Content uploaded by Michele Mauri
Author content
All content in this area was uploaded by Michele Mauri on Nov 17, 2017
Content may be subject to copyright.
Author accepted manuscript of:
Mauri, M., Pini, A., & Ciuccarelli, P. (2017). Designing diagrams for
Wikipedia. Information Design Journal, 23(1), 65–79.
The article has been accepted on Feb 08, 2017 for publication in the
Information Design Journal, Volume 23, Issue 1, 2017.
The article is under copyright, and the publisher (John Benjamins Publishing
Company) should be contacted for permission to re-use the material in any
form.
The final version of this article is available at this URL:
http://dx.doi.org/10.1075/idj.23.1.08mau
The final version will be published in Open Access and freely available as
sample issue on the website starting on January 1st, 2018, for an entire year.
1
Designing diagrams for Wikipedia
1 Abstract
Despite the high usage of diagrams and images in Wikipedia, as well as across all
Wikimedia projects, few studies have been conducted on the role of visual contents within the
online encyclopedia and on the collaborative creation of diagrams. With the present research,
we brought a group of designers into Wikipedia, with the goal of exploring how information
designers behave on this platform and how other users react to their involvement. Inspired by
the WikiEdu Program, we engaged postgraduate design students in the creation of diagrams for
Wikipedia and then followed the reactions of both users and designers. The results of the
experiment have been evaluated using built-in Wikipedia functions (e.g., page history and
discussion pages) and through an anonymous survey among the students involved. This
experience brought to light the different consideration granted to images and diagrams with
respect to texts on the online encyclopedia. It also allowed a reflection on the role of designers
in the knowledge production process, as well as on the meaning of producing “open” contents,
which are meant to be improved by other people.
2 Introduction
Today, Wikipedia represents one of the main access points to information: it is the 6th
most visited site in the world (Alexa Internet Inc., 2016), and at the moment of writing it counts
294 language editions (“List of Wikipedias,” 2016). Its value goes beyond the publicly displayed
information through its interface: the whole process of the creation and refinement of
information, the discussions about its policies, and the social relationships behind the project
have been stored and made publicly available, making this project one of the most intriguing
dataset on the collaborative creation of knowledge and meaning on the earth. Moreover, all this
information is free from copyright and without technical limits for its reuse.
Quite a solid research body is taking shape around Wikipedia as a significant data
source, thus suggesting the need to apply analysis and visualization techniques to extract and
make sense of massive amounts of information (Bao et al., 2012; Massa & Scrinzi, 2013;
Wattenberg, Viégas, & Hollenbach, 2007). However, very little research has been conducted on
the role of images and diagrams within Wikipedia articles and the collaborative process of their
creation (Viegas, 2007). Nevertheless images and diagrams are a relevant part of the Wikipedia
ecosystem: Creative Commons, Wikipedia’s side project focused on media, counts more than
one billion media items (“State of the Commons,” 2015), used across all Wikipedia-related
Revised Manuscript
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
2
projects. Images and diagrams are indeed a powerful device for information (Ricci, 2010) and
users rely on them as well as text, if not more.
By recognizing this dearth of scientific production, with the present research we brought
a group of designers into Wikipedia to analyze how information designers act on Wikipedia and
how Wikipedia users react to their involvement.
To reach the needed critical mass for a similar experiment, we involved students
enrolled in their second year of a master’s degree in Communication Design, thus young
professionals. Inspired by the WikiEdu Program, we asked our students to design a diagram for
a Wikipedia article. The WikiEdu Program is simple and effective: as an assignment, students
need to improve a Wikipedia page related to the course’s topic (“WikiEdu Website,” n.d.). By
doing so the students develop writing skills and critical thinking, and at the same time they
produce something valuable for the whole community. We saw a great potential in bringing this
kind of assignment in graphic design education, specifically in information design. This
assignment allowed us to investigate the relationship between Wikipedia and diagrams from two
different angles:
- An analysis of how information designers act on Wikipedia, and how Wikipedia users
react to designers’ involvement. As we will explain later, the assignment’s goal was to
produce a diagram for a specific Wikipedia page, and for the final evaluation students
had to engage with other users to include the diagram in the article.
- The definition of a new format for design education. We transposed the WikiEdu
guidelines from the creation of texts to the creation of diagrams, seeing it as an enriching
experience both for students and for the Wikipedia community. The experience was a
test to understand whether a format that proved to be valuable for texts could also work
in the field of information design.
