Authors Raimond-Hendrik Tunnel Ulrich Norbisrath
License CC-BY-4.0
Journal of Education and Learning; Vol. 11, No. 5; 2022
ISSN 1927-5250 E-ISSN 1927-5269
Published by Canadian Center of Science and Education
A Survey of Estonian Video Game Industry Needs
Raimond-Hendrik Tunnel1 & Ulrich Norbisrath1
1
Institute of Computer Science, University of Tartu, Tartu, Estonia
Correspondence: Raimond-Hendrik Tunnel, Institute of Computer Science, University of Tartu, Tartu, Estonia,
Narva mnt 18, 51009, Tartu, Estonia.
Received: June 2, 2022 Accepted: July 19, 2022 Online Published: July 29, 2022
doi:10.5539/jel.v11n5p183 URL: https://doi.org/10.5539/jel.v11n5p183
Abstract
Designing a video game design and development curriculum in higher education is a challenging task.
Information about the needs of the respective industry certainly helps. In this paper, we have surveyed Estonian
video game development companies to determine their current needs when it comes to knowledge areas,
software tools, languages, abilities, and contextual fluencies. The survey is based on a similar survey conducted
a decade ago and this paper compares the current results with those found earlier.
Compared to the prior survey, we have found significant differences in the rated importance of knowledge in
optimization, version control technologies, the C, C++, and C# programming languages, and the time
management ability for video game development companies looking to hire university graduates. We have also
extended the previous survey to include a contemporary selection of game design and development tools. Based
on that, we have determined a strong need for graduates with skills specifically in Unity and Unreal Engine game
engines, Photoshop raster image editing software, and Git version control software.
While most of our results are largely consistent with the previous research, our added survey items like visual
languages and game engines bring the results to the modern context. This allows curriculum designers and
managers to see the differences regarding the landscape of industry needs for their graduates and thus make more
informed decisions in their work.
Keywords: contextual fluency, curriculum design, game design tools, game development tools, game engines,
higher education, programming languages, skills, video games, video game industry
1. Introduction
Over the years, the video game industry and video game curricula in higher education institutions have become
increasingly common. Popular game engines such as Unity and Unreal Engine make video game development
available for everyone with sufficient programming skills. There is very good, established literature as well as
practices for designing video game user experiences (Rogers, 2014; Salen & Zimmerman, 2003; Schell, 2008;
Swink, 2008). Publicly available recordings of talks from conferences such as the Game Developers Conference
make even the bleeding edge techniques available for everyone.
While readily accessible, truly learning and mastering video game design and development takes time and effort.
Higher education institutions offer many Bachelor’s and Master’s level curricula in different aspects of video
games (Valentine, 2014), ranging from programming and development, media and design, artistry, marketing,
business, and entrepreneurship, to game studies and analysis (Bouchrika, 2021). No matter the path, these fields
are not easy and require dedication from the students.
Thus, when designing such a curriculum, it is essential that the skills and knowledge it provides form a useful set
for the graduates. They need to be well prepared to continue their studies at the next level and find
junior-position employment in the industry. Balancing or focusing the curriculum among these complexions of
the video games field can be quite challenging for the curriculum designer (Mateas, 2007).
At the University of Tartu, we are deliberating on a new video game designer-developer Bachelor’s level
curriculum. We have chosen specializations in video game design and development as these are of primary
importance in making games. Some of our current computer science curriculum courses can be used in the
development specialization. However, including the teaching of video game design is something we feel very
passionate about as a video game with mediocre design but excellent implementation would still provide just a
183
jel.ccsenet.org Journal of Education and Learning Vol. 11, No. 5; 2022
mediocre experience.
Our experience in teaching video games comes from running elective courses and conducting thesis supervision
in our existing computer science and software engineering curricula. It has become obvious that students in such
general studies make a lot of extra effort in their final year to prepare themselves for working in the video games
field. And even doing so will, at times, still leave them inadequate in the eyes of the respective companies.
When designing our new video games curriculum, we want to make sure that the three years of study will be
efficient for both students and educators. To make informed decisions regarding curriculum design, we need to
know what the industry expects from our graduates. As the video games industry in Estonia is relatively young,
it is unclear if the learning outcomes and practices from countries with well-established industries would work
here or not. As such, we surveyed the video games, computer graphics, and virtual and augmented reality
companies to map out their expectations.
