System for reproducible code environments - Dohrnii

Abstract

The reproducibility of code and the whole environment in which that code needs to be executed takes up much of the time of software engineers, computer scientists, and anyone who writes code. There are several reasons why we can not get the same results if the original data is available to us. This may be because the environment is not configured the same way as the source code author, we do not have enough knowledge of the technology used, poorly written documentation , lack of documentation, different versions of libraries, tools, development frameworks, mistakes made by one who wants to reproduce the code or there are errors in the code or errors that occurred after a certain period of time. In this paper, we introduce a new system for reproducible coding environments called Dohrnii, as a cloud-based solution. The main purpose is to save the time lost when we want to reproduce the results of a project for the first time. Every Dohrnii environment contains a video, description, instance (virtual machine), resources, and evaluation. This means that with this system it will be possible to reproduce environments in which the code needs to be executed, for that there will be a video where it will be shown how to reproduce the code, instance (virtual machine), description, additional resources, and part for an evaluation to determine if what is in the video corresponds to what is in the instance.

Full text

System for reproducible code environments -
Dohrnii
Dimitar Mileski
Ss. Cyril and Methodius University
Faculty of Computer Science and Engineering
Skopje, North Macedonia
dimitar.mileski@ieee.org
Marjan Gusev
Ss. Cyril and Methodius University
Faculty of Computer Science and Engineering
Skopje, North Macedonia
marjan.gushev@finki.ukim.mk
Abstract—The reproducibility of code and the whole envi-
ronment in which that code needs to be executed takes up
much of the time of software engineers, computer scientists,
and anyone who writes code. There are several reasons why we
can not get the same results if the original data is available to
us. This may be because the environment is not configured the
same way as the source code author, we do not have enough
knowledge of the technology used, poorly written documenta-
tion, lack of documentation, different versions of libraries, tools,
development frameworks, mistakes made by one who wants to
reproduce the code or there are errors in the code or errors
that occurred after a certain period of time. In this paper, we
introduce a new system for reproducible coding environments
called Dohrnii, as a cloud-based solution. The main purpose is
to save the time lost when we want to reproduce the results of
a project for the first time. Every Dohrnii environment contains
a video, description, instance (virtual machine), resources, and
evaluation. This means that with this system it will be possible to
reproduce environments in which the code needs to be executed,
for that there will be a video where it will be shown how
to reproduce the code, instance (virtual machine), description,
additional resources, and part for an evaluation to determine if
what is in the video corresponds to what is in the instance.
Index Terms—code reproducibility, code environment, web
platform, cloud computing
I. INTRODUCTION
The reproducibility of the code is the process by which we
will get the same results as the author of the code if we have
access to the original data [1].
It takes time to configure and set up the environment when
we try to reproduce code, to get the same results using the
original data. In many cases, we spend a lot of time so that
in the end we can not reproduce the results even though we
have the original data. This is a big problem because software
engineers, computer scientists, and anyone who writes code
spends a lot of time trying to reproduce the results. Part of
the job of software engineers and computer scientists is to try
out different projects. Many hours and even days are spent
trying to reproduce some code, which in the end may not be
reproducible. There are several reasons why we can not get
the same results as the author of the paper, or the project.
A. Code reproducibility problems
There are a number of errors that can lead to not reproduc-
ing the results. This is because the environment in which the
code is executed is not the same as the environment in which
the author of the code executed it and obtained the results.
Common problems are identified by:
•not being familiar with technology,
•not having the opportunity to take the time to learn new
technology,
•trying to see the end results, insufficient expertise in the
field.
The problem is usually found in a new programming
language, a new library, or a new software development
framework. Extensive and complex documentation, poorly
written documentation, and often incomplete or no docu-
mentation at all can be a big problem. Different versions of
libraries, programming languages, and software development
frameworks are one of the most common reproducibility
issues. Software version incompatibility is often a problem.
In order to solve this problem, the author of the source code
needs to write text documentation write requirements, Docker
File [2], or similar configuration files where the versions
will be documented. Often package management tools do
automatic version updates which can cause conflicts. The
one who wants to reproduce the code also makes mistakes.
In many cases, there are errors in the code from the start,
and we spend time trying to reproduce the results from the
code that is not working. Library support ends after a certain
period, there are errors in the library or others problems with
the libraries used. Some systems are not reproducible because
they depend on a service that may have problems and cause
errors. In all these cases, the computer scientist, the computer
engineer, and all those who write the code will spend time
without getting the same results as the author of the project.
