Imprint: The content of this blog entry is based on VR experiments in the context of projects carried out in the scope of the authors masters degree studies at FHNW University of Applied Sciences and Arts Northwestern Switzerland under the supervision of Prof. Dr. Doris Agotai.

The result of our research so far is a functional prototype, implemented with Unity using the VRTK framework, optimized for the HTC Vive headset and controllers. Following the Visual Information Seeking Mantra it contains a map offering an overview of the full dataset, a body referenced on demand menu for filtering data points, and visualization concepts for direct integration of filter results in the map representation in the form of a Choropleth map encoding in combination with volume height adjustments. This immediate spatial representation of search results offers rapid insight on the contents of the dataset to the user. Thoughtful attention was given in respect to level of detail and visualization fidelity in the design and implementation of exemplary samples of collection objects. Applying the “details on demand” principle, a detail view with additional information is displayed by pointing at an item.

The application provides system control by the means of a perspective controller, allowing zooming in and out to shift the user’s perspective back and forth between bird’s eye view and ground level view. Spatial movement on the map is offered by physical locomotion and teleportation. And data management is provided through a JSON interface.

User tests have resulted in generally positive feedbacks, revealing that this form of spatial visualization and interaction with data provides benefits for the data exploration task and indeed offers a fresh perspective on data.

some impressions

outlook

Future work includes refinement of menu design and introduction of more elaborated search and filtering options and giving a deeper look into orientation. For the detail view the display of additional attributes or behaviors of objects such as its provenance journey, plus interactions like marking favorites or comparing the behavior of multiple selected objects are envisioned.

Furthermore, freehand system control should be considered. The benefit of gesture input lies in the natural characteristics of direct mapping of user movement which are easier to learn than virtual control performed by an input device. Studies have shown that bimanual input in particular has the potential to allow users to perform tasks faster, leverage existing skills, and increase expressiveness. A drawback is the need for clear delimiters to indicate initialization and determination of gesture, otherwise normal human motions may be interpreted as gestures while not intended as such.

Recognizing that spatial visual representation, filtering and interaction capabilities provided by this virtual data exploration space result in more rapid understanding and unique insight on data in general, applying these findings to other datasets with geographical reference should be considered. Museums in particular have a significant amount of reusable data about their collections that often remains largely untapped, yet has the potential to create real value if presented in a meaningful way and made accessible for interaction, exploration and discovery, eventually allowing the documentation layer to become an actual experience. For this reason, every endeavor should be made towards reusability. This implies in particular advancements in the data management interface and modularity of the application architecture as well as on the elaboration of concepts to embed such a VR experience into the context of an online presence or as part of an exhibition.

Imprint: The content of this blog entry is based on VR experiments in the context of projects carried out in the scope of the authors masters degree studies at FHNW University of Applied Sciences and Arts Northwestern Switzerland under the supervision of Prof. Dr. Doris Agotai.

Because VR is still a rather new technology and there are not many established interaction patterns yet, it is important to continuously get user feedback during the development process and apply the learnings immediately to the next iteration of the prototype. This frequent lean feedback loop happens in informal test settings as opposed to periodical more formal user and usability testing procedure, which may include a variety of aspects such as comprehensibility, effectiveness, efficiency, satisfaction, learnability, retainability, ease of use, speed of performance, rate of errors and user comfort.

Testing procedure

In our testing scenario the context, overarching idea and features of the application were briefly explained to the probands prior to launching the test user into the VR experience where they had the opportunity to explore the virtual data space applying the thinking aloud method. Then they were given specific tasks to solve, for instance changing the perspective, retrieving detail information about a collection item, applying filters to the dataset, or sharing their interpretation of the visualization concept. Following the test the probands were surveyed about their experience by the means of a questionnaire asking questions on the usability of specific features of the application, how they estimate their own performance in the given test tasks as well as their overall user experience.

Test subjects and summary of results

In total a sample of 8 users participated in this more comprehensive testing procedure. The probands were 75% male, 25% female, aged from 22 to 35 years with diverse occupational backgrounds. Most of them had little experience with VR stating that they had only tried it once or twice, or a few times before, yet some were more experienced with computer games.

More than the majority found the experience easy or rather easy to use and most had no problems with locomotion in the virtual space. The operation of the perspective controller was a little bit more challenging. The application of filters, obtaining detail information about collection items and comprehensibility of the data visualisation were perceived differently by the users resulting in mixed feedback. Nevertheless all users found the given task easy to solve. 7 out of 8 considered the experience to be very enjoyable or enjoyable, however there were mixed reactions concerning the insight provided, yet all reported that it offered them a fresh perspective on the dataset.

Insights from Testing

The testing observations resulted in many valuable findings. The most important learnings are summarized below:

• Regarding the perspective controller, many test users commented that plus and minus icons (as in the previous version of the prototype) would be more intuitive for them than the arrow icons for up and down. In addition, seamless zooming instead of stepwise adjustment of zoom levels for shifting the perspective between overview and detail was suggested. And the user experience of this feature might be further improved if the the user would be moved along the line of sight instead of the vertical axis. It was also mentioned that it is difficult to keep orientation when on ground level. Stronger orientation support could probably be provided by the control center or additional orientation aid such as a compass.
• Concerning filter interactions the consensus was that filters should be applied instantly. An additional option to select all or deselect all would be desired by some. And the default setting has to be a filter combination that provides an interesting selection of items.
• Most users intuitively discovered how supplementary information about collection objects can be retrieved on demand and for those who did not discover this feature on their own, it was easy to apply the corresponding interaction pattern once a hint was given. Due to the orientation challenges on ground level perspective some users suggested that it would be convenient to find out more about a collection item by the means of magnification of the selected item, triggered by simply pointing at it instead of having to approach it for this purpose. However, we hold the view that the spatial movement intensifies the exploration process.
• No attention was given to the control center. The reason for this is that it was almost invisible for most users, for some it was out of the field of view or illegible due to too small font size and others just suppressed it because the information on display was not considered to be relevant. A better solution might be to display the information only on demand, though larger in size and more prominent in position.
• The abstract visualisation style of the application was well received by the majority of the test users, yet a visual differentiation between countries and surrounding might increase clarity. Only one person expressed the desire for a higher level of detail of geographical features such as rivers, mountains and so on. Additional landmarks might offer help for orientation in the virtual world. The higher-level data visualization by the means of the choropleth map was not obvious for the majority of test user, yet after being pointed out the feature was well received and comprehensible for all.
• What may constitute legitimate reasons for the mixed reactions concerning the insight provided by the experience could be the fact that on one hand side the tested prototype only included a small portion of the original data set because the data owner hasn’t provided full access to the archive database yet and on the other hand side the intended filtering options have not yet fully been implemented. It is assumed that more data will lead to more connections and correlations between individual data points and more interaction possibilities will result in a more interesting exploration space. And the combination will eventually lead to the discovery of more insight.
• Last but not least, most users of the target group are inexperienced with VR and therefore need clear instructions about the use of the hardware (the controllers in particular). Explanation on how to navigation in the virtual space as well as how to make use of the features provided by the application are required to take full advantage of the proposed data exploration experience. The best way to learn this would be by the means of an interactive tutorial in the way it is often done in games. To compensate for the lack of such an introduction, clear instructions need to be given to the user and the operative rules have to be explained by the host. In addition it can be useful to guide the user through a specific usecase to become familiar with the application.
  A way to demonstrate usage and interaction principles in one specific use case is by applying a narrative pattern called Martini glass structure, which begins with a single-path author driven approach following a tight narrative path (representing the stem of the glass). Once the intended narrative is complete, the experience is opening up to a user driven explorative stage where the user is free to interactively explore the data (representing the body of the glass). [1]

