CoCalc is a cloud-based platform that provides a variety of services and tools for open science education. It allows users to create, share, and collaborate on computational documents that can run various programming languages and frameworks. In this paper, we explore the potential of CoCalc as an integrator of services for learning informatics and mathematical disciplines within the context of open science. We aim to identify the structural elements of the CoCalc environment that are suitable for these disciplines and to examine the prospects of their use. We analyze the structure of the CoCalc kernel and highlight the features that can support diferent kinds of learning activities, such as interactive coding, data analysis, visualization, simulation, testing, and assessment. We also discuss the challenges and opportunities of using CoCalc in open science education, such as accessibility, reproducibility, transparency, and ethics. We conclude that CoCalc is a promising environment that can enhance the quality and efectiveness of informatics and mathematical education by providing a rich and flexible set of services and tools.
Programming skills are essential for many disciplines, not only for computer science. Therefore,
it is important to introduce students to programming concepts and methods in diferent courses,
especially those that involve practical applications. This way, students can learn how to use
computational tools to solve problems in their fields of interest. However, learning programming
is not only about writing code, but also about collaborating with others and sharing knowledge.
This is in line with the principles of open education and open science, which are becoming
more prevalent in higher education in Ukraine. In an open learning environment, students,
researchers, and teachers are equal participants of the same information community, without
LGOBE
(S. V. Shokaliuk)
(S. V. Shokaliuk)
https://kdpu.edu.ua/personal/pvmerzlykin.html (P. V. Merzlykin);
https://kdpu.edu.ua/personal/svshokaliuk.html (S. V. Shokaliuk)
© 2023 Copyright for this paper by its authors. Use permitted under Creative Commons License Attribution 4.0 International (CC BY 4.0).
CEUR
Workshop
Proceedings
a rigid hierarchy. However, in reality, there is still a formal hierarchy in most universities.
Therefore, the role of teachers is not only to deliver the course content, but also to manage the
information environment and support students with technical issues. Moreover, teachers need
to foster a culture of cooperation and communication among all users of the digital environment
[
Another challenge in teaching programming across disciplines is the diversity of computing systems, methods, and concepts that are used in diferent fields. This makes it dificult to compare, reproduce, and communicate the results of computational tasks. It also hinders the development of common teaching standards and methodologies. Therefore, there is a need for integrating educational computer systems and research support systems that are tailored for specific disciplines. This will help to enhance the scientific component of students’ education, not only in humanities but also in technical specialties. Furthermore, such an integration will facilitate the formation of a community of students and researchers within a single information space. The single digital environment can provide such integration tools.
Science is inherently a collaborative endeavor. Researchers often work in teams to conduct experiments, and this trend has been growing in recent years. Moreover, many experiments are performed using cloud services or platforms, which require suitable tools to support the experimental activities. However, not all tools are designed to facilitate collaboration among scientists. Some tools focus on computation, but neglect the communication and coordination aspects of group work. Even if a cloud-based environment ofers some features for work or learning management, it may not address the specific pedagogical challenges of collaborative experiments. Therefore, our research aimed to explore the available tools for students to perform group tasks, conduct joint research, and share their results openly. We believe that conducting experimental research with a group of students, faculty, and researchers is a timely and important issue. We also think that it is necessary to examine every aspect of collaboration among these participants, and to identify the current problems and solutions in this area. In particular, we will present some evidence in the following paragraphs that shows how cloud service tools can be used as a means of open science.
SageMath is an open-source computer algebra system that has been widely used for research
problems related to algebra and geometry. However, it has also evolved into an open-source
cloud service that supports collaboration, and allows users to work with Python, R, Jupyter,
LaTeX, and more. Furthermore, the CoCalc cloud service enables teachers to customize their
own LMS environment. Programming, LaTeX usage, simulation – these are new skills for
mathematics students, and such environments help them develop these skills [
Klaßmann et al. [
De Assis Zampirolli et al. [
MEGUA also has a feature for adding parameterized graphs to exercises, but it does not have automatic correction for printed copies of questions, or a function for grading hundreds of users.
The problem of developing a curriculum for operations research courses has been addressed
by Vlasenko et al. [
Bobyliev and Vihrova [
Gavrilyuk [
The aim of the study is to identify the structural elements of the CoCalc environment that are appropriate to use in the educational process in the context of open science.
CoCalc (Collaborative Calculation and Data Science; mode of access cocalc.com) is a virtual online workspace (cloud-based environment) for calculations, research, authoring documents in collaboration mode.
The learning and scientific activities in the CoCalc environment involve working on a project. The elements of a project are folders and files in diferent formats.