B. Methods to reproduce the results
When we try to reproduce code we first see if the author
of the code has documentation that we can follow. If it is not
there or it is not complete, problems with reproducing the
code are possible. In such cases, we can ask for help from
the official documentation of the programming language, the
library, and the development framework. More and more
video platforms are used on which there are videos showing
software development. Forums and blogs are also used so
that we can solve a certain problem or replace part of the
code. If the problem is in the environment in which the
code is executed, which is initially set before the code is
executed over the data set, configuration files, installations,
tools, programs, integrated development environments, devel-
opment frameworks, and libraries can solve the problem, and
finally reproduce the code. If the code has documentation
that lists the libraries used and their versions and all the
details about the environment in which the code needs to
be executed, we can reproduce the results.
For some projects, library version file, such as require-
ments.txt, package.json, pom.xml, and others is sufficient.
Docker containers [2] make the environments in a container
easily reproducible. Google Colab [3], and Jupyter Notebook
80
The 19th International Conference on Informatics and Information Technologies – CIIT 2022

[4], are also platforms for reproducibility of code that is
stored and executed in the cloud. The problem arises because
the code and data are often not sufficient to obtain the same
results as those obtained by the author. The environment
in which the solution is developed is different from the
environment in which we try to reproduce the results.
C. How the Dohrnii system solves the problem
The system for reproducible code environments (Dohrnii)
tries to solve the problem with a unique combination of the
previous methods. The system is based on Dohrnii environ-
ments. Each environment is composed of a video recorded
by the author of the code using the screen recording tool on
the Dohrnii platform. The video explains how to reproduce
the results.
Short description of the Dohrnii environment, a description
of the code, the data set, and the results to be obtained at
the end. Each environment has link to the code repository.
There is a list of additional resources. This section lists links
to additional resources that are not in the code repository.
It could be a link to a data set, a link to documentation,
or some other useful link. All previous parts of the Dohrnii
environment will help us to be able to reproduce the code in a
virtual machine created by the Dohrnii system itself, and that
virtual machine is in the cloud. SSH can be used to connect to
the instance (virtual machine) [5] SSH Web client of choice
or one proposed in Dohrnii system. If the instance has a
graphical user interface, then we can connect to with Remote
Desktop [6], Web or Desktop version of Remote Desktop
[6]. Finally, in each Dohrnii environment, there is a section
for evaluation. In the evaluation section, two answers are
possible, the environment is reproducible or the environment
is not reproducible, i.e. it means that what is in the video cor-
responds to the current state of the instance (virtual machine).
This combination of screen recording, description, instance
(virtual machine), code repository, additional resources, and
evaluation section should provide a Web-based, cloud-based
solution for reproducible code environments.
The paper is further organized as follows. An overview
of similar systems for reproducible coding environments is
given in Section II. The system is described in Section III
with details on architecture, implementation technologies, and
functional description. Section IV addresses the conclusions,
including a discussion on advantages and disadvantages, and
reveals a perspective on future work.
II. RELATED WORK
This section will look at existing solutions that try to solve
the problem of reproducibility. The Jupyter Project [4] is
a project and community that aims to develop open-source
software, open standards, and interactive computing services
for multiple programming languages. The Jupyter project is
made up of several subprojects such as Jupyter Notebooks
[4], Jupyter Hub [7], Jupyter Lab [8]. Jupyter Notebook is
an open-source web application that allows you to create
and share documents with code, equations, visualizations,
and text. Usage includes data cleaning and transformation,
numerical simulation, statistical modeling, data visualization,
machine learning, and more. Jupyter Notebooks if it is hosted
can be a good platform for reproducibility. JupyterHub is a
platform for hosting Jupyter Notebooks. It is used by students,
scientists, and software developers.
JavaScript
library for
user
interfaces
Cloud
computing
SDK for
Cloud
Automation
Web
Framework
Database
Video
Streaming
Platform
Git
Repository
SSH
Remote
Desktop
Graphing
library
REST API
...
Dohrnii
Environment
Fig. 1. System Architecture
Another project based on Jupyter is Binder [9]. Binder
is an open-source web service that allows users to create
interactive, reproducible cloud environments. Some of the
technologies included in Binder are JupyterHub, Kubernetes
[10], Repo2Docker [11]. Binder works with pre-existing
community projects to create interactive versions of code
repositories that exist on sites like GitHub [12].