Testing has provided many valuable insights that will feed into the further development of the prototype. Although pointing out some minor issues it has revealed that this form of spatial visualization and interaction with data provides benefits for the data exploration task, that the application is engaging and enjoyable and that the experience indeed offers a fresh perspective on the given dataset.

[1] Edward Segel and Jeffrey Heer (2010): “Narrative Visualization: Telling Stories with Data” in IEEE Transactions on Visualization & Computer Graphics.

Imprint: The content of this blog entry is based on VR experiments in the context of projects carried out in the scope of the authors masters degree studies at FHNW University of Applied Sciences and Arts Northwestern Switzerland under the supervision of Prof. Dr. Doris Agotai.

Jeremy Bailenson, founding director of Stanford University’s Virtual Human Interaction Lab, states in his recent publication that “VR is about exploration whereas storytelling is about control” [1]. Yet even while striving for offering the highest degree of freedom by handing full system control over to the user and allowing him/her to explore the virtual (data) space at is own discretion, the user needs some support for orientation, navigation and wayfinding to take advantage of this freedom, so that the exploration process may not solely be based on serendipity and the vastness of the data space doesn’t feel like a maze. .

localization, wayfinding and navigation

The authors of “Understanding Virtual Reality” define navigation in VR as “the way of traveling through the space” and wayfinding as “knowing where you want to go” [2]. In the context of exploration one might need go even one step further back, namely knowing where your are. The term localization is used to determine the user’s current position within the virtual space and give him/her a system of reference that will be an assistance for the continuation of the exploration journey.

Some practical wayfinding aids for localization and navigation include

• (world) maps provide a mental picture that is deeply rooted in the user’s mind. Information on the current position such as the country the user is located in serve as a trigger to activate existing knowledge about the surrounding.
• offering a system of reference such as geographical coordinates.
• orientation aid via the indication of true north or placing a compass at the user’s disposal.
• hints such as leaving bread crumbs, some kind of guideposts or punctual instrumented guidance.
• marking spots that have been popular with other users, because insight on an individual level can be gained from information that might not be new to other members of the community.

The purpose of such measures is to foster the discovery of insight by offering light guidance in the individual exploration process without reducing the freedom of choice of the user.

localization in the context of this project

While the user observes the data space from the bird’s eye perspective in our scenario, orientation is obvious, yet while flying over the world, the precise corresponding location on the ground might not be so obvious. Plus the real challenge is to keep orientation when the user switches perspective to ground level view. Support in this matter can either be offered directly within the virtual world, for instance by indicating the current location by the means of an avatar or marker on the ground, visually highlighting the underlying corresponding country or placing a label on it. Another approach that separates this additional information layer from the actual data space would be some kind of cockpit or control center, that provides the user with facts such as the geo coordinates of the current position, corresponding country, “flight” altitude, etcetera. In addition to localization related facts a summary of further relevant information could be offered on demand. This could include what filters have been applied to the currents selection and how many items are included as a result.

[1] Bailenson, Jeremy (2018): “Experience on Demand: What Virtual Reality is, how it works and what it can do”. W. W. Norton & Company, New York, USA.
[2] William R. Sherman, Alan B. Craig (2003): “Understanding Virtual Reality”. Morgan Kaufmann Publishers, San Francisco, CA, USA.

Imprint: The content of this blog entry is based on VR experiments in the context of projects carried out in the scope of the authors masters degree studies at FHNW University of Applied Sciences and Arts Northwestern Switzerland under the supervision of Prof. Dr. Doris Agotai.

In the previous article we explained the reasoning for some of the overall design decisions of our worldmap data exploration prototype. In this article aspects of consideration for the display of individual collection items such as the form of representation, their position on the map, the display of meta information and so forth are explained.

form of representation of the data points

Even though the collection mainly consists of physical objects, the database contains no 3D models but only photographs and metadata of the objects. On one hand side, this limits the forms of representation of the objects in virtual 3D space, yet on the other hand it poses a new challenge of finding an adequate display mode for 2D artefacts in a 3D environment. In order to create a spatial representation of the item, the photograph can be applied as a texture to a cube or a cylinder or a more common practice for this purpose is placing pictures on camera facing billboards. Concerning the camera facing mechanism, it has to be decided if all axes of the object should get aligned with the users line of sight or if the object should simple rotate around the z axis. With regard to the various perspectives the user can take from bird’s eye view to ground level view and everything in between, full adjustment of the object’s orientation is recommended.

An alternative approach would be to visualize merely the item’s meta data in the overview by the means of some abstract form of representation applying the item’s attributes to various dimensions such as size, shape, colour, transparency, etc. of a basic 3D object as it is done in many spatial data analytics tools, and only show the photograph of the item on demand. However this idea was not further pursued due to the fact that the attributes of the dataset contain mainly textual descriptions and categorical or discrete values, yet very few continuous values. Consequently there is no meaningful mapping of attributes to visual variables possible with the data at hand.

placement of objects

There are many aspects for consideration regarding to the placement of the the collection item objects and most of them relate to the issue of uncertainty visualisation because the dataset is rather imprecise in regards to the location of origin of the individual objects. For many of them only the country or broader region of origin is known, but there is no specific geographical information available, that could be translated into precise geo coordinates. This issue has been brought up previously and we have also discussed various approaches for the visualization of uncertainty.