In this article, we will focus mainly on the first angle.
3 Experiment setup
The formulation of the assignment was designed to follow the basic rule of Wikipedia:
write on something you know. Each student was encouraged to select a known topic to be able
to evaluate if there was a lack of diagrams in the relative article. Since “diagram” can have a
vast range of meanings (Ricci, Ciuccarelli, & Valsecchi, 2008), we suggested focusing on a
topic belonging to these four categories:
- Processes: the making of something, phases, materials, and techniques;
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
3
- Technological descriptions: how something works, its components, the relationships
among them;
- Historical events: temporal evolution of an event, the main actors, their movements
- Quantitative phenomena: create a visual representation of a dataset, highlighting
relevant correlations.
The diagram had to be designed for a specific existing article. In this way, it was simpler
for students to identify which part of the topic should be visualized and what kind of knowledge
could they add to it. They were free to create a new diagram or to improve an existing one
within the article. Since on Wikipedia there are some meta pages containing lists of requested
materials, we pointed them to the list of requested diagrams1 to understand which kind of
diagrams are usually demanded by other users. Finally, no technical constraints were given:
students were free to choose any visual language and technique.
Students had seven weeks to create the diagram, upload it, and advocate for its
inclusion. Preparing this assignment, we took advantage of the knowledge gathered through
other projects related to Wikipedia (Borra et al., 2015; Mauri, 2011). These experiences allowed
us to better understand how the encyclopedia works, particularly its social dynamics and its
policies, formalizing them in the following four guidelines we gave to the students.
3.1 Understand the project
Wikipedia is just the most visible part of a much bigger ecosystem of projects, operated
by Wikimedia Foundation.2 Each project focuses on a specific type of knowledge: Wikiquote for
quotation, Wikisource for historical books, Wikinews for current topic, Wikipedia for the
encyclopedic knowledge, and so on.
Wikimedia Commons is the project focused on media: images, videos, and audio files
are collected there. Images can be then linked in any Wikipedia page, in any language. Before
uploading images, our students had to become familiar with the projects and understand their
dynamics.
3.2 Follow the standards
There are no clear standards on Wikipedia about images. The only requirement is to
adopt a free license (like as a Creative Commons license). We asked students to identify
whether best practices exist and are already shared among Wikipedia users. It turned out that
1 It is possible to see the full list of requested diagrams visiting this page:
https://en.wikipedia.org/wiki/Category:Wikipedia_requested_diagram_images
2 At the moment of writing, the Wikimedia Foundation operates 12 projects; here is possible to see the full
list: https://en.wikipedia.org/wiki/Wikipedia:Wikimedia_sister_projects
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
4
while there are several guidelines for articles (“Wikipedia:Policies and guidelines,” 2016), little
information is available for designing graphics. Only for geographic maps there are community-
defined conventions. Even if this could be seen as a drawback, as our students had no clear
instruction on how to create diagrams, it pushed them to experiment with information
visualization.
3.3 Follow the ethics
The third guideline is the most important. Even if there are neither strict standards nor
conventions, we asked the students to find technical solutions to preserve Wikipedia’s ethics.
Wikipedia presents itself as the “encyclopedia anyone can edit.” This also means that anyone
can improve its contents and that contributors must detach from the idea of being the
recognized authors. As will be described later, this is a very sensitive topic for graphic
designers. In technical terms, this means, for example, that vector formats should be favored
over raster images, that text paths should be preferred over traced texts, and that the file
layering should be understandable by other people.
3.4 Engage with the community
Finally, we asked the student to actively engage with the community: after the diagram
has been created, they must upload it and try to push the insertion of the graphic in the related
Wikipedia article, engaging with the other users in the talk pages.
4 Evaluation
To evaluate the reaction of other users, we repurposed the technical and social
mechanisms underlying the collaborative encyclopedia: analysis of discussions, analysis of
page history, analysis of diagram reuse, analysis of diagram modifications, and remixes. Below,
we briefly introduce the Wikipedia technical features we used for the evaluation.
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
5
Figure 01. A screenshot of the page history for the “2015 Ankara Bombings” article. It is possible to see the list of
edits: for each one it is possible to see the timestamp, the user, and the description.