2. Background
There are many success stories of curriculum design in different ecosystems and with different purposes. For
example, Mikami et al. (2010) describe their very practical 4-year curriculum developed and run at the Tokyo
University of Technology. Fachada and Códices (2020) propose a top-down interdisciplinary curriculum for the
Portuguese private university Universidade Lusófona. Kenwright (2016) has developed a holistic curriculum,
where the program provides a wholesome learning experience, giving both a broader understanding and the
essentials required in the field.
These examples of innovation in curriculum design certainly provide insight and offer options to mix and match
in designing a new curriculum in another ecosystem. However, they do not describe a general baseline upon
which to build it. This becomes apparent even in the details, for instance, what tools or languages are being
taught. Kenwright (2016) mentions that they do not focus on one language so as to give the students better
problem-solving and learning skills. They mention, as examples, the Python, Java, C++, and HTML languages.
However, it is unclear if these are the languages also valued by the industry and, if so, how much. A similar
example can be found in the paper about the curriculum in Tokyo (Mikami et al., 2010), where they describe
using Maya when teaching students to create 3D graphical assets. This begs the question if Maya is the best
choice for the purpose. Do the industries value the graduates’ skills with Maya, would they prefer that they knew
another tool, like Blender, instead, or does this particular choice matter to them?
An extensively referenced document for the baseline of a video games curriculum is the IGDA Curriculum
Framework from 2008, developed by the International Game Developers Association (IGDA, 2008). The
framework, albeit more than a decade old now, lists nine different core topics with many sub-topics under each.
Fields range from all the different aspects of video games mentioned earlier (game studies, development, design,
business, etc.). The document is extensive, but there is no ranking of the different sub-topics. For example,
databases and machine architecture are on par with software design patterns and game logic scripting languages.
It is unclear how much focus these fields should have in a video games curriculum.
For a better overview, McGill (2009) has made a thorough survey and comparison of both industry needs and
existing emphases in curricula in higher education institutions. They had conducted content analysis on
advertisements for game developer positions (McGill, 2008). From that they developed a survey, which
consisted of five categories: 1) Knowledge Areas, 2) Languages, 3) Software Tools / Environments, 4) Abilities,
and 5) Contextual Fluencies. They then had game development companies and higher education institutions with
a game developer field in the US and Canada rate different items in these categories on a 5-point Lickert scale.
This comprehensive survey not only allowed to identify mismatches between the industry needs and curriculum
focus but also to acquire informative overviews of both individually.
The downside of the survey by McGill is that it is also more than a decade old now and was done before the
popularization of game engines (Andrade, 2015). One decade is sufficient time for the landscape of a new field
like video game development to have changed considerably. Furthermore, McGill surveyed the companies in the
regions of the US and Canada where game development was an established field. It is possible that in
ecosystems where the field is still in the starting phase, the needs of the industry will be considerably different.
Thus, a comparable survey made in Estonia seemed like an important and potentially very informative first step
in designing a new curriculum.
3. Method
We created a questionnaire to collect the data and analyze the needs of Estonian video game, computer graphics,
and virtual reality development companies. The questionnaire was based on the survey conducted by McGill
184
jel.ccsenet.org Journal of Education and Learning Vol. 11, No. 5; 2022
(2009) with some of the response options modernized. We also asked about specific tools used in these
companies and to what extent the skills in these specific tools are necessary for their applicants.
3.1 The Questionnaire
Similar to the previous study, we asked about Knowledge Areas, Languages, Software Tools / Environments,
Abilities, and Contextual Fluency. We mostly used the same items for each category as McGill, which were
established in their previous paper (McGill, 2008) based on real-world job applications. The questionnaire asked
to rate the importance of each item on the Lickert scale from 1 (Strongly disagree) to 5 (Strongly agree). As such,
direct item-wise comparisons of the results could be made. The questionnaire was in English.
However, during pre-interviews with university colleagues and industry practitioners, it became obvious that
some of the items from the 2009 survey had become outdated. Furthermore, the ecosystem of different software
tools and environments has grown considerably. Thus, we made a few changes to the original item lists.
Naturally, the added or removed items would no longer allow for individual comparisons for these specific items.
Most items did remain unaffected and thus most results do remain comparable. The changes are as follows:
• Added Software Testing, Video Game Analysis, and Video Game Design to the Knowledge Areas
category.