In regards to visualization fidelity it is not advised to anchor the items in a precise location by the means of a flag pole, a pin or another kind of marker because that level of information about the location of origin simply is not known. Let the objects float over a certain area is a better expression of the level of accuracy. In addition, a random distribution factor could be added on purpose to the position. This would also prevent the probability that multiple items are located at exactly the same location. Another way of dealing with multiple objects in proximity would be some form of clustering or stacking of objects.

display of meta information

While the photograph of the item is considered to attract the highest interest of the target audience and therefore is displayed permanently, following the “details on demand principle” additional meta information should be available if requested. An intuitive interaction to express this request is simply pointing at an object.

Aspects of consideration for the display of meta information include which information should be displayed as well as the position where this information should be displayed, for example on top of the item itself, in close proximity, on some kind of menu floating in the virtual world or attached to the controller or in a fixed position on the user’s display. Further aspects include font size, font colour, background and text orientation. One has to balance these factors between readability, ensuring that the reference to the object is unambiguous, usability and aesthetics.

And on a side note, spoken text would for various reasons such as the general legibility of text in VR due to the limited resolution of current headsets for instance be a better option than text display, although audio presents its own challenges like leveling up with the noise level of the surrounding (taking into account that this application might be used in public spaces) or accessibility for users with hearing problems.

The data management of the meta information is provided through an interface, which enables automated import and export of data by the means of JSON files. Museum management database offers for each entry an object an identification number, name, short description, place and time of origin (both in textual format), material and dimension of the object as well as information about its provenance. This information is transformed into more structured data by adding additional attributes such as country code, start and end time of the time period (as numbers) to enable the filtering mechanisms offered by our prototype.

visual and interaction design

There are a few more points to mention in regards to the visual and interaction design of the collection items. But first of all it has to be acknowledged that the mARChive project by Sarah Kenderdine and Jeffrey Shaw which allows users to explore the database of the collection of Museum Victoria in Melbourne [1], has been a great inspiration for the visual design of the collection items. Even though this project is not realized in full immersive VR, learnings are applicable because they work with similar types of content and their experience is directed towards a comparable target audience.

One remaining point for discussion is the size of the collection item objects. A question of principle is whether the object’s dimensions should be fix or if it should dynamically change depending on the distance of the viewer. It can be argued that a set size is more realistic and reinforces the discovery approach as items get larger when the user approaches it. On the contrary, if the size adapts to the user’s perspective in the same way as the width of the country outlines, it can be ensured that items are recognizable at all times and a more consistent impression is conveyed. Moreover, it has to be decided if the object should have some kind of frame or border around the photo in order to stand out against the background. Such a measure for a clearer distinction seems expedient, yet the width and style should match with the surrounding.

When it comes to the filtering interaction with the objects, an additional aspect for visualization would be to somehow visually display the degree to which an item matches the selection criterions. And last but not least, some additional general interactions with the object could include to mark items as favorites, share them with others and eventually display which items are most popular among all users.

[1] Shaw, Jeffrey: “mARChive”. URL: https://www.jeffreyshawcompendium.com/portfolio/marchive, https://vimeo.com/137801550

Imprint: The content of this blog entry is based on VR experiments in the context of projects carried out in the scope of the authors masters degree studies at FHNW University of Applied Sciences and Arts Northwestern Switzerland under the supervision of Prof. Dr. Doris Agotai.

Because maps are something most users are familiar with and therefore offer a natural interface to access and explore the given data set, we chose to pursue this scenario for our initial prototype. In this article we give reason for the design decisions that were taken over the iterative implementation process of this prototype.

map representation

We have discussed various forms of representation of the world in VR space in a previous article and observed that it depends on the objective of a VR experience whether virtual globes are the best way to display global data in immersive environments or if other representations are more eligible. A significant limitation of globes is that only one hemisphere can be observed at once and another observation is that the unfamiliarity with this perspective leads to disorientation of users. Given the objective of fast perception, we concluded that the presentation of data on a flat map offering a comprehensive overview is more suitable for the given use case.

navigation in the virtual world

A further general design question regarding the map representation is whether the map should be positioned below, in front of or tilted towards the user. This question is closely related to how the user should navigate around the virtual world. Having the map on the ground allows for more natural forms of spatial movement such as free walking (thanks to the support of physical locomotion in room-scale VR setups) and locomotion by the means of teleportation. And zooming between overview and close-up view can happen by allowing the user to be lifted up and down. Flying up naturally offers a bird’s view perspective whereas lowering down brings him/her to ground level view.
In contrast, when the map is in front of or tilted towards the user, he/she would float in space and other less natural and therefore presumably less intuitive navigation concepts – for example via indirect navigation through an input device such as motion controller or joystick – would be required for all directions of locomotion, specifically the vertical and horizontal movement in front of the map as well as for the “zoom factor” which is realized by moving closer towards or further away of the map.
Some of the benefits of simpler navigation and interaction techniques in regards to usability and user experience are that they are faster perceived and mastered particularly by inexperienced VR users such as the target group in our usecase, that they require less cognitive effort in the execution and consequently offer a more pleasurable spatial experience. Additionally, being in the world and being able to freely move around offers a higher degree of immersion and overall supports better the explorative approach pursued in this project, allowing the users to gain spatial memory and understanding about the dataset and eventually discover their own insights.

design of the map

There are a number of design question in regards to the visual style of the map:

• What is an appropriate level of detail for country shapes? On a side note, it has to be considered that this is not only a design but also a performance question…
• Should the countries have outlines or gaps to indicate the borders? And if so, how do we keep a consistent visual impression as users are looking at the map from different perspectives? A gap could look immense from a close up perspective, but be hardly recognizable from far away and the same applies to outlines. To deal with this challenge a concept was applied that is similar to the zoom level technique of geo information systems where, depending on the current zoom level, cartographical material or orthophotos with a different levels of detail are displayed. Country outlines with different line thickness tailored to the viewing distance of the user are dynamically loaded to convey a consistent visual language from every perspective.
• Will the shape alone be sufficient for the user to recognize the country or is additional information such as a label with the country name, country code or a flag needed? And in case of a label, additional points of discussion include aspects such as if the full country name should be displayed and if so, in what language should the names be displayed (e.g. english or local language)? How to deal with autonomous regions like Tibet for example? Where to display the label in the case of countries with a rather small surface? How can the reference to the corresponding country be ensured if placed next to the area or would the country code alone be understandable for the majority of users? How can the legibility be ensured given that the user looks at it from different angles, distances and changing background conditions. Some approaches to deal with this issue is to to adapt the font size to the users distance and to adjust the orientation of the text towards the viewing direction of the user, which is a technique that is often used in VR and commonly referred to as camera facing billboards.
• Additionally it needs to be determined what is the right measure of information density in each situation and for the given use case in general. For example should country labels only be displayed for countries that are of particular interest for a user at that moment (for example those closest to his/her position or those with the largest amount of collection items)? Or should collection items which are close to each other be displayed as clusters in the overview perspective in order to to prevent information overload?

