// HCU visualizes

On Thursday, October 20th, 18h, at HCU Campus City Nord we will give a warm welcome to Hansi Senaratne from University of Münster!

We are looking forward to talk to her about her work on the field of usability of uncertainty vis:

This research has made its means to provide the users of uncertainty visualisation methods, an ease of selecting suitable methods to visualise their spatio-temporal uncertain data. Usability varies between different uncertainty visualisation methods as well as between users of different background experiences who are utilising these methods to visualise their spatio-temporal uncertainties. Thus, it was realised that, even if the users are provided with which methods to use, from a categorisation based upon their data type, uncertainty type, data format and preferred interaction type, these methods were not necessarily usable by all kinds of users. Building upon this, a web-based usability study was conducted on selected uncertainty visualisation methods on their learnability aspect. 81 participants took part in this study and were categorised into the user domains: map visualisation, urban planning, decision support, GIS and statistics. Any participant not belonging to either of these domains was grouped as “Other”. Through a measurement of correspondence between their performance and preference, the most suitable uncertainty visualisation method(s) were derived for each user domain.

Subsequently, the categorisation of spatio-temporal uncertainty visualisation methods was implemented in a web application, the Uncertainty Visualisation Selector. From the usability study, user domain as another parameter was added to the implemented design of the categorisation in order to further characterise the methods, in terms of their suitability to individual user backgrounds. The Uncertainty Visualisation Selector enables the user to specify the data requirements (data type, data format and uncertainty type) and user requirements (domain he/she belongs to) upon which, suitable uncertainty visualisation methods are generated through realising the categorisation.

Good news ev'ryone, our next meeting is coming up:

Gather at HCU Campus City Nord on Tuesday, July 26th, 18h!

This time Tonio gives an introduction to visual metaphors and spatialization:

Throughout the centuries maps have proven to be a powerful tool to display information of the most diverse tyes and sources simultaneously. Indeed, they have become so outstanding at this task, that they are also used to display information which is not of an originally spatial nature. Instead, information from high-dimensional abstract information spaces can be mapped onto a spatial layout. This process is also referred to as spatialization.

The mapping of an abstract property onto a less abstract – e.g., spatial - property is a metaphor. Although we are often not aware of them, metaphors structure how we perceive and reason about those things which are either not immediately palpable or trivial. For understanding complex cocepts, we use complex metaphors which consist of several individual metaphors.

When we want to depict complex information spaces as landscapes, we are employing complex landscape metaphors. In these metaphors, it is important to structure the metaphors in such way that a consistent mapping is achieved. The overall goal of a successful spatialization must be to present information from abstract spaces to a user in a fashion that it becomes intuitively understandable.

In the presentation I want to further dig into this topic by showing examples, pointing out pitfalls in spatialization design, and discussing the potential of employing new metaphors to communicate special kinds of information.

Interested? Feel invited and join us!

...and bring some drinks as well ;-)

Come to D105 at HCU Campus City Nord on Wednesday, April 27th, 18h!

Let’s discuss our prototypes!

We assume that the need for visualizing spatiotemporal information will further increase, e.g. for change detection and analysis. This information is often displayed in animated maps, but the problem with animated maps is that the users’ working memory might not be able to process the incoming stimuli if the amount of information is high, which is likely in maps. Also the fact that our attention can only be directed to small parts of a map to follow the animation and thus to recognize the change in the map shows that the concept of animated maps has to be reconsidered.

Therefore we suggest integrating sound into animated maps. The usage of sound for displaying quantitative information leads to a multimodal communication of map content. We assume that it facilitates a more comprehensive and flexible communication of geoinformation and helps to overcome so called “change blindness”.

In this session we would like to introduce and discuss our first prototypes and additional options to convey quantitative information as sound.

Check out Beate's ideas about personalizing maps on Wednesday, March 30th, 18h at City Nord, D105!

Let's discuss her research questions:

The following describes the situation we are facing when we use a map: „If two people with the same task in mind both request maps of the same region, current interactive map applications will typically present both users with exact the same map“ (Wilson 2010: 742). This is “communication of geoinformation in an unbiased way to diverse users” (Meng 2003) and “can easily hinder individual users’ particular task goals by including irrelevant spatial feature content” (Wilson 2010: 742).
To make maps more efficient, effective and their handling satisfying we suggest personalized maps that adapt to a specific user, her interaction, her technology in use or her information demand. Personalized maps are supposed to focus on the users' needs and preferences and thus present information in an appropriate form to the individual user.
The aim of this presentation is to give an overview of strategies for map adaptation and personalization and to discuss possible research questions in this research field.