• Added GLSL / HLSL, Go, JSON, Rust, TypeScript, and Visual Languages to the Languages category.
• Changed the HTML/DHTML/XHTML item to just HTML.
• Removed OpenGL from the Languages category because it is not a language.
• Removed ActionScript from the Languages category because the Flash technology from Macromedia
(later Adobe), which mainly used that language, had become obsolete in the interim.
• Removed Ceramics from the Contextual Fluency category as it was found the least important in the
original survey, and we deemed it unnecessary in this context.
The exception is the Software Tools / Environments category. Based on the input from pre-interviews, that
category was completely remade. Instead of individual tools, we opted for the following more general items:
• 3D content creation tool (Blender, Maya, Houdini, 3DS Max, Modo)
• Audio editor (Audacity, …)
• Database server (SQL, NoSQL)
• Digital audio workstation (Ableton Live, Audition, FL Studio, REAPER)
• Game engine (Unity, Unreal Engine, Godot, GameMaker Studio, Open 3D Engine, CryEngine,
Construct 3, Source 2)
• Graphics API (DirectX, OpenGL, Vulkan, WebGL)
• IDE (Visual Studio, Visual Studio Code, Monodevelop, Rider)
• Project management tools (Confluence, Jira, Trello)
• Raster graphics editor (Photoshop, PaintShop Pro, Affinity Photo, Krita, GIMP)
• Software runtime (.NET Framework / Mono, Java, Node.js)
• Texturing tools (Substance 3D, Material Maker, ArmorPaint, Quixel Mixer)
• Vector graphics editor (Affinity Designer, Adobe Illustrator, Inkscape)
• Version control technology (Perforce, Git, SVN, Mercurial, Unity Collab, Plastic SCM, Dropbox,
Google Drive)
• Video editor (DaVinci Resolve, Adobe Premiere, OpenShot, Kdenlive)
• Visual effects tools (After Effects, Nuke, DaVinci Resolve).
The questionnaire proceeded with individual questions about all the specific tools previously listed in brackets
for each item (except for the Database server). In these questions, the participants were asked to mark for each
category the tools used in the company. Next to each question, there was a secondary question “How important
are the skills in these specific tools for a candidate?” with the following options:
• It is OK if they do not have skills in these or alternative tools.
• It is OK if they have skills in alternative tools.
185
jel.ccsenet.org Journal of Education and Learning Vol. 11, No. 5; 2022
• They need to have skills in these specific tools.
The answers to these questions would allow us to home in on what specific tools are required in the given
companies and, thus, what should be used when teaching the curriculum. Among the last questions, we asked
how many new graduates the company would be willing to hire every year, assuming they meet their recruitment
criteria. That number would allow us to quantify the previous answers and make sure that we include in our
teaching the technologies in which skills are needed in the industry.
3.2 The Participants
The participants for the survey were found from six different sources. The first source was the website
shipped.ee (Anonymous, 2022), which is a privately managed website listing released Estonian PC games and
their developer companies. From the list, we filtered out companies that had released a video game during the
last 10 years. That strategy excluded companies that had only been active in game development by releasing a
relevant product more than a decade ago and thus most likely were irrelevant for the current job market. The
second source was a members list of the Estonian Virtual and Augmented Reality Association (EEVR, 2022).
From that list, we looked for private companies that develop or publish virtual reality software themselves. This
excluded labs from universities, museums, and government institutions. The third source was a list of gaming
startup companies in Estonia made by Mark Kendall and published on beststartup.eu (Kendall, 2021). The fourth
source was the list of Top Game Development Companies in Estonia on goodfirms.co (GoodFirms, 2022). The
fifth source was a short list of collaborating companies that the Computer Graphics and Virtual Reality Study
Lab at the University of Tartu had listed on their webpage (CGVR Lab, 2022). The final source was the member
list of GameDev Estonia, which is a relatively new grouping organization representing many Estonian video
game development companies (GameDev Estonia, 2022). These sources also included newly established
companies, which were important to include as they represent the field from the perspective of startups.
The companies on the lists from these six sources were then further researched. We looked to see if the company
had a valid entry in the credit registry E-krediidiinfo (http://e-krediidiinfo.ee) and if they had some internet
presence to indicate that they are still an active company. In total, there were 70 companies deemed potentially
active and thus valid participants of the survey.