data visualisation and interaction with the dataset

Our data visualisation concept aims to combine the visualization of information about the dataset as a whole with detailed information about specific data points and therefore operates on two layers: the former is incorporated into the map visualization and the latter is represented in the display of the individual collection objects as shown in the following figure.

In order to provide a fast overall impression about the contents of a given dataset, aggregated general information about the dataset such as the prevalence of objects matching certain criterias is directly embedded into the map representation itself in the form of a Choropleth map colour encoding in combination with volume height adjustments. While visualizing quantities through the height dimension might seem to be the most obvious form of spatial data visualisation, it should be noted that this form of expression alone might be difficult to read because scale variation makes judging distances, heights and areas in 3D space harder, and occlusion might at times remove relevant information from the field of view. Therefore, a combination of spatial data representation together with a visual encoding in the texture of the surface of a geographic object provide a more effective solution allowing fast interpretation of data. In concrete terms, this means that the texture of countries with many collection items on display is highlighted and the height of the 3D models of these countries is elevated.

Additionally, interaction with the dataset is an essential factor for gaining a better understanding of its content. Our approach is to allow the user to filter the dataset by various criterions such as time span of creation, material, geographical origin, etc. and apply the filter results in real time to the global data visualization rendering concept explained above. And in addition to the overall presentation, collection items that are included in the selection are highlighted by being expanded. This immediate spatial representation of filter results offers rapid insight on the contents of the dataset to the user.

visual design

And last but not least there are the obvious design questions regarding the general aesthetic such as overall style, colour schema, lighting mood and so forth which exhibit a certain degree of subjectivity and can be rather hard to measure. Filonik and Baur have researched the topic of measuring aesthetics for information visualization and conclude that “aesthetics is an unsolved problem of information visualization, because there is no satisfactory understanding of what constitutes aesthetic effect”. Yet they also point out that empirical studies have shown a correlation between perceived aesthetics and usability. [2] And Katja Kwastek states in Interaction in Digital Art that “materiality and interpretability should be understood as complementary components of aesthetic experience” [3] confirming that the choice of material and style can contribute to the comprehensibility of the data on display.

The leading design principles for our VR experience are simplicity, information density, focus and comprehensibility.

Simplicity in a broader sense applies to being intentional on how much visual detail information is really required. Lindeman and Beckhaus who coined the term “experimental fidelity” encourage VR creators to carefully curate the user experience yet not by increasing the level of detail, but rather by reducing it to the essential components [4]. And Salen and Zimmerman address this issue in the domain of computer games and use the term “immersive fallacy” to denote the widespread misbelief that illusions should be as realistic as possible [5]. For this reason we intend to apply an adequate level of detail in the visual design of the components. For instance not every minor detail of the world map is needed, simplified shapes and outlines fulfill their purpose (and might improve the performance as a side effect).

Simplicity concerns not only the visual aspect but also an adequate level of information density. According to Claudia Giannetti, the theory of apperception states that a “oversupply” of information causes irritation of the user, whereas “undersupply” can lead to boredom. [6] Dealing with larger amounts of data, the lower bound is never an issue, the challenge is to show just the right amount of visual information to the user to prevent a cognitive overload. It is interesting that the author subsequently recommends to use a multilayer model to keep the right balance between the two extremes which is exactly the approach taken with this prototype.

The virtual environment should not take away the user’s attention from the actual content and colour can plays an important role in guiding the user’s focus. Therefore only the collection items are represented in their original colour to be in the center of attention whereas the chromaticity of the surrounding has deliberately been reduced to grayscale with low luminosity for countries with little or no prevalence of collection items and increasing brightness for countries with higher concentration of collection items.

Last but not least emphasis should be given to comprehensibility. It is important that the chosen forms of visualisation can be interpreted by the target audience, which are by the majority not expert users nor data scientists. While we strive for simplicity and higher degree of abstraction might offer interesting visual appeal, employing familiar forms of representations allow to tap into prior knowledge. For this reason a conventional world map was chosen to convey the geographical context.

[1] GisGeography (2018): “Choropleth Maps – A Guide to Data Classification”. URL: https://gisgeography.com/choropleth-maps-data-classification/
[2] Filonik, Daniel and Baur, Dominikus (2009): “Measuring Aesthetics for Information Visualization”. Information Visualisation, 13th International Conference. 579 - 584.
[3] Kwastek, Katja (2013): “Aesthetics of Interaction in Digital Art”. MIT Press, Cambridge, Massachusetts.
[4] Lindeman, Robert W. and Beckhaus, Steffi (2009): “Crafting memorable VR experiences using experiential fidelity”. Proceedings of the 16th ACM Symposium on Virtual Reality Software and Technology, Seiten 187-190. ACM, New York.
[5] Katie Salen and Eric Zimmerman (2004): “Rules of Play - Game Design Fundamentals” MIT Press, Cambridge, Massachusetts.
[6] Giannetti, Claudia (2004): “Ästhetik des Digitalen: Ein intermediärer Beitrag zu Wissenschaft, Medien- und Kunstsystemen”. Springer Verlag, Wien/New York.

Imprint: The content of this blog entry is based on VR experiments in the context of projects carried out in the scope of the authors masters degree studies at FHNW University of Applied Sciences and Arts Northwestern Switzerland under the supervision of Prof. Dr. Doris Agotai.

Interaction and UI design for virtual environments are still a rather novel discipline where not many standard design patterns have been established yet. This leaves much room for exploration and therefore present an interesting challenge. In this article we will look into theory and current research on this topic and investigate if proven principles from 2D interaction and UI design can be transferred to virtual space.

