=Paper=
{{Paper
|id=Vol-2827/KBS-Paper_1
|storemode=property
|title=Knowledge in Computational Design: Typography, Emoji and Flags
|pdfUrl=https://ceur-ws.org/Vol-2827/KBS-Paper_1.pdf
|volume=Vol-2827
|authors=João M. Cunha,Pedro Martins,Penousal Machado
}}
==Knowledge in Computational Design: Typography, Emoji and Flags==
https://ceur-ws.org/Vol-2827/KBS-Paper_1.pdf
Knowledge in Computational Design: Typography,
Emoji and Flags
João M. Cunhaa , Pedro Martinsa and Penousal Machadoa
a
CISUC, Department of Informatics Engineering
University of Coimbra, Portugal
Abstract
This paper describes a showcase of Computational Design projects from three different areas: Type
design, Icon design and Flag Design. The nature of these tools is also different, going from a creativity
support tool to a fully automatic system. We analyse these tools in terms of use of knowledge and user
interaction.
Keywords
Computational Design, Typography, Emoji, Flags
1. Showcase Description
The ability to create new concepts by making connections between existing ones is, in part,
related to our creative behavior [1]. Such ability is highly needed in design processes and has
to be considered when implementing Computational Design systems.
The use of knowledge-based systems has great potential in computational design, especially
for solving creative problems based on previous solutions. In this paper, we describe a showcase
of three projects from three different domains: type design, icon design and flag design.
In this section, we describe the projects featured in the showcase, which consist of a creativity
support tool (TypeAdviser), a co-creative system (Emojinating) and a fully automatic one (Ever-
changing flags). All the projects use knowledge to help or guide the process of design in a
computational system.
1.1. Type Design
A typeface is a set of characters that work as a family – even though they all look different,
they share characteristics. The first project that we want to describe, Typeadviser [2], is the
result from a master’s thesis [3] and it aims at helping type designers in their creative process.
It consists of a creativity-support tool that explores anatomic relations among characters of a
typeface to provide help in the early stages of designing a typeface by using semi-automatic
letter-part sharing and allowing the users to compare their design with existing typefaces. In
Joint Proceedings of the ICCC 2020 Workshops (ICCC-WS 2020), September 7-11 2020, Coimbra (PT) / Online
" jmacunha@dei.uc.pt (J. M. Cunha); pjmm@dei.uc.pt (P. Martins); machado@dei.uc.pt (P. Machado)
~ https://cdv.dei.uc.pt/ (J. M. Cunha)
0000-0001-6502-3500 (J. M. Cunha); 0000-0002-3630-7034 (P. Martins); 0000-0002-6308-6484 (P. Machado)
© 2020 Copyright for this paper by its authors. Use permitted under Creative Commons License Attribution 4.0 International (CC BY 4.0).
CEUR
Workshop
Proceedings
http://ceur-ws.org
ISSN 1613-0073
CEUR Workshop Proceedings (CEUR-WS.org)
Figure 1: Interface of TypeAdviser
[3], we conducted a study to identify rules for the development of typefaces that could be used
in a computational system to design a typeface based only on a few characters given by the
user. Despite not being able to fully produce a typeface, the system gives support to the design
task stimulates creativity and helps to reduce the time spent by the designers in the initial stage
of the design process. The system is composed of two components: the Part-sharing (see Fig. 1),
which is based on the consistency criterion, and the Adviser, which uses Self-Organizing Maps
to suggest similar designs to the user.
The user inputs previously drawn vectorial glyphs and then identify their parts. The system
then generates other letters, based on the predefined rules of part-sharing – this is controlled
by the Part-sharing component, which was built using type design principles. The Adviser
uses Self-Organizing Maps (soms) to suggest similar designs to the user. A som is used to
produce a similarity map of different styles of character for each letter, grouping similar styles.
With this approach, the user is able to draw a letter and then see possibilities for the other
letters, based on the drawn one. There are two main uses for this: (i) to serve as inspiration by
allowing the designer to see characters of similar typefaces; but also (ii) to help in distancing the
typeface being designed from existing ones. Our system could be combined with approaches
for computational generation of glyphs [4]. For a more detailed description, we refer the reader
to [2].