Figure 02. A screenshot of the discussion page for the “2015 Ankara Bombings” article. Each title defines a new
discussion, each comment has a different indentation according to the reply chain.
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
6
Figure 03. A screenshot of the image page on Wikimedia Commons.
4.1 Page history
All the revisions of each Wikipedia article are stored online, and they are publicly
available. In Figure 01, it is possible to see their appearance through the interface. This is
originally intended as a versioning tool for editors: in this way, it is simpler to identify the latest
edits, who made them and when, compare the latest version with the previous one, and
“rollback” it if needed, meaning restore an old version of the page. Each time a user saves a
new version of the page, the system asks for a short description of the edits. In addition, when
new images are added or removed, the edit is traced. The revision history of an article is useful
to determine whether images were deleted and the reason, how long they lasted, and the pace
of editing for a specific page.
4.2 Discussion pages
Each Wikipedia entry is composed by the article page and the discussion page. The
“talk” tab opens the discussion page—as in Figure 02—where users coordinate on the article
evolution, and, if needed, solve controversies avoiding the raise of edit wars3. The discussion
3 According to Wikipedia policies, an edit war occurs when editors who disagree about the content of a
page repeatedly override each other's contributions.
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
7
pages are organized with a tree structure to show which comments respond to whom.
Commonly, new topics are defined by a new title. When commenting, users leave their
signature, meaning a link to their user page and the timestamp of the comment. The edit history
of discussion pages, as for the articles, is stored and browsable. Discussion pages were useful
to understand why some diagrams were rejected or moved to other pages. We encouraged the
students to actively engage in discussions, explain the reason for their designs, and accept
suggestions and critiques.
4.3 Image pages
Images are stored in the Wikimedia Commons website, the “sister project” focused on
media (pictures, videos, sounds). When a new image is uploaded, a wiki-page is created, as in
Figure 03. The new page shares all the basic characteristics of articles: history and talk pages.
There are some additional functions to image pages, such as the revision history of the picture
and the link to all Wikipedia articles using it. These pages are useful to trace the image use, in
which pages it has been inserted, its usage in other linguistic versions, if derivate works have
been created based on it, and if other users have modified and improved it.
4.4 Poll among students
Finally, we set up an anonymous form to collect feedback from our students, in
representation of the feedback from the information design area. In the poll, we asked for
feedback on their experience regarding whether they found the contribution to Wikipedia
interesting, the main difficulties, and their relationship with the project beyond this exercise.
5 Results
The students produced and uploaded 73 different diagrams on Wikimedia Commons 4.
Students covered a huge variety of topics, creating both still images and animations. The
created diagrams are mainly about processes (such as Jack Daniel’s production, olive oil
extraction, etc.), technical descriptions (how an astrolabe works, why we see mirages), and
historical events (the Battle of Thermopylae, or the Cambodian Campaign in 1970). Few
students worked on “classical” data visualizations (e.g., the evolution of Haiti economy, the
results of Olympics Games). Students created both static images (56) and animations (17).
Static images are mainly in vector format: only a third of the diagrams are PNG or JPG.
4 The results of the exercise can be found at this link:
https://commons.wikimedia.org/wiki/Category:Contents_produced_by_DensityDesign_students
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
8
As in Figure 04, the community accepted 45 diagrams, while 13 were rejected. In 15 cases,
students were not able to add the image to a Wikipedia article.
We found a correlation between the grades we assigned to the diagrams and their performance
on Wikipedia: the lowest graded diagrams were also not accepted in the article.
Figure 04. Accepted, rejected, and discussed diagrams per topic.
5.1 Debate
Very few diagrams generated an actual debate in Wikipedia talk pages. All 13 rejected
diagrams were at least commented on, while only 8 of the accepted ones created some debate.
We explore in detail the discussion topics in the next session (See the section “Results
analysis”).
5.2 Re-use of uploaded diagrams
Fifteen images were reused in other pages. In most of the cases, diagrams were added
to another linguistic version of the same article. Apparently, only one diagram was fully
translated in another language. We are saying “apparently” because there are no automatic
ways to follow potential derivative works on Wikimedia Commons. When users download and
modify an image and then re-upload it, they should put a link to the original work, but this is not
automatic.
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
9
6 Results analysis
6.1 Type of discussion
Most of the diagrams were accepted without any kind of comment. At the same time, we
can identify the main concerns of other users from the few discussions that arose and the main
reason for an image’s rejection. Four main types of discussions emerged.