VR pioneer LaViola et al. write in the introduction of their reference book on 3D User interfaces that „although we live and act in a 3D world, the physical world contains many more cues for understanding and constraints and affordances for action that cannot currently be represented accurately in computer simulation. Therefore great care must go into the design of UIs and interaction techniques for 3D applications. It is clear that simply adapting traditional WIMP interaction styles to 3D does not provide a complete solution to this problem. Rather, novel 3D UIs based on real-world interaction or other metaphors must be developed.“ [1] With this statement the authors point out the importance of developing adequate interaction patterns and menu designs for virtual environments.

Basics of spatial interaction design

Concerning types of interaction, 3D interaction theory distinguishes between direct user control, virtual control and agent control. In the first case, interface gestures that mimic real world interaction are applied meaning that the user directly interacts with a virtual object as he would with a physical object. Interaction fidelity is a term that is frequently used in this context to describe the objective degree of exactness with real world actions [2]. An example for high interaction fidelity is the isomorphic outlook, where strict geometrical, one-to-one correspondence between hand motions in the physical and virtual worlds is suggested, reasoning that more natural interaction patterns are better for users. Yet this is often impractical and ineffective because of user’s limitations such as arm length meaning he/she can only reach items within that limited radius. In the second case of virtual control, system control is effected by forms of indirect manipulation, for instance by the means of menus, commands, physical controllers or laser pointers among many others.

Another general question regarding interaction design is whether interactions should be effectuated single handed (unimanually) or with both hands (bi-manually) and in the latter case, if the actions of the two happen synchronously or asynchronously and an additional distinction according to Guiard’s framework for classification is made depending on the role of the dominant and the non-dominant hand. Studies have shown that bimanual input in particular has the potential to allow users to perform tasks faster, leverage existing skills, and increase expressiveness [3]. However it is also more difficult and time-consuming to acquire that skill and input gestures may be prone to errors if they are not carried out properly. Consequently, bi-manual control is recommended for VR experiences targeted to expert users who will presumably use that application recurringly, whereas for single usage and target groups with little experience in VR unimanual interaction patterns are more sensible.

abstract, real and magical 3D user interfaces

When designing 3D user interfaces, the degree of abstraction is a big question. Some school of thoughts believe that the closer an interface or a virtual environment resembles the real world, the easier the usage. The grounds for this assumption are that in a more realistic setting, novice users can transfer precognition from real life to the virtual environment and recognize affordances from their everyday life experience. However if the virtual representation differs in some minor aspect from the user’s expectation, the transfer learning effect is lost.

In fact, it has long been argued that the real power of 3D UIs lies not only in imitating physical reality and simulating real-world features, but also in creating a “better” reality by utilizing magical interaction techniques. By creating mappings and interactions that are specifically tailored to 3D environments, actions and processes can be made easier than in real life and an advantage of VR is that it allows users to overcome many human limitations that are so prominent in the real world such as limitations of our cognitive, perceptual, physical and motor capabilities [4].
In some case this is done by enhancing the user’s natural capabilities (e.g. offering the ability to fly to allow the user to view the world from bird’s eyes perspective). Yet in other cases, constrained interfaces that are designed to be simpler than the real world by restricting movement, limiting interface action and keeping interface objects in a plane are more suitable for a given task. An intriguing possibility is that enhanced 3D interfaces might offer simpler navigation, more compelling functionality, safer movements and less occlusion, than 3D reality, especially for information exploration and visualisation tasks. In conclusion, clever 3D designs prioritize facilitation and simplification of user tasks over mimicking reality [5].

Placement of Menus in VR

According to the initial frame of reference for the design of virtual menus offered by Jacoby and Ellis, invocation, location, and reference frame are considered as the most important aspects [6]. In the virtual space, there is no menu bar as it is the standard in desktop applications nor are there any borders or corners to which the menu could be attached to. The term spatial reference is used to describe the placement of the menu in virtual space. The distinction is made between world referenced (attached to a specific location in the virtual space), object referenced (attached to an object), head referenced (showing the menu in a constant ratio to the users position, for instance on a “head up display” at a fixed position of the users field of view) and last but not least body or device reference (for example attached to the hand controller or wrist of the user).

Related to menu placement is the issue of occlusion. Is a menu permanently shown at a fixed position of the screen, it might hide other relevant information from the user’s field of view. The challenge lies in finding the right balance between visibility of the content and accessibility of the menu. For this reason, it is recommended to display menus prominently when needed, for example by evocation on demand. Although a concern with on demand is the following question: how will the user know about the functionality and remember the evocation mechanism when requested? The distinction is made between recall and recognition. Recall refers to knowledge in the head and recognition refers to knowledge in the world. In the context of this issue recall would mean that the display of the menu is evoked by pressing a certain button, applying a gesture or voice control command whereas recognition means that there is some memory aid in the virtual environment such as an icon, object or another visual representation, that indicates the on-demand availability of the menu. The latter is recommended because it lowers the cognitive load placed on the user.

3D Menu design

While over the past decades of software development interaction patterns for application control techniques in 2D applications were established, there is not yet a proven set of 3D menu techniques available to designers and developers. Oftentimes, well-known desktop interaction techniques are adapted to virtual environments, although menu solutions integrating 2D approaches into space face problems such as the greater skills required in reaching a menu item in space as well as the lack of haptic constraint and tactile feedback. Freehand menu selection in particular is an inherently difficult task, especially with increasing menu breath, and free moving hands in space cannot achieve the same precision levels as with physical input devices such as mouse or stylus. For this reason effectiveness is severely reduced when adapting a traditional 2D drop-down menu to virtual environments, plus it leads to exhaustion.

Dachselt and Hübner devised a taxonomy with classification criterias for 3D menus which includes the following aspects [7]:

• intention of use: number of displayed items (e.g. limited or not, range or definite value), hierarchical nature
• appearance and structure: geometric structure, structural layout, type of displayed data (e.g. text, image, 3D object, combination), size and spacing of items
• placement: frame of reference, orientation (e.g. always facing user), repositioning
• invocation and availability: visibility (e.g. whole time, temporarily, user-dependent), invocation, animation, collapsibility
• interaction and I/O setting: interaction device dependence, application type and setting, dimensionality, feedback (e.g. highlighting selection, audio cue, etc.), visualization of selection path
• usability: evaluation criterions such as selection speed, error rate, efficiency, user comfort, ease of use and learning as well as comparison with other menu solutions
• combinability

Some prominent early examples of menus that were deliberately designed for immersive virtual environments include inter alia the TULIP Menu and rapMenu.