It is through the project files that the student and/or scientist accesses the main components of CoCalc explicitly (figure 1) or through an “intermediary” (file type “X11 desktop”, figure 2).
According to CoCalc’s statistics over the last month, the most popular environment instrumental and applied components are Jupyter Notebooks, Sage Worksheets, LaTeX Documents and R Markdown Documents.
The popularity of Jupyter Notebooks is obvious. Because it is on Jupyter Notebooks that you can modeling (calculate, programming, etc.), with the functionality of SageMath or Python or R or Julia.
Before talking about the already popular tools (SageMath, Python, R, LaTeX), let’s focus on the latter mentioned, Julia.
Julia is a high-level, high-performance programming language with dynamic typing for mathematical calculations. The syntax is similar to the matlab family, the language is written in C, C++ and Scheme, it is possible to call C libraries.
Julia was designed from the beginning for high performance. Julia programs compile for eficient native code for multiple platforms via LLVM.
Julia plays dynamically, is a scripting language and has good support for interactive use.
Playable environments make it possible to play the same Julia environment every time, on diferent platforms, with pre-built binaries.
Julia uses multiple sending as a paradigm that facilitates the expression of many objectoriented and functional programming patterns. Provides asynchronous I/O, metaprogramming, debugging, logging, profiling, package manager, and more. You can create entire programs and microservices in Julia.
Julia is an open source project with more than 1,000 authors. It is provided under MIT.
But first of the stages in the development of the CoCalc is a web Computer Mathematical System (web-CMS) SageMath.
SageMath is a free open-source mathematics software system based on many existing opensource mathematical packages – FLINT, GAP, Matplotlib, Maxima, NLTK, Numpy, Pandas, Scikit Learn, Scipy, Statsmodels, SymPy, and many others. They can be accessed using a generalised language based on Python, or directly through interfaces or shells.
The available web-CMS tools of SageMath version 4.6 (the latest version before the advent of
CoCalc, even earlier than SageMathCloud) were not suficient to organize all types of learning
activities under distance learning or its elements. It was necessary either to organize training or
with the involvement of two systems – web-CMS SageMath and any system to support distance
learning, such as Moodle, or to integrate them. The first method proved to be inconvenient for
neither teachers nor students, the second method – continues to be widely used [
Since 2014, more than 80 students have completed the courses “Computer Technologies in
Research” and “Computer Mathematics” for future computer science teachers with the
additional qualification “applied programmer”. The SageMath toolkit in CoCalc became especially
popular with the advent of the ability to work on interactive Jupyter Notebooks instead of
Sage Worksheets [
In addition, future teachers of mathematics and computer science were ofered to master the tools of SageMath in CoCalc within the optional course “Using SageMathCloud in learning mathematics” (by Maiia V. Marienko), the course “Numerical Methods / Methods of Computing / Computational Mathematics”, “Discrete Mathematics”, “Operations Research”, “Mathematical Programming”, as well as to perform independent work on the courses “Linear Algebra and Numerical Systems”, “Analytical and Diferential Geometry”, “Calculus”, “Probability Theory and Mathematical Statistics”.
The mathematical packages FLINT, GAP, Matplotlib, Maxima, NLTK, Numpy, Pandas, Scikit Learn, R, Scipy, Statsmodels, SymPy, TensorFlow are known as members of the Python Scientific Computing Ecosystem or more simply Scientific Python because they provides data processing (modeling, experiment control) and visualize results for quick analysis with high-quality metrics for reports or publications.
Among the tools mentioned, the packages TensorFlow and R are of particular note.
TensorFlow is a comprehensive open source platform for machine learning. It has a comprehensive flexible ecosystem of community tools, libraries, and resources that allows researchers to advance the latest advances in machine learning, and developers can easily create and deploy machine-based applications.
R is an integrated suite of software facilities for data manipulation, calculation and graphical display. Among other things it has • an efective data handling and storage facility; • a suite of operators for calculations on arrays, in particular matrices; • a large, coherent, integrated collection of intermediate tools for data analysis; • graphical facilities for data analysis and display either directly at the computer or on hardcopy; • a well developed, simple and efective programming language (called ‘S’) which includes conditionals, loops, user defined recursive functions and input and output facilities. (Indeed most of the system supplied functions are themselves written in the S language.) R is very much a vehicle for newly developing methods of interactive data analysis. It has developed rapidly, and has been extended by a large collection of packages.
Since September 2018, almost 50 PhD candidates have been involved with the R toolkit in CoCalc and have successfully completed the Modern Information and Communication Technology in Research course.