Another cloud system based on Jupyter projects that pro-
vides computer resources such as GPU and TPU is Google
Colab [13]. Collaboratory, or Colab for short, is a Google
Research product that allows developers to write and execute
Python code through their browser. These are Jupyter note-
books that do not require installation and have a free version
that gives free access to Google’s computer resources, such as
GPUs and TPUs. Another such project is JupyterBook [14].
Jupyter Book is a collection of Jupyter notebooks.
There are also interactive environments for executing code
in the field of Web development. Some of the client-side code
development environments are:
•CodeSandbox is an online web development editor.
CodeSandbox [15] is a tool for quickly building proto-
types, experimenting, and sharing code with other devel-
opers. Used to create static Web pages, Web applications
as a whole, or components of any Web browser device.
•JSFiddle [16] is an online service, integrated devel-
opment environment, and community for testing and
displaying user-generated code snippets. HTML, CSS,
and JavaScript code are written and shared via JSFiddle.
JSFiddle also enables the simulation of AJAX calls.
•CodePen [17] is a system for testing and displaying
code snippets created by developers. It functions as
a code editor and open-source learning environment,
where developers can create snippets of code called
”pens” and test them.
III. DESCRIPTION OF THE DOHRNII SYSTEM
A. System architecture
Figure 1 shows the system architecture with the following
components:
•Dohrnii environment - Dohrnii environment is an
abstraction of all the technologies needed to have a
reproducible project.
•Video platform - Screen recording and instructions on
how to reproduce the environment are uploaded on a
81
The 19th International Conference on Informatics and Information Technologies – CIIT 2022

Fig. 2. Dohrnii environnment - video, description, VM controls, code
repository, resources
video platform. These are external video platforms that
are not part of Dohrnii.
•Git repository - Code is versioned with Git [18] and is
hosted by external Git repository hosting services.
•SSH - SSH connections to instances (virtual machines)
are made with Web-based SSH clients, which are free
to use and open source.
•Remote Desktop - In order to be able to see the graphi-
cal user interface of the instances (virtual machines), we
need to use the Remote Desktop client. SSH and Remote
Desktop clients may not be Web-based, but Web-based
RD clients are recommended to have a fully Web-based
system.
•Visualization Library - Graph for the evaluation section
of the Dohrnii environment is ploted with the help of
visualization library.
•Javascript library - The client application is built with
Javascript library, which uses the REST API [19] to
perform basic CRUD operations, create, read, update
and delete. This library includes other packages such
as a screen capture package using the Screen Capture
API [20], an icon package, and a Bootstrap [21] CSS
framework.
•Web framework - The web framework should be com-
patible with the database. The choice of the framework
also depends on the programming language in which the
framework is written and the programming language of
the library for cloud automation, starting and stopping
instances (virtual machines) in the cloud.
•Database - The database stores user information so
that we can implement authentication and authorization.
Data for Dohrnii environments such as title, description,
evaluation section data, link to Git repository, and links
to additional resources are persisted in the database.
•Cloud computing - Cloud Instances (virtual machines)
are created, started, stopped, and cloned for the users to
reproduce the results.
•Cloud computing automation - The Automation Part
uses a Software Development Kit (SDK) for automating
cloud operations.
B. Implementation technologies
Technologies used for each of the components of the
system architecture are given in this section. As shown in
Figure 2, the Dohrnii environment contains title, video, de-
scription, instance (virtual machine) controls, code repository
and resources.
•Video platform - We use YouTube for the video
platform in the Dohrnii [22]. YouTube is a Google-
owned video-sharing platform for the Internet and social
media. The advantage of an existing platform is that
users can monetize videos through existing platforms.
The user who creates the Dohrnii environment should
upload the screen recording with the code reproducibility
instructions on the YouTube platform and insert the
link to the YouTube video when creating the Dohrnii
environment.
•Git repository - A link to GitHub is placed in each
Dohrnii environment. One Dohrnii environment can have
multiple code repositories. Other services such as GitLab
[23] and BitBucket [24] can also be used.
•Visualization Library - To display the evaluation sec-
tion shown in Figure 3 of the Dohrnii environment we
will use the Plotly [25] library for data visualization.
Plotly is an open-source library for data visualization.
Plotly libraries are available for multiple programming
languages including Python, Julia, R, ggplot2, MAT-
LAB, and FSharp. We will use the plotly.js React com-
ponent from Plotly to visualize the Dohrnii environment
evaluation section.