TULIP is a menu system using Pinch Gloves for tracking the position of the user’s hand. It’s name stands for “Three-Up, Labels In Palm”, meaning that three items are active at one time while the rest of the menu items are arranged in columns of three along the palm of the user’s hand. To access an active item, the user simply pinches the thumb to the appropriate finger. To access other items, the user pinches the thumb to the little finger until the desired item appears on one of the fingers [8, 9].
A further evolution of the TULIP menu is Leap Motion’s Hovercast menu, which radiates from the palm of the user’s hand. A wide arc of menu items extend his/her fingertips and follow the hand movement. The user can interact with menu items using the index finger of the opposite hand. On rotation of the palm towards the eyes, the menu fades into respectively out of view. Hovercast is highly customizable, and can include many nested levels of selectors, toggles, triggers, and sliders. [10]

The rapMenu (roll and pinch menu) developed by Ni, McMahan and Bowman is a hybrid of the best features of the ring and the TULIP menu. It is based on gestural commands, by rotating the wrist the user makes a pre-selection of a sector with four items and then selects one out of them via pinch gesture. The menu works for up to 16 options (constrained to groups of four) and can be extended by the means of nesting of hierarchies. To prevent unintended menu invocation, gestural commands are only interpreted when the user’s hand is point at the display which reveals, that this menu was not designed for immersive VR environments. However, this “effective zone” could be substituted by another mechanism. User studies comparing pie menus to linear menus have shown, that radial placement reduces target selection time and lowers error rate. [11]

Guidelines for 3D UI design

Shneiderman, who is also the author of the originator of the Visual Information Seeking Mantra, points out that advanced 3D UI designs are marked by their support of (1) rapid situation awareness through effective overviews, (2) reduced numbers of actions to accomplish tasks, and (3) prompt, meaningful feedback for user actions. He offers the following guidelines for 3D UI design [5]:

• minimizing the number of navigation steps for users to accomplish a task
• concern for readability of text, in particular good contrast with background and minimization of tilt orientation
• avoidance of unnecessary visual clutter
• simplification of user and object movement
• error prevention
• organisation of information or groups of items in structures that allow rapid visual search and enable spatial recall
• enriching interfaces with haptic feedback and audio cues might provide additional benefits
• overviews to provide users with big picture of an application
• permit user actions on objects
• give user control over amount of information, offer details on demand
• implement dynamic queries to rapidly filter out unneeded items
• recognizable and memorable icon designs

Applying these twelve guidelines to the implementation of a 3D UI increases the usability and user experience of a spatial environment.

Application to the Worldmap Data Exploration Prototype

Applying these learnings to the context of our prototype, that has been built on the hypothesis that real user value is created when data is presented in a meaningful way and made accessible for exploration and discovery, notably if the user is able to interact with it, and combining it with the previously presented Visual Information Seeking Mantra “overview first, zoom and filter, then details on demand” which also points out important aspects for interaction with data, leads to the following interaction concepts:

• perspective control: facilitating a mechanism to switch between overview and detail, allowing to shift the user’s perspective back and forth between bird’s eye view (presentation of the dataset as a whole) and ground level view (individual collection item and its surrounding)
• importance of filtering mechanisms: filtering is essential in order to make sense out of large amounts of data, therefore the application should allows to apply various filter combinations such as category, material, technique, purpose, origin, epoche, objects on display in current exhibition, etc. to the dataset
• providing details on demand: allowing to interact with specific data points and displaying additional information for selected items. Further interaction possibilities could include marking favorites or comparing the behavior of multiple selected objects

Concerning interaction control, it is estimated that body referenced controls attached to the handles are the best solution for this use case as they minimize occlusion of view yet remain available anytime and are therefore easy to recall. And for menu design, given the many benefits we came to know through the research above, radial menus should be explored.
In any case, rapid prototyping should be applied and great attention be given to ongoing testing while iteratively devising interactions and menus suitable for the given task.

[1] Joseph J. LaViola, Ernst Kruijff, Ryan P. McMahan, Doug A. Bowman, Ivan Poupyrev (?): „3D User Interfaces - Theory and practice“. Addison-Wesley Educational Publishers Inc, New Jersey, USA.
[2] McMahan, Ryan Patrick (2011): ”Exploring the Effects of Higher-Fidelity Display and Interaction for Virtual Reality Games”. Virginia Polytechnic Institute and State University, Blacksburg, VA.
[3] Russell Owen, Gordon Kurtenbach, George Fitzmaurice, Thomas Baudel, Bill Buxton (2005): “When It Gets More Difficult, Use Both Hands – Exploring Bimanual Curve Manipulation”. GI 2005 Conference proceedings: Graphics Interface Conference, pp. 17-24.
[4] William R. Sherman, Alan B. Craig (2003): “Understanding Virtual Reality”. Morgan Kaufmann Publishers, San Francisco, CA, USA.
[5] Shneiderman, Ben (2003): „Why not make Interfaces better than 3D reality?“. IEEE Computer Graphics and Applications, Volume: 23, Issue 6.
[6] Jacoby, Richard H., Ellis, Stephen (1992): „Using virtual menus in a virtual environment“. Proceedings of SPIE - The International Society for Optical Engineering.
[7] Dachselt, Raimund and Hübner, Anett (2007): “Three-dimensional menus: A survey and taxonomy”, Computers & Graphics, Volume 31, Issue 1, Pages 53-65.
[8] D. A. Bowman and C. A. Wingrave (2001): "Design and evaluation of menu systems for immersive virtual environments". Proceedings IEEE Virtual Reality 2001, Yokohama, Japan, pp. 149-156.
[9] D. A. Bowman, C. A. Wingrave, J. M. Campbell, V. Q. Ly, C. J. Rhoton (2002): “Novel Uses of Pinch Gloves™ for Virtual Environment Interaction Techniques”. In “Virtual Reality”, Volume 6, Issue 3, pp 122–129. Springer, London, UK.
[10] Kinstner, Zach (2015): “Hovercast VR Menu: Power at Your Fingertips”, URL: http://blog.leapmotion.com/hovercast-vr-menu-power-fingertips
[11] Ni, Tao, McMahan, Ryan, Bowman, Doug (2008): “Tech-note: rapMenu: Remote Menu Selection Using Freehand Gestural Input”. 3DUI - IEEE Symposium on 3D User Interfaces 2008.