3.3 The Survey
The chosen participants were sent a link to the survey to their official email found in the official business entry
in E-krediidiinfo credit registry. If another email address was found on the company website or similar source,
the given address was used in combination with the official one. The first email was sent on 4th of April, and if
the company had not participated in the survey yet, the follow-up emails were sent on 18th and 26th. The emails
explained that the survey is used to develop a Bachelor’s level video game designer-developer curriculum. The
participants were offered a free overview of the survey results. They were also informed that answering the
questionnaire indicates that they allow their answers to be used anonymously in research. The data were
collected for one month.
4. Results
In total 28 different companies answered the survey. A few explicitly declined as they did not consider
themselves as the correct target audience. A few also filled the survey multiple times, in which case we
contacted them and agreed on which of the answers they deemed the most accurate.
4.1 Knowledge Areas
Compared to the results from the survey by McGill (2009), there were a few noticeable differences (see Figure
1). We performed the Welch’s t-test of unequal variances to determine if an item was rated statistically different
(α=0.05). The differences were found with Optimization (average 0.81 points higher in our survey, p=0.0017)
and with Version Control Processes (average 0.73 points higher in our survey, p=0.0140).
186
jel.ccsenet.oorg Journal of Edducation and Leearning Vol. 11, No. 5; 2022
Figuure 1. Our and M
McGill surveyy results for thee Knowledge A
Area category
Our addedd items Video Game Design,, Video Game Analysis, andd Software Tessting were rateed with averag ges of
4.36, 3.96,, and 3.93 resppectively. Ordeering by the mean, these resuult in the 2nd, 99th, and 10th rranks.
4.2 Languaages
In the Lannguages categoory we observeed several diffferences (see FFigure 2). Therre was an increase in the ave erage
importancee of JavaScriipt (+0.80 pooints, p=0.0311) and C# (+ +0.75 points, p=0.031), whhile a decreasse of
importancee was observeed with C++ ((-1.07 points, pp<10-4), C (-00.98 points, p= =0.008), and X XML (-0.73 po oints,
p=0.043). It is worth notting that while the largest deccrease was witth C++, it still ranked third inn our results.
187
jel.ccsenet.oorg Journal of Edducation and Leearning Vol. 11, No. 5; 2022
F
Figure 2. Our aand McGill surrvey results forr the Languagee category
Our addedd items Visual Languages, JS SON, GLSL / H HLSL, TypeSccript, Go, and Rust were rateed with averag ges of
Ordering by thee mean, these result in the 2nd, 4th, 13th, 15th,
4.07, 3.75,, 2.64, 2.50, 2..29, and 2.07 rrespectively. O
17th, and 118th ranks.
4.3 Softwaare Tools / Envvironments
As the Sofftware Tools / Environmentss category waas completely reworked for our survey, a direct comparison
with the MMcGill results cannot be madde. It is still nooteworthy thatt in our resultss the Game Enngines item haas the
first rank w
with a mean off 4.68 and a reelatively smalll standard deviiation of 0.6700 (see Figure 33). Perhaps witth the
exception of Director anda Flash, thee survey by M McGill did noot feature gam me engines att the time. Th his is
understanddable as, while there existedd some game engines like GameMaker, a more substaantial rise of game g
engines caame in the midd-2010s (Andraade, 2015).
Figuure 3. Our survey results for tthe Software T
Tool / Environm
ment category
188
jel.ccsenet.oorg Journal of Edducation and Leearning Vol. 11, No. 5; 2022
In our survvey, all the 288 companies annswered that thhey would eveery year hire inn total 53 empployees who would
w
f a video gaame designer / developer pposition. We uused that numbber of annual new
fit their hiiring criteria for
employeess for each com mpany to scalle the answers to the questtions that askked for the usee and necessity of
individual tools. This givves a clearer picture of the acctual needs in the job markett (see Figure 44).
Figure 4.. Usage and skkill expectationns for individuaal tools in the surveyed comp
mpanies, scaled by the number of
estimatted annual empployment posiitions for the juunior designer / developer joob
We found that the most specific needs were in the gaame engines U Unity (43/53) aand Unreal Enggine (29/53). These
T
were follow
wed by the rasster graphics editor Photoshoop (12/53) andd version controol software Giit (11/53).