•Screen capture - Figure 4 shows the screen capture
interface for creating the Dohrnii environment. The
Dohrnii screen capture uses a React js component called
the React-media-recorder [26]. React-media-recorder is
a React component that can be used to record audio,
video, and screen. React-media-recorder uses the Medi-
aRecorder API [27]. The MediaRecorder interface of the
MediaStream recording API provides easy media record-
ing functionality. Created using the MediaRecorder ()
constructor. An option that is important to us is that the
recorded video is available for download.
•Javascript library - The React js [28] library is used
to create the client application. React (also known as
React.js or ReactJS) is a free, open-source JavaScript
library for building user interfaces or components. React
can be used as a basis for developing single-page appli-
cations or mobile applications. The most important Re-
act components we have used are React-media-recorder
component and Material-UI. Material-UI [29] is a simple
and customizable component library for building React
applications.
•Web framework - We have used Django for the Web
framework [30]. Django is a free web framework written
in Python that follows the model-template-views (MTV)
Fig. 3. Dohrnii environnment - evaluation
82
The 19th International Conference on Informatics and Information Technologies – CIIT 2022

Fig. 4. Create Dohrnii environment view
architecture. The framework emphasizes the usability
and ”connectivity” of the components, less code, fast
development, and the principle of non-repetition. Python
is used for settings, files, and data models. Django
also provides an optional administrative interface for
creating, reading, updating, and deleting dynamically
generated data.
•Database - PostgreSQL is used as the database in the
Dohrnii system [31]. Django has built-in support for
PostgreSQL which facilitates the process of developing
Django applications with PostgreSQL. PostgreSQL, also
known as Postgres, is a free and open-source rela-
tional database management (RDBMS) that emphasizes
scalability and is SQL compliant. PostgreSQL features
transactions with properties of atomicity, consistency,
isolation, endurance (ACID), automatic updates, mate-
rialized views, external keys, and stored procedures. It
is designed to handle a wide range of workloads, from
single machines to data warehouses or Web services with
multiple concurrent users.
•Cloud computing - The instances that are created,
started, and terminated on the Dohrnii system are AWS
EC2 [32] instances. Amazon Elastic Compute Cloud
(EC2) is part of Amazon.com’s cloud computing plat-
form, Amazon Web Services (AWS), which allows users
to rent virtual machines to run their own computer
applications. EC2 encourages the scalable deployment of
applications by providing a Web service through which
the user can upload an Amazon Machine Image (AMI)
to configure a virtual machine, which Amazon calls an
”instance” that contains the desired software. The user
can create, start and terminate instances in the cloud.
•Cloud computing automation - The AWS SDK for
Fig. 5. SSH to Dohnrii environment
Fig. 6. Remote desktop conntecton to Dohrnii environment
Python (Boto3) [33] is used to create, configure, and
manage AWS services such as Amazon Elastic Compute
Cloud (Amazon EC2) and Amazon Simple Storage Ser-
vice (Amazon S3). The SDK provides an object-oriented
API as well as low-level access to AWS services. Boto3
is a software development kit for Amazon Web Services
(AWS) using the Python programming language, which
allows Python developers to write software that uses
services such as Amazon S3 and Amazon EC2. Listed
below are just some of the features of the AWS SDK
for Python (Boto3). Manage Amazon EC2 instances, get
basic information about your Amazon EC2 instances,
start and stop detailed instance monitoring of Amazon
EC2, start and stop an instance of Amazon EC2, restart
an instance of Amazon EC2, working with Amazon
EC2 key pairs, get information about your key pairs,
generate a pair of Amazon EC2 instant access keys,
delete an existing pair of keys, working with security
groups in Amazon EC2, get information about your
security groups, create an Amazon EC2 instance access
security group, delete an existing security group.
•SSH - Figure 5 shows Sshwifty, SSH and Telnet [34]
client made for the Web. Can be deployed to our server
to provide SSH and Telnet access interface for any
compatible (standard) Web browser. Sshwifty is an open-
source Web SSH client that allows you to connect to
SSH on a remote machine without downloading addi-
tional software.
•Remote Desktop - Remote Desktop Protocol (RDP) is
a commercial protocol developed by Microsoft that pro-
vides the user with a graphical interface for connecting
to another computer over a network connection. The user
in Figure 6 uses RDP client software for this purpose,
while the instance (virtual machine) must be an RDP
server. RDP client exist for most versions of Microsoft
Windows (including Windows Mobile), Linux, Unix,
macOS, iOS, Android, and other operating systems.
RDP servers are built into Windows operating systems.
An RDP server for Unix and OS X also exists. By
default, the server listens on TCP port 3389 and UDP
port 3389. To use RDP with Dohrnii environments it
83
The 19th International Conference on Informatics and Information Technologies – CIIT 2022

is recommended to use Web RDP clients to have a
completely web-based experience.