6.1.1 Misleading/confusing. Some diagrams were rejected based on the argument that they
were not clear or, even worst, misleading. Three images received this kind of comment, and as
teachers we agreed: The graphics were indeed not visually efficient or the information source
was not clear.
6.1.2 Does not add information. If the diagram is simply a visual translation of the text, it will
probably be discarded. On Wikipedia, apparently, the text has priority over images. This makes
sense thinking about how simple is to translate, transform, and repurpose text rather than
images, especially in the process of adaptation for different supports (mobile, desktop, etc.) or
for visually impaired people. Many images have not been accepted because they were mere
visual translations of the information contained in the text. As an example, Figure 05 represents
how the production process of chocolate works. It was refused because, according to users, the
space for this image would be better used for a textual description:
I don't think this diagram adds enough to make it worth the page space it will
take up. (Wikipedia user, commenting on the “Chocolate Production” diagram)
A second example is Figure 06, refused because it is too illustrative and seen as a
decoration rather than real information:
I'm not seeing an advantage to the reader in replacing the current interactive
graph with the new one. It looks prettier, but is less helpful. I think if there were
a way of introducing graphics and color into the present graph, that would be
great! (Wikipedia user, commenting on the “Brewing Process” diagram)
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
10
Figure 05. (on the left) Diagram representing the chocolate production process.
Figure 06. (on the right) Diagram representing the beer brewing process.
6.1.3 Not suited for Wikipedia UI. In the actual interface of Wikipedia, images are usually
rendered as 220 pixel wide thumbnails. This means that diagrams meant to be read at full size
risk becoming completely illegible. For example, for the diagram in Figure 07, a user
commented:
The main reason for not adding this to the article is that it does not work well
as a thumbnail. As the image on the right shows, it doesn't work well as a
thumbnail on a web page. Even the clickable version is not easy to read, as
the text is too small. It would work best printed on an A4 sheet of paper.
(Wikipedia user, commenting on the “Google Acquisitions” diagram)
This kind of critique raises a general question: Should be diagrams designed for the
current Wikipedia UI, or can they go beyond it?
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
11
Figure 07. (on the left) Timeline of the acquisition made by Google, colored by typology.
Figure 08. (on the right) A possible interpretation of Odysseus journey.
6.1.4 Original research
In two cases, the discussion focused on the “no original research” (also called WP:NOR)
Wikipedia’s policy:
Wikipedia does not publish original thought: all material in Wikipedia must be
attributable to a reliable, published source. Articles may not contain any new
analysis or synthesis of published material that serves to advance a position
not clearly advanced by the sources. (“Wikipedia: No original research,” 2016)
Since many people edit Wikipedia, and it is not possible to verify their competences,
every user is required to cite a proper source for the claims added to the page. A person cannot
insert his own analysis of a topic: he should first publish it in a journal, conference or known
newspaper and then add it to Wikipedia. This rule applies also to images, but users are allowed
to create original work:
Original images created by a Wikipedian are not considered original research,
so long as they do not illustrate or introduce unpublished ideas or arguments,
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
12
the core reason behind the NOR (No Original Research) policy. (“Wikipedia:
Policies and guidelines,” 2016)
The diagram in Figure 08 indeed was rejected for this reason. The diagram represents
Odysseus’ journey on a map, showing with the thickness of the line the number of man with
him. The user rejecting it stated:
[…] I've removed it. No one knows whether the places mentioned in the
Odyssey were meant to correspond to real geographical locations, so any
map like this would be highly speculative at best. In any case, and more
importantly, such a map, not coming from a reliable source would constitute
original research which is not allowed on Wikipedia. (Wikipedia user,
commenting on the “Possible causes of Colony Collapse Disorder” diagram)
A more complex discussion arose on the diagram “Possible causes of Colony Collapse
Disorder” (Figure 09). In this case, the diagram was a visual translation of the possible causes
listed on the page. This became an interesting discussion about how these policies should apply
to images:
That the image is based on the article is actually a bit of the problem. What
you’ve basically done is a novel synthesis not only of the subject matter, but
we also should not be using Wikipedia as a source (WP:CIRC). If we were
going to have such an image, it would need to have been published
somewhere, as [previous user] alluded to, so we could verify it came from a
source with expertise to review this topic. The effort is definitely appreciated
though, as I’m sure some time went into making the image. (Wikipedia user,
commenting on the “Possible causes of Colony Collapse Disorder” diagram)
A second user disagreed with him:
I think it should go into the article, until we have a better replacement. My
understanding is that the WP:IMPERFECT policy allows imperfect things in
the articles. (Wikipedia user, commenting on the “Possible causes of Colony
Collapse Disorder” diagram)
In the end, the diagram was not added to the article not due to the “no original research”
policy, but rather because images are more difficult to edit than text:
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
13
That policy you mention applies much more to articles that we can easily
improve over time. That spirit relates more to starting with something sub-
standard, but acceptable that can be improved over time. When a flaw is
pointed out from the beginning though, we don’t need to automatically accept
that. The other problem is that images cannot be easily edited like text in an
article. You need access to GIS software for this. (Wikipedia user,
commenting on the “Possible causes of Colony Collapse Disorder” diagram)
Figure 09. Visual synthesis of “Possible causes of Colony Collapse Disorder.”