To support cumbersome scientific calculations, there is a need to reduce the computational delay. Edge computations adopt a decentralized model that brings cloud computing capabilities closer to the user equipment to reduce computational latency. There are two types of projects in CoCalc: “trial (free) projects” and “participating projects”. Trial projects run on computers that share the same node with many other projects and system tasks. These nodes may also stop at any time, causing the current project to interrupt and restart.
Projects accepted by members are transferred to less loaded machines, which are reserved only for users who have purchased one of the proposed licenses (tarif plans). Those servers are not being restarted daily. The cluster is dynamically scaled to accommodate diferent numbers of member projects.
Work on members projects is much smoother because commands are executed faster with less delay, and heavy operations of the processor, memory and I/O work faster.
By default, free projects stop working after about 30 minutes of inactivity. This makes the calculations quite time-consuming.
There is an advanced license option to completely prevent downtime. Processes can still stop if they use too much memory, crash due to an exception, or or being restarted by the server on which they are running.
That is, for users who have purchased one of the proposed tarif plans, there are more opportunities to use edge calculations.
Also, it is possible to change the free tarif plan (default) Hub server by clicking “Reconnect”
(figure 3). To some extent, this setting may also be considered as a practical use of edge
computing [
In addition, we should mention Big Data. The complexity arises from several aspects of the Big Data lifecycle, such as data collection, storage on cloud servers, data cleaning and integration. But edge computing solves this problem, which is an essential point for working with CoCalc.
CoCalc ofers a wide collection of software environments and libraries (see tables 1-4).
A complete list of the current versions of CoCalc (1267 Python packages, 4472 R packages, 447 Julia libraries and more than 243sd files have been installed) can be obtained by using the command $ sudo dpkg --get-selections.
Detailed information on the specified in tables 1-4 and other CoCalc components (at the time of publication) can be obtained by direct link https://cocalc.com/help on the oficial website of the CoCalc project.
Implementation of research projects, term papers with the use of CoCalc involves two ways: 1. Using the individual tools presented in CoCalc. 2. Execution, writing and registration of results of educational and research work in CoCalc without involvement of auxiliary software.
At the same time, teachers and a group of students can be involved in the research project.
The IPython interpreter in the process of training future mathematics teachers can be used to develop dynamic models with semi-automatic / automatic demonstration modes.
The first way involves creating a model (models) of the phenomenon under study on a worksheet using standard controls, HTML tags, LaTeX commands and using CSS.
The disadvantages of this use are that in the process of registration of the obtained results have to involve other software: text editor, software for creating presentations, video editor (if necessary). As a result, only a certain point of the research work was performed using the CoCalc toolkit. In addition, in the process of presenting scientific findings, the student will have to demonstrate to their colleagues in addition to the presentation of the developed model using a browser (or video editor). This can be avoided by using CoCalc tools not only to perform the research part of a particular job. Therefore, it is better to use the built-in LaTeX editor as a CoCalc tool.
LaTeX is a high-quality text document program.
LaTeX is a TeX-based macrosystem that aims to simplify its use and automate many common formatting tasks. This is the de facto standard for academic journals and books, and it ofers one of the best free typography programs it has to ofer.
Performing a term paper or a thesis in the LaTeX editor, the student has the opportunity to print it, preformed on the basis of a resource such as tex PDF-document.
That is, at the same time there is a process of registration of the obtained results, calculations, presentation and presentation of the main provisions of the study (using the presentation developed in the LaTeX editor) and demonstration of the created model. The student does not need to include additional software to perform, design or present the results, because all the work is completely unified within one cloud service – CoCalc. \usepackage[utf8]{inputenc} \usepackage[ukrainian]{babel} \usepackage{sagetex} \title{Sharing Sage and LaTeX} \author{M. V. Marienko}
You can of course ofer an alternative to CoCalc – Jupyterhub and Zoom. However, they do not include the ability to synchronize with other community members in a text file, although Zoom has a basic real-time chat feature. Of course, you can ofer to integrate the Markdown hypertext into the configuration by using the Jupyter Notebook, which seemed to be the ideal solution to enable collaboration in a browser-based text document in real time using Zoom, for example in workshops. In addition, HackMD Markdown files will be available to students at any time and will be used for notes during the workshop. In this way, you can create joint documents that implement synchronous and asynchronous discussions. In addition, HackMD will provide tools for documenting group work sessions so that it is easy to share with other users. In this way, you can create templates for courses that will be used later for notes, discussion of seminar topics outside the classroom. Currently, Jupyterlab does not allow real-time collaboration on real-time collaboration due to technical limitations.