C. Functional description
The Dohrnii system is designed to be used by a variety of
users who write code from developers, software developers,
computer scientists, electrical engineers, scientists in other
fields who write code.
Examples of use cases for the Dohrnii code reproducibility
system include:
•Users can browse the system and reproduce the code in
the environments. Developers who work in a company
can share the environments so that they can reproduce
a project. These could be projects in Web Development,
Artificial Intelligence, or other areas.
•Professors, assistants, and laboratory assistants can insert
link to the Dohrnii environment at the end of the
presentation or exercise so that students can reproduce
the results.
•Scientists can use Dohrnii environment as an open-
source environment and enable others to reproduce the
results. The link from the Dohrnii environment should be
placed on the platforms for hosting scientific papers, or
on the servers for the preprinted versions of the papers.
•Developers who ask questions on one of the software
development question platforms, where the answer with
the most votes to the question can be a link to the
Dohrnii problem-solving environment or there is an
environment that has a similar solution for the problem.
•Users can also access the Dohrnii system from existing
code repositories with a link to the Dohrnii environment.
•Content creators on one of the video platforms that cre-
ate videos in the field of computer science and computer
engineering to put a link in the video description to the
Dohrnii environment.
Use of the Dohrnii system requires basic knowledge and
understanding of SSH, Remote Desktop Protocol, SSH Tun-
neling, and Port Forwarding.
IV. CONCLUSION
A new system for reproducible coding environments
Dohrnii was introduced. The main purpose of the system is
to save the time lost when we want to reproduce the results
of a project.
The system was developed with sophisticated cloud-based
architecture supported by modern technologies.
The advantages of the Dohrnii system are that it is a
complete Web-based solution that uses cloud computing. All
the tools needed to successfully reproduce an environment
are available in the Web version. Such are screen capture,
SSH, and Remote Desktop, clients.
The biggest disadvantage of the Dohrnii system is that it
requires a large amount of resources. Because to reproduce
the environment we use instances (virtual machines) that
need working memory, storage, and processing resources.
Disadvantages of the system include the disadvantages of
virtual machines as a technology. The cost of creating and
launching multi-user instances can be very high, depending
on the cloud. The system is intended to be able to support
the concurrent creation, start-up, and termination of instances
for more users at the same time and the demand for the
resources can be high, thus the cost of operating the system.
The Dohrnii system now uses external SSH Web clients and
external Web Remote Desktop clients. This is a potential se-
curity issue for the system as instance connection information
will be shared with systems other than Dohrnii.
Possible improvements to the system include additional
development including cloud automation with SDK Boto3.
There would be a limit to the number of instances that can
be run by a single user, we can also limit the time that
each instance has from the time it is started in order to save
resources. One of the priorities would be to have an SSH
client built into the system. Additionally, SSH clients can
be deployed supporting the free and open-source. Remote
Desktop client built into the system to avoid using external
clients. Consider possible ways to improve visualization in
the Dohrnii environment evaluation section. Development of
a built-in program for editing screen recording. That program
would be Web-based and would have only basic features like
trimming the screen recording and inserting text over the
video itself if something needed to be further explained.
REFERENCES
[1] V. Stodden, “The scientific method in practice: Reproducibility in the
computational sciences,” 2010.
[2] C. Anderson, “Docker [software engineering],” Ieee Software, vol. 32,
no. 3, pp. 102–c3, 2015.
[3] M. J. Nelson and A. K. Hoover, “Notes on using google colaboratory
in ai education,” in Proceedings of the 2020 ACM conference on
innovation and Technology in Computer Science Education, 2020, pp.
533–534.
[4] F. P ´
erez and B. E. Granger, “Ipython: a system for interactive scientific
computing,” Computing in science & engineering, vol. 9, no. 3, pp. 21–
29, 2007.
[5] T. Ylonen, “Ssh–secure login connections over the internet,” in Pro-
ceedings of the 6th USENIX Security Symposium, vol. 37, 1996.
[6] R. Rahim, J. Simarmata, A. Purba, M. A. Prayogi, A. Sapta, O. Krianto,
and S. Suharman, “Internet based remote desktop using indy and socket
component,” Int. J. Eng. Technol, vol. 7, no. 2.9, pp. 44–47, 2018.
[7] L. Fern´
andez, R. Andersson, H. Hagenrud, T. Korhonen, E. Laface,
B. Zupanc et al., “Jupyterhub at the ess. an interactive python com-
puting environment for scientists and engineers,” in This conference,
2016.