6.2 Students’ feedback
At the end of the course, we asked students to participate in an anonymous survey to
understand whether they found the exercise interesting, the main difficulties, and the
unexpected results. Out of 74 students, 32 agreed to participate to the survey. Of them, 84%
said that the creation of a diagram specifically for Wikipedia and not for other uses was
enriching for them. When we asked them why, the main cited reasons are that they felt they
were actually contributed to the enrichment of someone else and that they felt responsible for
the quality of their work:
You have to design something for everybody. It’s an extra responsibility for
you. (Anonymous student #18)
Other students said that it was interesting to understand what is beyond Wikipedia (its
dynamics and rules) and had the feeling of “giving back” something to a project they use daily,
as stated by this student:
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
14
At first I didn’t understand the utility of this exercise. When it “forced” me to
enter into the Wikipedia community and to understand the underlying
dynamics, then I found it extremely interesting. (Anonymous student #8)
Finally, few of them said that having a diagram accepted on Wikipedia is valuable for
their portfolio visibility.
When discussing the difficulties, 31% found difficulties in contributing to Wikipedia. Most
said that the main difficulty is the use of the SVG format, as MediaWiki (the software running
Wikipedia) can render only few graphic filters and effects. In particular, a strong limitation is the
small number of available fonts: even if one can use any font in an SVG image, MediaWiki is
able to render few of them. Sometimes, even for available fonts, the render is not perfect.
Other difficulties were related the choice of a topic—some of the students realized the
complexity only after starting to visualize it:
The chosen topic, in hindsight, was too complex and historically not clear;
therefore, the project realization was complex, and in the end, it was not
accepted. (Anonymous student #5)
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
15
Figure 10. Most of the students found the exercise “useful” or “very useful.”
When we asked if they found the exercise useful for their design education, most found it
very useful (see Figure 10). Finally, when asked whether there was something unexpected
(positively or negatively) in this experience, most of the students said that working on Wikipedia
has been enriching beyond what is strictly related to design education. The question “Did you
learn something beyond information visualization creating a diagram for Wikipedia?” produced
several insights. Many students said that it is simpler than thought:
We never think about it, but it should be something to do more often, it is
something for everybody. (Anonymous student #10)
Other students were impressed in realizing how many people read Wikipedia, and
therefore see their diagrams:
I learnt something new not related to design. It is nice to think that many
people around the world will see my diagram. (Anonymous student #12)
Only one student cited the community feedback as something enriching:
The positive comments left by other users on the utility of my diagram and its
reuse on other language editions made me understand the relevance of
graphic design and its power in transforming culture (Anonymous student #32)
Being forced to expose their work to unknown people, collecting in most cases a positive
reaction, made clear to our students their role as communication designers in the dissemination
of knowledge.
Even if most of the students followed the fate of their diagram after the end of the course
(75%), and most found the experience interesting, when asked if they continued to contribute to
Wikipedia, only one answered affirmatively.
7 Discussion
As we stated in the introduction, our aim was twofold: the definition of a new format for
the education of communication designers, and the analysis of how information designers can
act on Wikipedia. As a didactical activity, the experience proved to be valuable under several
points of view. Students felt that their work was useful for other people, and not just to improve
their visual skills in information design. The feedback gathered from external people, not
necessarily experts in graphic design, were useful to test their ability to convey meaningful
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
16
information. The experience helped them to reflect on the role of designers in communicating
information and knowledge as well as on the tradeoff among visual clarity and information
richness. Finally, it was an opportunity for them to understand what a collaborative open project
is by doing it.
