Monitoring and Visualizing Last.Fm Beobachtungen Verschiedener Aspekte Des Musikkonsums Im Sozialen Netzwerk Last.Fm

MONITORING AND VISUALIZING LAST.FM BEOBACHTUNGEN VERSCHIEDENER ASPEKTE DES MUSIKKONSUMS IM SOZIALEN NETZWERK LAST.FM

DOCUMENTATION MONITORING AND VISUALIZING LAST.FM Final Year Project of Christopher Adjei and Nils Holland-Cunz Supervised by Prof. Eva Vitting and Prof. Philipp Pape University of Applied Sciences Mainz 2008 Motivation 06-07 Formative Elements Table of 01 The Topic and our Intention 05 #01 Composition of the Poster-Label 72 #02 Typeface 73 Research Contents 02 #01 Research and Analysis of other Projects 08-09 Resumé #02 Searching for Suitable Data Sources 10-11 06 What have we learned? 74 #03 What means Last.fm? 12-13 #04 What is an API? 14-15 Appendix Bibliography 74 Impressum 75 Finding the Concept 03 #01 Questions 16-17 #02 Scopes of Topics 18-19 and Form of Presentation #03 Getting and Restructuring the Data 20-21

Design and Visualizing 04 Introduction: Distribution of Users #01 Research 22-23 #02 First Visualisations 24-25 #03 Final Graphic 26-27 Topic Scope 1: Comparing Fan-Groups #01 Question/Idea 28-29 #02 First Visualisations 30-31 #03 Approach Circular Diagram 32-35 #04 Approach Circular Diagram without 36-37 Connections #05 Approach Cochlea-Shape 38-39 #06 Approach with Individual Forms of Genres 40-41 #07 Approach with Mirrored Form 42-43 #08 Approach with Mirrored Form and Overlay 44-45 #09 Final Approach »Amplifier« 46-47 Topic Scope 2: Fluctuation of Fans #01 Idea Development 48-49 #02 First Realisation 50-51 #03 Finalizing the Graphic 52-53 #04 Final Poster and Colour-Coding 54-55

Topic Scope 3: Album-Release #01 First Scribbles 56-57 #02 First Realisation 58-59 #03 Finalizing the Graphic 60-61 #04 Poster and Colour-Coding 62-63

Topic Scope 4: Cumulation of Genres #01 Question/Idea 64-65 #02 2D und 3D 66-67 #03 Realizing Application 68-69 #04 Realizing Poster 70-71 Motivation

01 Motivation The Topic and our Intention 06 | 07

The Topic and our Intention

Turntable Technics 1200 Score of „Die Zauberﬂöte“ by W. A. Mozart

At the beginning of our working process there were various ideas programming language »Processing«, in other words: a combina- coming to our minds. We connected these ideas with each other tion of programming, concept and design. First experiences with and out came, as a result, this diploma project. The ideas refered »Processing« were made during winter-semester 07/08. The to here range in a wide scale: From an interactive appplication reason why we were interested in a topic dealing with music from including the possbillity for the user to call up online current the beginning goes, amongst others, back to both our musical concerts and events within his environment by his location data, background as a dj for electronic music, on one side, and as a up to an analysis of music preferences of various social groups. vocalist in classical music, on the other. It was our aim to widen our knowledge and abilities in the ﬁeld of data visualizing, and to deepen our competence in handling the

Monitoring and VisuAlizing last.fm 02 Research Research and Analysis of other Projects 08 | 09

Research #01 Research and Analysis of other Projects

offf.ws bestario.org manyeyes.alphaworks.ibm.com processing.org

visualcomplexity.com benfry.com senseable.mit.edu/nyte/ felixheinen.de

At first, we researched in theme-specific and other internet portals design. These informations now served us as a basis for orienta- dealing with the object of data visualizing. We wanted to find out tion in order to produce primary visualisations for our project. At what makes these works interesting and what is »State of the that point, however, it was not yet clear to us how our view on Art«. Thereby, we came across projects by Ben Fry, Aaron Koblin these works was to change as the project went on. and the works of other design students as e.g. Michael Groß, Felix Heinen and Stefan Bräutigam. We then noticed that certain visualizing methods and trends, as e.g. the semi circle or the circular diagram were predominant, as also in other fields of graphic 02 Research Research and Analysis of other Projects 10 | 11