4.4 Abilitiees
The Abilitties category (see Figure 5) had only one statistically siignificant diffeerence. The O
Organization / Time
T
<10-4).
Managemeent item was raated higher (+00.84 points, p<
189
jel.ccsenet.oorg Journal of Edducation and Leearning Vol. 11, No. 5; 2022
Figure 5. Our and McGill suurvey results fo
for the Ability category
4.5 Contexxtual Fluenciess
There wass also only onee statistical diffference in thee Contextual F
Fluencies categgory (see Figuure 6). This wa
as the
item Sculppture, which was
w rated higheer in our surveyy (+0.71 pointss, p=0.017).
McGill survey rresults for the Contextual Fluuency categoryy
Figuree 6. Our and M
5. Discusssion and Concclusion
The results of our surveyy are in large part similar too the results byy McGill from m 2009. This leeads us to conc
clude
that the neeeds of the beginning Estonnian video gam me developmennt industry aree comparable tto the needs of o the
well-established industrries in the US S and Canada. The implicattion of this iss that graduatees of a curriculum
designed aafter these resuults could find employment w with industriess in different sttages of growthh.
The new findings are somewhat exxpected. In thhe Knowledgee Areas categgory, the rise of importanc ce in
Optimizatiion could be due
d to the rise in the market share of mobile games (Walllach, 2020). IIt could also be
e due
to the fact that developerrs use game enngines not beinng careful withh performance management as the engine hides
h
a lot of thee underlying coomplexity from
m them.
The imporrtance of the Version
V Controol Processes wwas also signifi
ficantly higher.. This could be due to the riise of
the populaarity of versiion control teechnologies, nnamely Git, wwith the launcch of GitHubb in 2008 and its
overwhelm
ming success (K Kashyap, 20200).
It is notable that our addded items Video Game Desiign, Video Gaame Analysis, and Software Testing were rated
quite impoortant. We ackknowledge thatt the original sstudy focused on the video ggame developer position and d our
study extended this to a designer posittion. Items likee Video Gamee Design mightt be importantt for a designerr or a
designer-ddeveloper, but not necessarilyy for a mere ddeveloper. How
wever, as many curricula aree a mix of designer
and develooper studies (BBouchrika, 20221; IGDA, 20008), then we ffeel this choicee to be justifieed here. We think a
190
jel.ccsenet.org Journal of Education and Learning Vol. 11, No. 5; 2022
video game designer should be able to write a quick prototype code and understand what is technically possible
for the developer to implement. The developer should also understand game mechanics and user experience
design and be able to clearly grasp the designer’s intent. Thus, these two roles are highly coupled in practice.
The rise of C# and possibly the fall of C++ in the Languages category is most likely due to the great popularity
of the Unity game engine, which uses C# as its language. The increase in the importance of JavaScript is
possibly due to the advancement of web technologies. However, the importance of web development itself for a
video game designer / developer was still rated quite low. Out of our added items, Visual Languages was rated
quite important. This is likely due to the increase in the popularity of Unreal Engine, but also because of visual
languages in other popular technologies, for example, Unity’s Shader Graph, and content creation tools like
Blender and Arcweave. On the other hand, the shader programming languages GLSL / HLSL were rated quite
low in importance. This makes sense as game designers and regular game developers are not likely to program
computer graphics in a shader. That would be done by game developers with an interest in computer graphics, or
more accurately, computer graphics programmers.
From the answers to the questions about individual tools (Figure 4), it is evident that the game development
companies need designers and developers who have specific knowledge about the Unity and Unreal Engine
game engines. There were also many answers that listed both engines together, indicating that both are in use in
their company, and skills in both are sought after.
There was a significant number of answers indicating the need for skills specifically in Adobe Photoshop. This
implies that having skills in alternative (and cheaper, but similarly capable) tools like Affinity Photo would be
insufficient for a significant proportion of the jobs.
In the Ability category, the item Organization / Time Management was rated more important than before. This
could be due to a more widespread access to the internet, as more work is done remotely and thus employees are
also expected to be more self-managed.
The results of our survey give a revised perspective on the needs of the video game industry compared to the
previous study by McGill made over a decade ago. These results can help curriculum designers make more
informed choices when establishing the learning outcomes and the use of specific tools in video game designer
and developer curricula.
Acknowledgments
This research has been funded by the European Social Fund. We are grateful to all the video game development
companies who participated in our survey and we hope our research helps you as well as curriculum designers.