Imprint: The content of this blog entry is based on VR experiments in the context of projects carried out in the scope of the authors masters degree studies at FHNW University of Applied Sciences and Arts Northwestern Switzerland under the supervision of Prof. Dr. Doris Agotai.

It is in the nature of cultural heritage data that datasets are often not complete or information is imprecise. We have briefly covered some of the challenges related to this issue in a previous blog post and will discuss this topic now more thoroughly.

information uncertainty

In natural science the term “uncertainty” is used to describe doubt about the validity of the result of a measurement. This could be due to for example an error in measurement, transmission or conversion. In the context of cultural heritage uncertainty is rather referring to missing information or imprecision of data. And more generally, information uncertainty refers to the impreciseness, vagueness, fuzziness, inconsistency, doubt or likelihood present in information. According to Pham et al. some of the sources of information uncertainty include limited accuracy in data entry or transformation (for example limited precision), missing data, incomplete definition, imperfect realisation of a definition, inadequate knowledge about the effects of the change in environment, personal bias, ambiguity in linguistic description, approximation or assumption embedded in model design methods or procedures [1]. Consequently, visualization fidelity describes the accuracy and exactness in the visual representation of such information uncertainties.

uncertainty visualization

According to Pang et al., uncertainty visualization is the “strive to present data together with auxiliary uncertainty information. These visualizations present a more complete and accurate rendition of data for users to analyze. [...] The ultimate goal of uncertainty visualization is to provide users with visualizations that incorporate and reflect uncertainty information to aid in data analysis and decision making. [2] Although information uncertainty brings more complexity to the visualisation problem, it is crucial to consider this issue in data visualisation and exploration. Brunet and Andújar state that “most immersive (and non-immersive) visualizations nowadays are fundamentally biased and can be even unreliable“ because “data is usually presented as realistic and plausible 3D information” and “uncertainty is largely ignored by present applications and virtual environments” [3]. And MacEachren stresses that despite the long history in uncertainty visualization research, there is no generally accepted strategy for leveraging visualization to cope with uncertainty. He suggests to take a visual analytics approach to tackle the larger challenge which is not just to present data accurately but ultimately to enable reasoning under uncertainty in all its forms. [4]
All sources agree that a visual distinction is required for portions of a dataset or model that contain aspects of uncertainty. Let’s have a look at the domain of archeology for learnings and inspiration how this can be done.

Learnings from Archaeology

Dealing with uncertainty and incompleteness is by nature a common task in the field of archaeology. Yet the emergence of 3D visualization tools resulted in a strive for an increasingly higher level of detail in digital reconstructions of historical and archaeological artefacts. Though the resulting almost photorealistic visualisations did according to Roussou and Drettakis not improve the experience but rather led to an information overload and diverted the users attention to irrelevant details. And additionally, such hypothetical renderings carry a deceptive connotation of a higher level of information instead of being truthful about the actual fragmented knowledge. In the meanwhile, the trend has shifted towards less photorealistic, yet more credible forms of visualization that place greater emphasis on the differentiation of facts and assumptions. this evolution has been labeled by the authors as “realness factor” [5]. Also Gershon puts emphasis on the importance of an accurate representation of the degree of incompleteness and uncertainty of information. He recommends to demonstrate this by purposefully choosing a lower level of detail or using visual metaphors conveying this notion of incompleteness [6]. In «Fantastic reconstructions or reconstructions of the fantastic? Tracking and presenting ambiguity, alternatives, and documentation in virtual worlds» [7] the authors describe a set of dynamically adjustable variables including color schemes and opacity levels to render images that are more transparent about the presence of ambiguity, evidence, and alternatives in virtual reconstructions. And Strothotte, Masuch and Isenberg make the concrete proposal to visualize the degree of uncertainty in reconstructions by visual variables such as line width, ductus or sketchy layout style and suggest to purposefully leave out details [8].

Robert W. Lindeman and Steffi Beckhaus share the viewpoint of reduction to the essential and introduce the term «Experimental Fidelity» which they define as follows: «Experiential Fidelity is an attempt to create a deeper sense of presence by carefully designing the user experience. We suggest to guide the user‘s frame of mind in a way that their expectations, attitude, and attention are aligned with the actual VR experience, and that the user‘s own imagination is stimulated to complete the experience.» [9]
All sources agree on the importance of visualization fidelity, and some also emphasize to focus on the essential which leads to better comprehension and prevents cognitive information overload.

Transfer into practice

How can the principles and learnings listed above be applied to projects in practice? It first starts with being conscious of the issue of information uncertainty, then carefully examining the dataset and questioning the apparent. A next step is searching for suitable forms of visualization in respect to this. Some techniques suggested by Pang et al. to make users aware of locations and degree of uncertainties in their data include adding glyphs, adding geometry, modifying geometry, modifying attributes, animation, sonification, and psychovisual approaches. [2] However there is no uniform solution and what visualisation strategies will be most expedient depends largely depends on the dataset, the presentation medium, goals to be achieved and last but not least the target audience. For this reason it is crucial to finally test and measure if the visualization concepts are understood by the target audience.
Some specific aspects where the topic of uncertainty visualisation needs to be considered in this project:

• geographical origin of objects: there are no geo-coordinates provided for the origin of the collection items and the textual description of the location of origin given in the dataset is rather vague. In many cases only the name of a region, province or even just the country of origin is stated. Anchoring items with a precise location marker such as a pin or a flag onto the map would imply a higher level of information than there is available. Some approaches to ensure visualization fidelity in this scenario would be to use a diffuse material (for example smoke, vapor or some kind of a particle system) or a rough sketched line for the anchor or leave out the fixture entirely and let the item float over an area.
• age determination: for most objects the dating is not a specific point in time but a textual description of a rather vague time interval such as an estimated time span, a century or an epoque. In order to dynamically process this information it has to be converted into an interval of numeric values. As in the example above, this implies a higher level of information than what is actually available. And when positioning items on a timeline it has to be considered if they will be displayed as a point and if so where this point would be located (start date, end date, mean value...) or rather as a vector with diffuse beginning and end points to reflect properly the information uncertainty aspect of this attribute.

Conclusion

There is not the one and only model solution how to deal with the issue of visualization of uncertainty, it always depends on the context, the objective and the target audience. Though it always starts with careful examination of the data at hand and challenging the obvious. And whilst aiming at the highest possible level of visualization fidelity one should at the same time be sensible to the user’s cognitive load. The goal is to find intuitive ways for the visualisation of uncertainty we are not just throwing even more information at the user.