Recherche #02 Searching for Suitable Data Sources

myspace.com/svenvaeth developer.myspace.com developer.yahoo.com code.google.com/apis

In search of suitable data for programmed visualisations we However, it turned out that appx. 50% of the artists' profiles on to observe by means of the data saved there. 3. In which towns do certain genres of music cumulate, and in next analyzed the social network »MyspaceMusic«. There bands MySpace were »joking-profiles« and therefore were not a What we wanted to find out in a next step were these questions: which countries is a newly released album heard first? as well as soloist artists, wishing to present themselves online reliable data source. In our further research we came across with their music, may start a profile site and upload their own the data base of Last.fm, which had a far better proportion of 1. Which artist is a »one-hit wonder«, and which has got a songs. Our first idea was, to sort out all the artists' profiles suitable to unsuitable data. To us, this data source opened up a constant fan-community? and to compare their touring-data. We wanted to find out if far wider range of possibilities than the MySpaceMusic data. there were interdependences among the artists' tours, and if We decided to use the Last.fm data base for our project and to 2. Are the supporters of a rockband more receptive forwards such correlations lay within a national or international scope. work out of it a series of interesing aspects which we intended various fashions of music than the fans of an hip-hop artist? 02 Research What means Last.fm? 12 | 13

Recherche #03 What means Last.fm?

lastfm.de/User/Linzi_Weir (Proﬁl Site of Linzi Weir) lastfm.de/music/Tokio+Hotel/+events

lastfm.de/music

lastfm.de/music/Tokio+Hotel/+listeners lastfm.de/music/Tokio+Hotel

As everybody knows, nowadays music is not only offered in the special interest in music as their target group. Last.fm is a and biographies of artists as well as lay out his own artist's or record shop or at concerts, but also in a large variety in the worldwide social online-network along with a personalized radio label's proﬁle. A further particularity is the so called »scrobbling« internet, whereby the offer of various webradios is very volumi- station. The network-user receives recommendations as to music of music. Thereby, the playing of each track with the Last.fm Player nous. Yet, in most cases we do not know the audience. A lot of and concerts. These recommendations are based on data generated or with an external plug-in is saved in a data base. This data base social networks such as »studiVZ« or »facebook« enter directly by the user's own listening-behaviour. Within this context he has is allways freely accessible by an API and served us as a source of into the user's whish to present himself with his hobbies and the possibilty of listening to various radio stations, of placing data for our Visualisations. interests and to communicate on that. Other networks such as »tags« for bands listened to as well as of communicating »Last.fm« go one step further and look for people with a very recommendations for music. He can also inscribe concert-data 02 Research What is an API? 14 | 15

Recherche #04 What is an API?

lastfm.de/api (Explanation of the API-Method: artist.get.similar) lastfm.de/api (Example of the response in the form of an xml-ﬁle)

An application programming interface (API) is a set of routines, of each »library« is termed »API«. Which way did we obtain the data structures, object classes and/or protocols provided by data? libraries and/or operating system services in order to support the building of applications. Besides the access to data bases and to 1. We sent a request to the data base of Last.fm by the program- hardware such as hard drives or graphic boards, an API may also ming environment Processing. enable a developer to produce the components of graphical user interfaces and to simplify this procedure. Nowadays, a lot of 2. We received data in form of xml-ﬁles, which we read out by internet providers also provide APIs. In a wider sense, the port means of a library and then saved for further processing in text-ﬁles. 03 Finding the Concept Questions 16 | 17