Some issues remain unresolved. Since for the students it was an exercitation, their
approach was more focused on convincing other users to accept the visualization rather than to
engage in a discussion, at the cost of refining and modifying the diagram. A second finding is
the difficulty to keep our students engaged in the discussion with other Wikipedia users. Our
students are more familiar with a “finish first, then publish” process. Diagrams, and artifacts in
general, are released only when complete and fine-tuned. Adopting an open source mentality,
based on the disclosure of the production process (Raymond, 1999), is key if we want to
contribute to collaborative processes as graphic designers. The approaches to establish this
mentality in our school are yet to be defined. Planning a future edition of this didactical exercise,
we would try to involve other users as tutors for our students, guiding them in the policies of the
encyclopedia. We would also set up a process based on quick releases, focusing on the
discussion with other users for the diagram improvement.
As an experimentation meant to bring information designers on Wikipedia, the
experience revealed several limits and, at the same time, opened interesting reflections. The
response from the other Wikipedia users, the high rate of acceptance and the low level of
discussion, is difficult to evaluate and led to four main interpretations or hypothesis. First, it
could be a tacit approval of the produced diagrams. Second, it could be related to the selection
of marginal articles by our students, leading to a low interest for other users. A third hypothesis
is that the addition or modification of images is seen as a marginal activity, less relevant than
textual edits. Finally, since images are seen as difficult to edit, the effort is awarded to
encourage further contributions. More research is required to evaluate these hypotheses and to
analyze the level of acceptance of visual edits compared to textual ones.
Another criticality is that among all the information designers (our students) we involved,
almost no one continued to edit Wikipedia. Nevertheless, all of them declared the experience
interesting and enriching. In this case, it is also difficult to evaluate the outcome. It could be a
matter of time (the creation of a diagram, its test on the Wikipedia platform, and the discussion
are time-consuming), or a lack of reward for the effort. In our experience, young graphic
designers are really motivated to work on a personal project when this brings visibility, adding it
to their portfolio. In a system that discourages the creation of diagrams and diminishes the
relevance of its author, the commitment could be easily lost. Lastly, a motivation could be a lack
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
17
of a community of users focused on the creation of diagrams and visualization: information
designers could feel lost, with no clear guidance and no one with an overall view on the project.
From a technical point of view, the creation and editing of an image proved to be complex and
intricate. As we stated in the introduction, the openness of a diagram is not just related to the
license one adopts, but also to the technology. If we want to enable other people to edit and
modify our work, we should provide files designed to be modified: for example, favoring the use
of vector images over raster ones.
With this experiment, we have seen a lack of technical tools to maintain the simplicity of
the openness of diagrams and their management. The main limitation is the difficulty in sharing
an editable version of the diagram: while it is possible to upload editable files, their rendering
and use across Wikipedia is highly unstable. Many students, in the end, opted for the raster
images to preserve the visual appearance of the diagram. The overall architecture, in the
practice, seems to discourage the creation of diagrams that are simple to edit and update. Even
if the project ethic is to produce artifacts meant to be free, reusable, and remixable, there are no
formalized tools to enable this approach. In this experiment, we repurposed tools provided by
MediaWiki (page history, talk page, Commons pages) to see the evolution of the diagrams
produced by our students, but there is a lack of dedicated tools for images. If the Wikipedia
community wants to extend to images the same ethics applied to texts, there is a need for new
tools to keep track of reuses and transformations of images. As an example, taking inspiration
from the Git5 environment, it could be useful to have tools to “fork” existing media and reuse
them in a new artifact, keeping track of the links.
In general, we felt that, from a technological point of view, the platform is still highly
focused on texts to the detriment of other media. Moreover, the problem for videos and non-
static artifacts remains open: if we represent something through an animation, how can we give
other people the ability to modify it? How can we provide access to the source files, particularly
when they have been created with proprietary software?
