final programming exam. We carefully observe Fig 4, most participants are confident also observed that a few slow learners could not catch and moderate about using a particular technology. up with the pace of class lectures and lab activities. We However, 47% of participants across the years are not scheduled special TA hours for personalized training confident about working with python, and 39% of the and helped come out of fear of programming.
and relevant assessments together is very crucial. As there are multiple concepts to be covered during the course, it is required to sequence the classes in a particular way to avoid complexities and confusion among students. There is a possibility of pressure on students to learn all these concepts in a short time. However, course instructors and teaching assistants should address students’ doubts on concepts in all available platforms. Scheduling TA hours with students is recommended to review and understand their learning levels and bring them up to speed.
tion was a dificult task for course instructors as we
had to weigh students’ learning patterns and plan them
carefully. We could not opt for too challenging or too
Pedagogy Methods: We initially practiced a few Researchers have conducted various studies and
pubpedagogy methods that were too challenging to execute. lished various experience reports on introducing new
As the enrollment was too high, planning one-one courses to freshman undergraduate students. An early
learning sessions or evaluation sessions were dificult study on course upgradation by Albert Crawford shows
to manage. Initially, we failed to meet the expectations that computing education should be reviewed and
conof students in the one-one evaluation sessions. Then stantly updated based on current trends [
This model course contains general concepts of those
times required for novice engineers to scale up for
future studies. Celina et al. [
Brown et al. [
This paper shares our experiences on formulating a twodecade-old introductory course on software systems for freshman engineering students with upgraded course curricula. We followed diferent pedagogical approaches to teach various concepts to students. We also conducted a multi-level assessment process to assess student performance during the tenure of this course. We captured a student feedback survey to understand students’ learning experiences. The feedback showed that the learning experience was unique and diferent. It helped them understand various concepts quickly. With a typical semester duration of four months, we encourage instructors across India to adopt this course curriculum to educate their freshman undergraduate students about an introductory course on software systems. It will eventually aid them in understanding more complex subjects without hassle.