Meng, L. 2004. About Egocentric Geovisualisation. In: Proceedings 12^th international Conference on Geoinformatics.

Wilson, D., Bertolotto, M. and J. Weakliam. 2010. Personalizing map content to improve task completion efficiency. International Journal of Geographic Information Science, Vol. 24, No. 5, May 2010, pp741-760.

See ya there (and bring some drinks)!

See you on Wednesday, Feb 2nd, 18h at City Nord, D105!

Christoph is developing a software prototype for the analysis of changes in remotely sensed data with the help of visual analytics methods.

What it's all about:

Methods for landscape monitoring play a major role for management and planning in a number of disciplines, e.g. in climate research, environmental science and urban planning. The analysis and documentation of spatio-temporal changes form the basis for such observations in which, apart from spatial relations, aspects of time can also be highly relevant. Depending on the application, diverse temporal scales can be of interest ranging from long-term changes (e.g. geologic processes) to relatively rapid developments (as in disaster management). A great many of these analyses are conducted based on remotely sensed data. Especially satellite based remote sensing data providing large-scale coverage of the earth at relatively low costs are widely used. For them, new fields of application are emerging since their spatial, spectral, radiometric and temporal resolution have been increased.

Our approach is to move away from common purely (semi-) automatic methods and put the user into the center of an iterative process. Visual analytics combine automated processes with user-centered, visual methods.

For this, we would like to gather ideas about how the change of topographic objects over time can be visualized in abstract spaces, e.g. changes in their appearance (spectrality, geometry) and their uncertainty. The aim is to communicate to the user visually how objects have changed.

First, we will have a look at typical examples for changes in remote sensing data and talk about the drawbacks of existing change detection / change analysis methods. Second, we will have a look at the specific task of visualizing the changes.

Prepare for a creative brainstorming!

Next: Johannes on 'Map styles and map perception'.

See you on Wednesday, Oct 13th at 18:00 @ City Nord, D105!

And: Still, we are open to everyone interested in Vis issues!

---

Map styles and map perception

How does the style of a city map affect our cognitive image of the depicted geographic space?

In most cases we use city maps for (re-)orientation when we are inside the respective urban area. But sometimes we also use them to plan a trip, to check out things on the map *before* we arrive in the city. This can be e.g. for touristy reasons, as a preparation for a business trip or if we want to move to that city to study there or because we got a new job.

So, what does the map tell us about the urban reality that we will explore soon? Do we find everything on it that's interesting for us - like the most famous sights, the shopping districts, the old town, the public transit network? Of course, this depends on the graphic style, the level of detail resp. the level of generalisation, the scale of the map and on other parameters.

One possible hypothesis of this research project is that the cognitive image coming up to our mind when we look at a 2D map of a city is determinated by these "map style" parameters. The research project will focus on paper maps and include a "map experiment". Its goal is to find out more about the coherence between map style and "cognitive spatial image".

In his Blog, Enrico Bertini formulated the following rules for a proper vis brainstorming, let's take those to heart!

1-Don’t speak. Draw.
It’s easy to fall into the trap of trying to explain the visual picture you have in mind verbally. As I said, it doesn’t work. Normally it goes like that: a bright ideas comes into your mind and words start coming off of my mouth. The person in front of me keeps saying yes, yes … and there’s no connection, no way to check we are talking about the same thing. Normally, the two pictures diverge a lot. Drawing doesn’t fix the problem completely but it puts some major constrains and makes the all thing run lot smoother.

2-Use low-fi tools.
The whiteboard is my favorite tool (checking that good pens are always around), but paper works perfectly fine if it’s a one-to-one game. But just make sure both of you have pen and space to draw! Sketching is a “power game” and the person who holds the pen has the power (thanks Andrews for this idea). Keep this in mind. Digital stuff just doesn’t work here, unless you want to show some mock-ups designed beforehand.  And yet, creativity might be heavily impaired because drawing several alternatives or variations on top of it might be really difficult.

3-Let everybody speak and don’t interrupt.
It’s too easy to start talking and forgetting that the best ideas often come from exchanging ideas. If you find your mates keeping their mouth shut there might be two problems: either you are not giving enough space to the other person or this person has nothing to say. In the first case, stop speaking. In the second, stop the meeting and find a better person to talk to.