[1] Binh Pham, Alex Streit and Ross Brown (2009): Chapter “Visualization of Information Uncertainty: Progress and Challenges” in "Trends in Interactive Visualization, State of the Art Survey".
[2] Pang, A., Wittenbrink, C., and Lodh, S. (1997). Approaches to uncertainty visualization. Visual Comput. 13, 370–390.
[3] Brunet, Pere and Andújar, Carlos (2015): “Immersive Data Comprehension: Visualizing Uncertainty in Measurable Models”. Frontiers in Robotics and AI. URL: https://www.frontiersin.org/articles/10.3389/frobt.2015.00022/full
[4] MacEachren, Alan M. (2015): “Visual Analytics and Uncertainty: Its Not About the Data”. In Proceedings EuroVis Workshop on Visual Analytics (EuroVA). The Eurographics Association.
[5] Roussou, Maria and Drettakis, George (2003): “Photorealism and Non-Photorealism in Virtual Heritage Representation”. Proceedings of the International Symposium on Virtual Reality, Archeology and Cultural Heritage. Eurographics, Brighton, UK.
[6] Gershon, Nahum D. (1998): Visualization of an Imperfect World. IEEE Computer Graphics and Applications, p.43-45.
[7] Kensek, Karen M., Swartz Dodd, Lynn and Cipolla, Nicholas (2004): “Fantastic reconstructions or reconstructions of the fantastic? Tracking and presenting ambiguity, alternatives, and documentation in virtual worlds”.. Automation in Construction 13, Seiten 175-186. Elsevier, Amsterdam.
[8] Strothotte, Thomas, Masuch, Maic and Isenberg, Tobias (1999): “Visualizing Knowledge about Virtual Reconstructions of Ancient Architecture”. IEEE Computer Society. Proceedings Computer Graphics International, Seiten 36-43. The Computer Graphics Society, Los Alamitos, CA.
[9] Lindeman, Robert W. and Beckhaus, Steffi (2009): “Crafting memorable VR experiences using experiential fidelity”. Proceedings of the 16th ACM Symposium on Virtual Reality Software and Technology, Seiten 187-190. ACM, New York.

Imprint: The content of this blog entry is based on VR experiments in the context of projects carried out in the scope of the authors masters degree studies at FHNW University of Applied Sciences and Arts Northwestern Switzerland under the supervision of Prof. Dr. Doris Agotai.

To visualize data means to form a mental model or image of something in order to gain insight. One particular way for doing this is by mapping, which is defined as “creating a graphic representations of information using spatial relationships within the graphic to represent some relationships within the data. The common and original practice of mapping is the scaled portrayal of geographical features, that is, cartography. In the contemporary sense of data visualization, it includes metaphorical extensions of geographical map conventions and literacies to other kinds of data, as well as innovative ways of visualizing data not clearly related to the geographical archetype. In popular vernacular, mapping can just mean organizing or systematizing information” [1]. Hans Ulrich Obrist further defines maps as “an abstraction of the physical or conceptual world – a symbolic depiction of a space or idea that allows one to understand and navigate an unfamiliar topography or complex topology. But while most conventional charts, plans and diagrams claim to offer an accurate, even objective picture of the world, each one is bound by the specific agendas of its creators and users” [2]. He points out that there is no neutral way for mapping data and stresses on the importance to carefully consider and choose factors such as visualisation methods, styles, type, medium and context when representing data because they influence how the data is perceived. Jay David Bolter writes in the context of medium and materiality that “the computer is not a neutral space for conveying information. It shapes the information it conveys and is shaped in turn by the physical and cultural worlds in which it functions” [3].

Geographical Maps

Maps and globes are widely used for visualizing qualitative or quantitative data with a spatial aspect. The choice of map projection type is one of the initial steps in the quest to represent the world, a sphere, on a flat planar surface. No projection is neutral, natural or given, they are all constructed, configured and underpinned by various and quite arbitrary conventions. The main distinction is made between conformal maps like Mercator’s projection and equal area maps like the Gall-Peters projection. The former has the appealing property of faithfully representing the shapes of regions on the map and has served as standard for atlases for centuries. It is fine around the equatorial areas, but it greatly exaggerates the sizes of features as it nears the polar regions. [4] The later depicts sizes accurately, the area covered by a feature on the map, is proportional to the true land area of the feature in real life, but now the shapes are distorted, again this distortion is greatest near the poles.

In between the two of them lies another common choice, the equidistant cylindrical projection (plate carrée). This projection gets neither the areas nor the shapes of map features correct, but adopts a compromise solution that distorts neither areas as badly as Mercator nor shapes as badly as Gall-Peters. [5] There are many other projections of the world, but all of them inevitably distort our view of the world because it is not possible to draw the surface of the earth on a 2D surface without significant spatial distortion.

Maps and Globes in VR

In contrast to the challenges of representing the world on a 2D medium outlined above, VR would allow to represent the world correctly in its original spherical shape. A key question however is if virtual globes are the best way to show global geographic data in immersive environments or whether maps or some other visualisations may be better. Some of the drawbacks are that a globe can only show one hemisphere which is a limitation for many global visualisations and another observation is that the unfamiliarity with this perspective leads to disorientation of users. In their paper “Maps and Globes in VR” the authors explore different ways to render world-wide geographic maps in VR. They compare the pros and cons of the 3D exocentric globe (user’s viewpoint outside the globe), the egocentric globe (viewpoint inside the globe) and the flat map and it’s variant the curved map (rendering map on a plane or section of a sphere). [6]

Other Types of Maps

As stated in the introduction, mapping can refer to any kind of systematic information visualization where data attributes are bound to visual encodings. Erik Champion states that “maps are not just instrumental artefacts, but also epistemic ones with a long history that have helped to organize our knowledge of the world for millennia.” [7] Besides of geographical maps there is a wide variety of other map types. For instance cartograms, a rather recent invention, are digitally modified maps which depict the areas and countries of the world not by their physical size, but by their demographic importance on a vast range of subjects. Each territory on a map displays its data graphically, becoming larger or smaller in proportion to other areas, which are in turn scaled according to their own data. On a population cartogram a country with twice as many people as another is drawn twice as large. This approach creates unpredicted views of the world and allows new perspectives for mapping the social dimension of our planet. [5, 8] In the context of the Serpentine Gallery Map Marathon Obrist even goes as far as denoting mapping as a curatorial device. [9]