CoCalc ofers shared computing capabilities to small groups of users. It also includes basic
chat and video conferencing features. CoCalc toolkit supports student projects and group
assignments that require synchronous collaboration in computer science and math. Because
CoCalc is also based on the Jupyter Notebook, integration with individual workspaces will be
seamless, as users in the same group can easily transfer individual files between CoCalc to
both the shared workspace and their own, private instance of Jupyterlab. Using the advanced
configuration with Zoom, HackMD and CoCalc, seminars can be organized completely remotely
[
Overall, this configuration is a good starting point for the further evolution of the digital
environment and the management of a group of students to increase digital literacy in
interdisciplinary research and the teaching of computer science and mathematics. To assess the
cloud environment, it is necessary to take into account both the student’s opportunities and
interaction with them, as well as the success in achieving interdisciplinary learning goals and
the level of discussion of the content achieved in seminars. CoCalc cloud service can be
recommended to groups of students of all academic levels, from bachelor to doctoral and teachers of
various fields of science. The use of a single cloud platform has certain advantages: it will help
to form and hold regular meetings to discuss modern computational approaches in
interdisciplinary research. This creates a digital environment for developing students and researchers
that goes beyond weekly seminars. From the point of view of teaching, seminars conducted in
one case study will confirm the potential of a common information environment for teaching
computational interdisciplinary research. Thus, students with limited programming experience
or no previous programming experience during distance learning workshops will be able to
fully learn the basics of Python programming and gain skills in discussing and implementing
high-level computational models [
The evolution of the configuration of the digital environment demonstrates clear progress, which is closely linked to the requirements of pedagogical and methodological practices within the developing free economic system, students and researchers. Thus, the resulting configuration for the introduction of computational thinking and digital literacy consists of the following tools that support the necessary functions in a single digital environment: • Jupyter Notebook, which is serviced through Jupyterhub, will provide a basic environment for notes, programming and working with computational methods and concepts without the need for local installation and maintenance. • GitHub, GitHub Pages, and GitHub Classroom will be used to track file versions, create a course website as an alternative communication channel, and support the logistics of issuing and submitting course assignments. • Zoom will provide a tool for interactive synchronous social communication in distance and face-to-face learning. • HackMD is used for synchronous co-writing of hypertext documents.
• CoCalc provides collaborative real-time programming based on the Jupyter Notebook.
Ukraine has adopted a roadmap for its integration into the European Research Area (ERA-UA) by the decision of the Ministry of Education and Science of Ukraine No. 3/1-7 on March 22, 2018. One of the priorities of this roadmap is to promote the development of open science in Ukraine. Open science means making the research process transparent by publishing all its results and details on how they were obtained, and making them publicly accessible on the Internet.
The practical implementation of the open science paradigm involves [
According to Shyshkina [
A common example of open source is the large number of open source virtual learning
environments used in the academic setting. The most notable example is Moodle, which is
widely used in educational institutions [
Therefore, the introduction of open science standards in Ukraine should lead to more exchange, accountability, reproducibility and reliability of scientific materials, and afect the learning process as a whole. In the process of studying domestic and foreign experience, we identified the following benefits of using cloud services for mathematical purposes: resource saving; mobility of access; flexibility.
The use of cloud platforms and services in the educational process leads to the emergence and
development of forms of education and research organization that are focused on collaborative
learning activities, and create more opportunities for educational and research projects [
CoCalc is a cloud service that provides a virtual workspace for computation, research,
collaboration and document creation [
CoCalc [
In this cloud service [
CoCalc [
The cloud service [
CoCalc [
The cloud service [
CoCalc [
The cloud service [
Although the cloud service is fully ready for use in research [
We can conclude that CoCalc meets all the principles of open science. And CoCalc tools can be considered as open science tools that have didactic potential in the learning process.
The chronology presented in this paper shows the creation and adaptation of the digital environment based on the specific needs and practical tasks of a group of students, teachers and researchers in interdisciplinary research and education. As the digital environment is constantly evolving, the research cannot be considered conclusive. We plan to integrate the configuration of CoCalc and the curricula of individual disciplines for a deeper understanding of the learning material and to expand the means of forming professional competencies of future specialists in various fields of education and science. CoCalc tools enhance students’ ability to organize and perform teamwork by implementing a joint project task. Thus, using the cloud service improves the indicators of scientific research, makes the educational process more open, relevant to human needs and content.
Given the growing popularity of free software and the wide range of applications and services ofered by CoCalc, it is important to note that there is a need to develop teaching materials for Computer Science and Mathematics.
The use of cloud services leads to the emergence and development of learning forms that are focused on collaborative learning activities on the Internet. Cloud services should be used in Mathematics teachers training as a means of: communication; cooperation; data storage and processing, which should be the subject of further research. It is advisable to focus further research on the dissemination of open science approaches in Mathematics teachers training process.