1.2. Icon Design
The visual representation of concepts or ideas through the use of graphic shapes has always
been explored in the history of humanity and has a key role in how we share information
among each other. This is the base idea behind our work on computational generation of
visual representations of concepts. We follow an approach that explores visual blending of
existing symbols to represent new concepts. Our contributions in this domain started with
a low scale system, in which three concepts (pig, angel and cactus) and corresponding visual
representations were used to produce combinations [5]. In a follow up project, we increased
faith log income kind obligation invitation anything aircraft
Figure 2: Examples of blends produced by Emojinating
Figure 3: Flags generated by Ever-changing Flags on November 15𝑡ℎ 2019
the scale by implementing a system that allows the user to introduce text to be represented [6]
– Emojinating. The system retrieves existing emoji that match the input text using semantic
information associated with them and also by exploiting ConceptNet [7] to extend introduced
concepts. The retrieved emoji are visually blended to produce representations for the input
text (see Fig. 2). The user is able to guide the production of solutions and the system is able to
change its own knowledge to constantly adapt to the user [8, 9]. Our approach is aligned with
other projects that address visual representation of concepts using emoji, e.g. [10]. For a more
detailed description about our system, we refer the reader to [6, 8, 9].
1.3. Flag Design
The last project that we want to present addresses flag design. Flags have an important role
when it comes to constructing and maintaining a sense of identity of a nation [11]. The symbols
that are included in a flag reflect changes that occurred in the entity that they represent and
their design is affected by a process of slow evolution, which you can see by looking at previous
versions. We go one step further and use the constant-changing nature of our society and the
easy access to global information to propose that, in addition to identity, a flag can also represent
“mood”. By “mood” we refer to a temporary, yet meaningful, state. Ever-changing Flags [12]
uses knowledge from three different sources: (i) meanings of existing flags, (ii) a dataset of
colour names, and (iii) semantic information associated with emoji. These knowledge sources
are used to produce changes to country flags based topic (see Fig. 3), while still maintaining
the resemblance to the initial flag. The system is based on the idea of “ever-changing flag”,
i.e. a flag that is in constant change and reflects the “mood” of the country that it represents.
This constant updating is implemented by using Google News rss feed to retrieve trending
topics, which are then used to motivate the changes in the flag. This sense of reactivity to the
environment is also explored by other authors, e.g. [13]. For a more detailed description, we
refer the reader to [12].
2. General Overview
We briefly described three projects (TypeAdviser, Emojinating and Ever-changing Flags) from
three different domains (Type Design, Icon Design and Flag Design). All these projects have one
thing in common – they all use domain knowledge to guide and/or aid in a process of design.
However, the way that they integrate and make use of this knowledge is not the same.
One issue with implementing systems for computational design is that they address tasks that
can be seen as open-ended – there might be no optimal solution since they often depend on user
preferences. For reason, the relation between user and system is of extreme importance when
developing such a system. This relation has an obvious impact on the way a system is designed,
especially concerning the use of prior knowledge. By analysing the described projects, it is easy
to identify several ways of using knowledge. TypeAdviser uses knowledge to reduce the time
spent in designing a typeface but also to present the user characters of similar style, thus making
it easier to either come closer to or avoid resemblances to existing typefaces. On the other hand,
while TypeAviser uses knowledge to merely aid the process of design, Emojinating highly relies
on domain knowledge. The capability of reaching solution that represents a user-introduced
concept depends on the existence of such knowledge. Moreover, TypeAviser can only be seen
as a creativity support tool. In contrast, Emojinating has a more cooperative relation with the
user, creating a co-creative dynamic between the two. In addition to this, Emojinating is able to
adapt itself to the user by changing its knowledge – e.g. by a keeping a record of what sort of
solutions match the user preferences.
The third project (Ever-changing Flags) can be placed in the opposite end of the spectrum to
TypeAdviser – it requires very little of the user. The system uses the domain knowledge as a
guide to produce its own solutions – e.g. using elements (or part of them) that have a certain
meaning to assign a similar meaning to a flag. In doing so, the only thing required of the user is
the selection of which country the system should produce a flag for.