[8] B. Granger and J. Grout, “Jupyterlab: Building blocks for interactive
computing,” Slides of presentation made at SciPy, vol. 2016, 2016.
[9] B. Ragan-Kelley and C. Willing, “Binder 2.0-reproducible, interactive,
sharable environments for science at scale,” in Proceedings of the 17th
Python in Science Conference (F. Akici, D. Lippa, D. Niederhut, and
M. Pacer, eds.), 2018, pp. 113–120.
[10] D. Bernstein, “Containers and cloud: From lxc to docker to kubernetes,”
IEEE Cloud Computing, vol. 1, no. 3, pp. 81–84, 2014.
[11] J. Forde, T. Head, C. Holdgraf, Y. Panda, G. Nalvarete, B. Ragan-
Kelley, and E. Sundell, “Reproducible research environments with
repo2docker,” 2018.
[12] L. Dabbish, C. Stuart, J. Tsay, and J. Herbsleb, “Social coding in
github: transparency and collaboration in an open software repository,”
in Proceedings of the ACM 2012 conference on computer supported
cooperative work, 2012, pp. 1277–1286.
[13] T. Carneiro, R. V. M. Da N ´
obrega, T. Nepomuceno, G.-B. Bian,
V. H. C. De Albuquerque, and P. P. Reboucas Filho, “Performance
analysis of google colaboratory as a tool for accelerating deep learning
applications,” IEEE Access, vol. 6, pp. 61677–61 685, 2018.
[14] D. Mileski, M. Jovanovik, and D. Trajanov, “Semantic web and data
science integration using computational books,” 2021.
[15] “Codesandbox, together.” [Online]. Available: https://codesandbox.io/
[16] JSFiddle. [Online]. Available: https://jsfiddle.net/
[17] “Codepen,” Feb 2022. [Online]. Available: https://en.wikipedia.org/
wiki/CodePen
[18] S. Chacon and B. Straub, Pro git. Springer Nature, 2014.
[19] M. Masse, REST API design rulebook: designing consistent RESTful
web service interfaces. ” O’Reilly Media, Inc.”, 2011.
[20] “Screen capture api - web apis: Mdn.” [Online]. Available: https:
//developer.mozilla.org/en-US/docs/Web/API/Screen Capture API
[21] J. Spurlock, Bootstrap: responsive web development. ” O’Reilly
Media, Inc.”, 2013.
[22] J. Burgess, “Youtube,” in Oxford Bibliographies Online, L. H. Meyer,
Ed. United Kingdom: Oxford University Press, 2011, pp. 1–1.
[Online]. Available: https://eprints.qut.edu.au/46719/
84
The 19th International Conference on Informatics and Information Technologies – CIIT 2022

[23] J. M. Hethey, GitLab Repository Management. Packt Publishing, 2013.
[24] Atlassian, “The git solution for professional teams.” [Online].
Available: https://bitbucket.org/product
[25] J. M. Perkel et al., “Data visualization tools drive interactivity and
reproducibility in online publishing,” Nature, vol. 554, no. 7690, pp.
133–134, 2018.
[26] “React-media-recorder.” [Online]. Available: https://www.npmjs.com/
package/react-media- recorder
[27] “Mediarecorder - web apis: Mdn.” [Online]. Available: https:
//developer.mozilla.org/en-US/docs/Web/API/MediaRecorder
[28] E. Elrom, “React router and material-ui,” in React and Libraries.
Springer, 2021, pp. 79–113.
[29] “The react component library you always wanted.” [Online]. Available:
https://mui.com/
[30] A. Holovaty and J. Kaplan-Moss, The definitive guide to Django: Web
development done right. Apress, 2009.
[31] B. Momjian, PostgreSQL: introduction and concepts. Addison-Wesley
New York, 2001, vol. 192.
[32] S. Ostermann, A. Iosup, N. Yigitbasi, R. Prodan, T. Fahringer, and
D. Epema, “A performance analysis of ec2 cloud computing services
for scientific computing,” in International Conference on Cloud Com-
puting. Springer, 2009, pp. 115–131.
[33] “Boto3 documentation.” [Online]. Available: https://boto3.amazonaws.
com/v1/documentation/api/latest/index.html
[34] J. Postel, “Telnet protocol specification,” in RFC 854, ISI. Citeseer,
1983.
85
The 19th International Conference on Informatics and Information Technologies – CIIT 2022