Even from the point of view of Wikipedia’s community policies, the discussions prove
how users struggled in trying to understand how to apply existing policies to images. While for
some policies the interpretation is quite simple (e.g., “Be bold,” meaning The Wikipedia
community encourages users to be bold when updating the encyclopedia), others are not. For
example, how does the “no original research” apply to images? Until where can the visual
transformation of knowledge be considered a linear mapping of already existing information,
5 Version control is a system that records changes to a file or set of files over time so that it is possible
recall specific versions later. (Chacon & Straub, 2014)
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
18
and when does it become original research? Also related to sources, there is a need for
reflection on how they could be cited in a diagram.
To conclude, while the experience proved to be interesting and valuable for design
education, many efforts are required both to involve information designers and to enable the
creation of truly open diagrams on Wikipedia. While a strong community is guiding the textual
enrichment of the encyclopedia, there is not yet any form of organized support for its visual
dimension.
In our opinion, there are two ways to address such criticalities. The first one is
technological and is related to Wikipedia/Wikimedia: it should provide tools for sharing source
files, not just the end results. The second one is conceptual and concerns information
designers: they should move from the idea of being authors to the idea of being contributors of a
diagram.
8 Bibliography
Alexa Internet Inc. (2016). The top 500 sites on the web. Retrieved October 5, 2016, from
http://www.alexa.com/topsites/global
Bao, P., Hecht, B., Carton, S., Quaderi, M., Horn, M., & Gergle, D. (2012). Omnipedia: bridging
the Wikipedia language gap. In Proceedings of the 2012 ACM annual conference on
Human Factors in Computing Systems - CHI ’12 (p. 1075). New York, New York, USA:
ACM Press. http://doi.org/10.1145/2207676.2208553
Borra, E., Laniado, D., Weltevrede, E., Mauri, M., Magni, G., Venturini, T., … Kaltenbrunner, A.
(2015). A platform for visually exploring the development of Wikipedia articles. Proceedings
of the 9th International AAAI Conference of Web and Social Media.
Chacon, S. & Straub, B. (2014). Pro Git (2nd ed.). Berkeley, CA: Apress.
http://doi.org/10.1007/978-1-4842-0076-6
List of Wikipedias. (2016). Retrieved October 5, 2016, from
https://en.wikipedia.org/wiki/List_of_Wikipedias
Massa, P. & Scrinzi, F. (2013). Manypedia: Comparing language points of view Wikipedia
communities. First Monday, 18(1). http://doi.org/10.1145/2462932.2462960
Mauri, M. (2011). Share your knowledge: Mapping project’s influences on the web. Retrieved
September 30, 2016, from http://www.densitydesign.org/research/share-your-knowledge-
mapping-project’s-influences-on-the-web/
Raymond, E. (1999). The cathedral and the bazaar. Knowledge, Technology {&} Policy, 12(3),
23–49. article. http://doi.org/10.1007/s12130-999-1026-0
Ricci, D. (2010). Seeing what they are saying: Diagrams for socio-technical controversies. In D.
Durling, R. Bousbaci, L.-L. Chen, P. Gauthier, T. Poldma, S. Rowoth-Stokes, & E.
Stolterman (Eds.), DRS2010 - Design & Complexity. Montréal.
Ricci, D., Ciuccarelli, P., & Valsecchi, F. (2008). Handling changes through diagrams: Scale and
grain in the visual representation of complex systems. In C. Cipolla & P. P. Peruccio (Eds.).
Turin: Allemandi.
State of the Commons. (2015). Retrieved October 5, 2016, from
https://stateof.creativecommons.org/2015/
Viegas, F. (2007). The visual side of Wikipedia. In 2007 40th Annual Hawaii International
Conference on System Sciences (HICSS’07) (pp. 85–85). IEEE.
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
19
http://doi.org/10.1109/HICSS.2007.559
Wattenberg, M., Viégas, F. B., & Hollenbach, K. (2007). Visualizing activity on Wikipedia with
chromograms. Methods, 4663/2010, 272–287. http://doi.org/10.1007/978-3-540-74800-
7_23
WikiEdu Website. (n.d.). Retrieved October 5, 2016, from https://wikiedu.org/
Wikipedia: No original research. (2016). Retrieved October 5, 2016, from
https://en.wikipedia.org/wiki/Wikipedia:No_original_research
Wikipedia: Policies and guidelines. (2016). Retrieved October 5, 2016, from
https://en.wikipedia.org/wiki/Wikipedia:Policies_and_guidelines
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