Another interesting aspect is the possible applications of these systems. One example is the
use of Emojinating engine for purposes of information visualization [14]. Another example
concerns one of the goals for the project Ever-changing Flags. In addition to being used in a
web-app, there is also the possibility of integrating in an installation. Such setup increases
the impact on the participant and addresses questions of perception – do the participants still
identify themselves with altered versions of the flag of their country? All these aspects gain
importance in our research work, which not only involves implementing systems but also
thinking about their impact on society [15].
Acknowledgments
This work is partially funded by the Foundation for Science and Technology, I.P., within the
scope of the project CISUC - UID/CEC/00326/2020 and by European Social Fund, through the
Regional Operational Program Centro 2020, and under the grant SFRH/BD/120905/2016.
References
[1] M. Boden, The Creative Mind: Myths and Mechanisms, Routledge, 2004.
[2] J. M. Cunha, T. Martins, P. Martins, J. Bicker, P. Machado, Typeadviser: a type design aiding-
tool, in: Proceedings of the Workshop "Computational Creativity, Concept Invention, and
General Intelligence 2016", 2016.
[3] J. M. Cunha, Dissertation on anatomical relationships among characters of a typeface
(Dissertação sobre relações anatómicas entre caracteres de um tipo de letra), Master’s
thesis, University of Coimbra, 2013.
[4] D. Lopes, J. Correia, P. Machado, Adea-evolving glyphs for aiding creativity in typeface
design, in: Proceedings of the 2020 Genetic and Evolutionary Computation Conference
Companion, 2020, pp. 97–98.
[5] J. M. Cunha, J. Gonçalves, P. Martins, P. Machado, A. Cardoso, A pig, an angel and a cactus
walk into a blender: A descriptive approach to visual blending, in: Proc. of the Eighth Int.
Conf. on Computational Creativity, 2017.
[6] J. M. Cunha, P. Martins, P. Machado, How shell and horn make a unicorn: Experimenting
with visual blending in emoji, in: Proceedings of the Ninth International Conference on
Computational Creativity, Salamanca, Spain, June 25-29, 2018., 2018, pp. 145–152.
[7] R. Speer, C. Havasi, Representing general relational knowledge in conceptnet 5., in:
Proceedings of the Eighth International Conference on Language Resources and Evaluation,
LREC 2012, Istanbul, Turkey, May 23-25, 2012, 2012, pp. 3679–3686.
[8] J. M. Cunha, N. Lourenço, J. Correia, P. Martins, P. Machado, Emojinating: Evolving
emoji blends, in: Proceedings of the Eighth International Conference on Computational
Intelligence in Music, Sound, Art and Design, EvoMUSART 2019„ Springer International
Publishing, 2019, pp. 110–126.
[9] J. M. Cunha, P. Martins, N. Lourenço, P. Machado, Emojinating Co-Creativity: Integrating
Self-Evaluation and Context-Adaptation, in: Proceedings of the Eleventh International
Conference on Computational Creativity, 2020.
[10] P. Wicke, M. Bolognesi, Emoji-based semantic representations for abstract and concrete
concepts, Cognitive processing (2020).
[11] G. Elgenius, Symbols of nations and nationalism: Celebrating nationhood, Palgrave Macmil-
lan, 2011.
[12] J. M. Cunha, P. Martins, H. G. Oliveira, P. Machado, Ever-changing flags: Trend-driven
symbols of identity, in: 8th Conference on Computation, Communication, Aesthetics & X
(xCoAx 2020), 2020.
[13] S. Rebelo, C. Pires, P. Martins, J. Bicker, P. Machado, Designing posters towards a seamless
integration in urban surroundings: A computational approach, in: Proceedings of the 9th
International Conference on Digital and Interactive Arts, 2019, pp. 1–12.
[14] J. M. Cunha, E. Polisciuc, P. Martins, P. Machado, The many-faced plot:strategy for auto-
matic glyph generation, in: Proceedings of the 22st International Conference Information
Visualisation (IV), 2018, IEEE Computer Society, 2018.
[15] J. M. Cunha, P. Martins, P. Machado, Ever-changing Flags: Impact and Ethics of Mod-
ifying National Symbols, in: Proceedings of the Eleventh International Conference on
Computational Creativity, 2020.