Concept #01 Questions

1 Countries 5 Preferences of music/gender 1.1 How large is the quantity of the users in one country? 5.1 Which tracks by one artist are more preferred by women Länder 1.2 What is the proportion of active to inactive users in a country? resp. men? 5.2 Which genres in country X are more favoured by women Wie ist das Verhältnis von aktiven zu inaktiven Usern in 1.1 2 Fluctuation resp. men? einem ausgewählten Land? 2.1 Who listens to which music in which country? Vorgehensweise: Passt zusammen mit: 2.2 How does the distribution of the 1,000 top listeners of an 6 Album-Release - Auslesen der Namen jeder 200. User-Verzeichnisseite eines Landes artist vary in relationship to place and time? 6.1 How does an album or a single-release influence artist’s 2.3 Do the touring-data influence this distribution? 5DXVÀQGHQZHUDNWLYXQGZHULQDNWLYLVW popularity? 2.4 Do the touring-data influence the above mentioned distribution 6.2 Does an album/single-release influence the popularity of his of »similar artists«? »similar artists« (scrobble-data) 6.3 How do the top-listeners of a track spread after an album or 3 Preferences of music single-release? 3.1 Are the listeners of the tag-genre X more receptive towards other tag-genres than the listeners of the tag-genre Y? Wie viele Datensätze werden aufgezeichnet? 3.2 Are the listeners of country X more receptive towards various tag-genres than the listenres of country Y? 3.3 Is there any correlation between the result of 3.2 and the GDP, Wie interessant sind die Daten? Parsing: Datenquellen: economic growth and religion? W. Last FM A.

4 Concerts 4.1 In which towns do the concerts in country X take place? (areas of urban cumulation?) Evaluation of a question on a schedule created by us 4.2 Which tag-genres do the concerts taking place in country X belong to? 4.3 Where do all concerts belonging to one tag-genre take place in country X? (areas of urban cumulation?) 4.4 Does the distribution of the concerts as to tag-genres depend on GDP, economic growth and religion?

As a conceptional basis we worked out a list of questions the answers to which were to be relevant to speciﬁc target groups. We assorted these questions from various points of view: Whom is the question interesting to? How often does a data set have be captured again? How many steps and stations must be passed and taken until the data can be used for a visualisation? 03 Finding the Concept Scopes of Topics and Form of Presentation 18 | 19

#02 Scopes of Topics and Form of Presentation

Introduction Topic Scope 1, Comparing Fan-Groups Topic Scope 2, Fluctuation of Fans

Topic Scope 3, Album-Release Topic Scope 4, Cumulation of Genres

As mentioned at the beginning our initial approach resulted in Introduction: Distribution of Users Topic Scope 02: Fluctuation of Fans Topic Scope 04: Cumulation of Genres various questions, which we first of all gathered. Next we filtered Representation of the distribution of users of Last.fm and explai- Demonstration of weekly listeners leaving and coming by of Presentation and Categorisation of the concert distribution within some of them out and grouped them in five scopes of topics. ning introduction for the following topic scopes. various music groups. Realisation: Poster, 1100x3235 mm europe. Realisation: Interactive Application & Poster, 1100 x 2215 Those are related to issues that either run over a longer period, Realisation: Poster, 1100x750 mm mm or are place-related, or show properties. Consequently, we had to Topic Scope 03: Album-Release capture all the data necessary for that and link them up to each Topic Scope 01: Comparing Fan-Groups Visualizing the place-related extension of the listeners of certain other. On the following pages, an overview of these data links is Comparing random samples of the »top-listeners« of Radiohead, albums after the release on Last.fm. presented for each scope. The five scopes are as follows: Moby and Nelly. Realisation: Poster, 1100x2690 mm Realisation: Poster, 1100x1890 mm 03 Finding the Concept Getting and Restructuring the Data 20 | 21