4-Listen carefully.
I said carefully. This is very much related to the previous one. Letting others speak is important but it’s not enough. Since this is an intense exchange of ideas aiming at boosting creativity, you have to make sure your mind is not only absorbed by what YOU think. Your brain is excited and it takes some effort to focus on the other and not on yourself.

5-Think constructively.
Well … once you succeed in keeping your mouth shut (but not forever!) and listen to the others, you also have to find the right way to criticize what you listen to. Don’t think “this is bad” too early. Actually don’t think “this is bad” at all. The best approach is to ask yourself: “how can we make this thing look better together? How can this improve my idea? And how can I add something to this idea to make it better.”

6-Be a flea and an elephant.
Generating lots of alternatives is very important, especially at the beginning, but from time to time it’s important to stop and  dig into a specific idea. Both are important and there’s no specific rule and when to do what. Use your intuition and just remember to be a flea and sometimes also an elephant (thanks to Alan for the metaphor).

7-Take pictures. Save it.
You don’t want to lose those ideas you have been generating for 1 hour or so. It’s easy to walk away and think you will remember what you have done, but it’s not true. Normally, after few days many details fade away, if not the whole thing, and details often matter a lot. To be sure, take a picture. It takes few seconds and it doesn’t cost a thing.

8-Take your time, it’s a process.
Finally, don’t pretend to solve everything in one shot. Designing a visualization, as any other kind of design activity, it’s a process. It takes time and it’s highly dependent on the state of your creativity. Stop when you see the spark of your creativity is fading, take your time and come back when the spark is back.

The next meeting of our group will take place on June 1st at 18h in D105 (City Nord).

This time Anna and Christoph will present their contribution to this year's IEEE VAST Challenge.The VAST Challenge is a participation category of the IEEE VAST 2010 Symposium as part of VisWeek 2010 held in Salt Lake City, Utah.

Christoph and Anna are with g2lab and will document the process of their visualization during our meeting. Finally, they will put their design up for discussion and hope to gather important feedback for their work.

Feel invited to this evening discussion - it is open to everyone interested in visualization topics!

The next meeting of our group will be

on April 6th at 18:00 in D105 (City Nord).

Alex will be the first to report on his topic which is
'Visualization of sound propagation in 3D'
A short summary of his approach can be found below.

We are looking forward to an evening with plenty of good ideas and
lively discussions!

---

Visualisierung von Schallausbreitung in 3D

Ohne näher auf die Hintergründe einzugehen, kann für ein 3D Gelände der Schalldruckpegel [in dB] für jeden Punkt in [x,y,z] angegeben werden. Allein die Darstellung dieser Schalldruckverteilung in 3D ist schwierig. Eine Lärmkarte ist dabei ein Darstellung der Schalldruckverteilung in zwei Dimensionen [x,y] (meist für einen festen z - Wert).

Man kann sich jedoch noch weitere "Dimensionen" denken: Die Zeit und die Frequenz!

Stellt man sich eine Quelle in der Schallausbreitungssimulation vor, wird diese durch einen "Pistolenschuss" nachgebildet. Der Schall breitet sich also kugelförmig um die Quelle aus. In 2D kann man sich das an einem Stein, den man ins Wasser wirft vorstellen, und die erste Welle sich langsam kreisförmig ausbreitet. So existiert der Schall direkt von der Quelle aus immer auf einer zeitlich größer werdenden Kugelschale. Der Schalldruckpegel nimmt dabei mit dem Radius, also der Zeit, ab. Dieses Verhalten wird allerdings erst nach Simulation mit Hindernissen interessant, an denen der Schall gespiegelt, gestreut oder gebeugt wird. Schließlich sind Reflexion, Streuung und Beugung frequenzabhängig, das heißt bei unterschiedlichen Frequenzen (="Tonhöhen") bilden sich verschiedene Schalldruckverteilungen.

Ergebnis einer Simulation ist also eine 5 -- dimensionale Matrix, mit dem Schalldruckpegel als Funkten von: X, Y, Z, Zeit, Frequenz

Gesucht werden soll hier nach einer möglichst guten Darstellung aller Parameter gleichzeitig, um die Schallausbreitung bildlich für jedermann verständlich und anschaulich zu machen.

Für eine Lärmkarte würde die Dimension Zeit wegfallen und in grober Näherung könnte man verschiedene Frequenzen zusammenfassen.

Ziel soll eine fruchtbare Diskussion zu Darstellung des beschriebenen Datensatzes sein, der später auch gerne interdiszipliär realisiert werden könnte.