References
Andrade, A. (2015). Game engines: A survey. EAI Endorsed Transactions on Game-Based Learning, 2, 8.
https://doi.org/10.4108/eai.5-11-2015.150615
Anonymous. (2022). Shipped.ee. Retrieved from http://shipped.ee/
Bouchrika, I. (2021, June 25). Video Game Design Degree Guide: 2022 Costs, Requirements & Job Opportunities.
Research.Com. Retrieved from https://research.com/degrees/video-game-design-degree
Computer Graphics and Virtual Reality Lab [CGVR Lab]. (2022). Companies. The CGVR Lab. Retrieved from
https://cgvr.cs.ut.ee/wp/index.php/companies/
Estonian Virtual and Augmented Reality Association [EEVR]. (2022). Estonian Virtual and Augmented Reality
Association. EEVR. Retrieved from https://eevr.ee/
GameDev Estonia. (2022). Database. GameDev Estonia. Retrieved from https://gamedevestonia.ee/location/
GoodFirms. (2022). Top Game Development Companies in Estonia 2022. GoodFirms. Retrieved from
https://www.goodfirms.co/game-development-companies/estonia
IGDA. (2008). IGDA Curriculum Framework 2008. Retrieved from
https://docplayer.net/1868720-Igda-curriculum-framework.html
Kashyap, N. (2020, March 4). GitHub’s Path to 128M Public Repositories. Medium. Retrieved from
https://towardsdatascience.com/githubs-path-to-128m-public-repositories-f6f656ab56b1
Kendall, M. (2021, June 28). 33 Top Gaming Startups and Companies in Estonia (2021). BestStartup.eu.
Retrieved from https://beststartup.eu/33-top-gaming-startups-and-companies-in-estonia-2021/
Kenwright, B. (2016). Holistic game development curriculum (pp. 1−5). SIGGRAPH ASIA 2016 Symposium on
191
jel.ccsenet.org Journal of Education and Learning Vol. 11, No. 5; 2022
Education. https://doi.org/10.1145/2993352.2993354
Mateas, M. (2007). Design Issues for Undergraduate Game-Oriented Degrees. Retrieved from
https://www.semanticscholar.org/paper/Design-Issues-for-Undergraduate-Game-Oriented-Mateas/c4dd333
871933a919cf2e3aac296e057d2b3941b
McGill, M. M. (2008). Critical skills for game developers: An analysis of skills sought by industry (pp. 89–96).
https://doi.org/10.1145/1496984.1497000
McGill, M. M. (2009). Defining the expectation gap: A comparison of industry needs and existing game
development curriculum (pp. 129–136). Proceedings of the 4th International Conference on Foundations of
Digital Games. https://doi.org/10.1145/1536513.1536542
Mikami, K., Watanabe, T., Yamaji, K., Ozawa, K., Ito, A., Kawashima, M., … Kaneko, M. (2010). Construction
trial of a practical education curriculum for game development by industry–university collaboration in
Japan. Computers & Graphics, 34(6), 791–799. https://doi.org/10.1016/j.cag.2010.09.015
Rogers, S. (2014). Level Up! The Guide to Great Video Game Design (2nd edition). Wiley.
Salen, K., & Zimmerman, E. (2003). Rules of Play: Game Design Fundamentals. MIT Press.
Schell, J. (2008). The Art of Game Design: A book of lenses. CRC Press.
Swink, S. (2008). Game Feel: A Game Designer’s Guide to Virtual Sensation (1st ed.). CRC Press.
https://doi.org/10.1201/9781482267334
Valentine, R. (2014, September 11). Video game degrees in the US increase by 50% in five years.
GamesIndustry.Biz. Retrieved from
https://www.gamesindustry.biz/articles/2014-09-11-video-game-degrees-in-the-us-have-increased-by-50-pe
rcent-over-the-last-five-years
Wallach. (2020, November 23). 50 Years of Gaming History, by Revenue Stream (1970−2020). Retrieved from
https://www.visualcapitalist.com/50-years-gaming-history-revenue-stream/
Copyrights
Copyright for this article is retained by the author, with first publication rights granted to the journal.
This is an open-access article distributed under the terms and conditions of the Creative Commons Attribution
license (http://creativecommons.org/licenses/by/4.0/).
192