#03 Getting and Restructuring the Data

Distribution of Users 04 Cumulation of Genres

List of Countries with Proportion Venue activ/inactiv

List of Countries User Index-Sites List of 100 coincidental Url-Profil-Site Town with Proportion of all Countries selected Users for each User activ/inactiv

Coordinates 01 Comparing Fan-Groups

List of Tags for each User List of Countries API All Event-Data within Europe of a Country Artists/ Bands

All Tags of Tag 1-3 an Artist (Genres) for each Artist (Genres)

Date Index-Sites 3 Random Samples API 50 Top-Artists Playcount of 1,000 Top Listeners with 100 Users each of each User for each Artist of an Artist/of a Band

Event-Title

02 Fluctuation of Fans

Index-Sites List with 500 Users API Country of a User Ranking of the of 1,000 Top Listeners Countries

03 Album-Release Getting and restructuring the data we used is shown by the a country index-site site on Last.fm for information. These are process diagrams. The procedure is always similar except the then saved in the form of text-files, as well. topic scope »Distribution of Users«. By an API method a request is made to the the data base of Last.fm. The data base sends back an xml-file which we decode. It is then being restructured and saved in the form of a text-file. An exception, as mentioned, Index-Sites List with 1,000 Users API Country of a User User-number is the scope »Distribution of Users« in which we directly search of 1,000 Top Listeners for each Country Design Introduction: 22 | 23 04 and Visualizing Distribution of Users

Introduction: Distribution of Users #01 Research

Examples in the book »Sémiologie graphique: Les diagrammes, Les réseaux, les cartes«, by Jacques Bertin

Example in the book »Atlas of Shrinking Cities« Typical Visualisations by Otto Neurath

As the headline already indicates we intent to mediate an coincidence. By this random sample we were able to re-con- describes the long-term stability of the mean of a random impression of the worldwide distribution of the Last.fm users. clude the proportion of active to inactive users as well as of variable. Given a random variable with a ﬁnite expected value, if These are found out by means of the country index-sites on Last.fm, mail to female users. At the same time we started the research its values are repeatedly sampled, as the number of these to which one can get by the user's searching mask. During a more of various possibilities of visualisation. Thereby we came across observations increases, their mean will tend to approach and precise investigation we found out that a large number of the among others the cartographer Jaques Bertin, the information stay close to the expected value. registered users had not visited their proﬁles for more than half a scientist Edvard Tufte and the Austrian philosopher and econo- year. We called these »inactive users«. On the basis of the »law mist Otto Neurath. of the large numbers«* we chose 100 users for each country by *The law of large numbers (LLN) is a theorem in probability that Design Introduction: 24 | 25 03 and Visualizing Distribution of Users

USERVERTEILUNG (ABSOLUT)

#02 First Visualisations

Map 1 Map 2

Die Kopfhörer stehen im auf der gesamten Karte im gleichen Verhältnis.

Polen 57.000 User

Polen 57.000

Deutschland 57.000

Deutschland Frankreich 57.000 57.000

Italien 57.000

Polen 57.000 User

Vereinigte Staaten von Amerika 533.780

Deutschland 57.000

Ägypten 57.000

Deutschland 57.000

Brasilien 57.000

Australien 57.000

Polen 57.000 User

Neu-Seeland 57.000

Map 3

<10.000 10.000-40.000 40.000-80.000 100.000-180.000 >500.000

AMERIKA NORTH 01 - KANADA 66400 02 - VEREINIGTE STAATEN 533780

NORWEGEN FINLAND SCHWEDEN EUROPA 1 - BELARUS 3780 11 - AUSTRIA 16340 21 - FINLAND 57060 29 - GERMANY 171500

2 - ESTONIA 6300 12 - BELGIUM 21180 22 - FRANCE 44380 30 - POLAND 113940 BRASILIEN

3 - GREECE 7260 13 - BULGARIA 11100 23 - ITALY 45180 31 - UNITED KINGDOM 172200 BRASILIEN RUSSLAND

4 - HUNGARY 8460 14 - CZECH REPUBLIC 18060 24 - NETHERLANDS 53660 DÄNEMARK BELARUS 5 - ICELAND 2560 15 - DENMARK 11820 25 - RUSSIA 77940 DEUTSCHLAND POLEN 6 - ISRAEL 8440 16 - IRELAND 10700 26 - SPAIN 54420 IRELAND UK NIEDERLANDE TSCHECHIEN 7 - CROATIA 8140 17 - NORWAY 24180 27 - SWEDEN 60840 RUMÄNIEN CANADA ÖSTERREICH 8 - LATVIA 6820 18 - PORTUGAL 19100 28 - TURKEY 33480 SCHWEIZ BRASILIEN FRANKREICH 9 - LITHUANIA 8420 19 - ROMANIA 11720

10 - SERBIA 5300 20 - SWITZERLAND 14440

SPANIEN ITALIEN TÜRKEI PORTUGAL GRIECHENLAND

ASIEN 34 - KASACHSTAN 1240 38 - CHINA 19680 USA

35 - PHILIPPINEN 7980 39 - JAPAN 38540 ISRAEL CHINA JAPAN 36 - INDONESIEN 3200

37 - MALAYSIA 4380

INDIEN MEXIKO

40 - SOUTH-AFRIKA 3320 AFRIKA PHILLIPINEN VENEZUELA 41 - GYPTEN 1260 COLUMBIEN

INDONESIEN COLUMBIEN

AMERIKA SOUTH Map 421: - COSTA RICA 1520 48 - Mexiko 25460 50 - Brasilien Map 119100 3:

43 - KOLUMBIEN 8060 49 - Chile 16640 BRASILIEN First we tried to visualize the number of users of a country on a After that we formed units. One headphone stands for 10,000 CHILE 44 - Equador 1000 AKTIV* EINHEIT: 10.000. USER 45 - Peru 4780 world map by colour shades and diverse patterns. users. Because of the proportions in europe we piled the head- BRASILIEN 46 - Uruguay 1060 INAKTIV* EINHEIT: 10.000. USER 47 - Venezuela 4800 phones diagonally on top of each other. CHILE COLUMBIEN INAKTIV* UND INAKTIV WENIGER ALS 10.000. USER

AUSTRALIA Map 512: - Neuseeland 7900 52 - Australien 42900 NEU SEELAND LÄNDER MIT WENIGER ALS In a further step we decided to picture the various amounts of Map 4 (next page): 1000 ANMELDUNGEN

*»AKTIV« BEZIEHT SICH AUF DIE NUTZUNG DES NETZWERKS, D.H. WENN EIN USER IN DIESEM JAHR DAS NETZWERK NICHT GENUTZT HAT, GILT ER users by a headphone icon. Final map, showing the headphones in a row. ALS »INAKTIV«.

MONITORING AND VISUALIZING LAST.FM OBSERVATIONS OF VARIOUS ASPECTS OF CONSUMING MUSIC IN THE SOCIAL NETWORK LAST.FM

26 | 27 Worldwide Distribution of the Network Members of Last.fm Status in August 2008

NORWAY

SWEDEN FINLAND

ESTONIA

LATIVA DENMARK LITHUANIA

GERMANY IRELAND UNITED KINGDOM NETHERLANDS POLAND RUSSIAN FEDERATION

CZECH REPUBLIC UKRAINE

SWITZERLAND AUSTRIA HUNGARY ROMANIA CROATIA CANADA SPAIN FRANCE BULGARIA

PORTUGAL GREECE TURKEY

ITALY

ISRAEL

UNITED STATES CHINA JAPAN

INDIA

MEXICO COSTA RICA THAILAND PHILLIPPINES VENEZUELA

COLOMBIA SINGAPUR

INDONESIA

PERU

BRAZIL

CHILE ARGENTINA SOUTH AFRICA AUSTRALIA

ACTIVE* UNIT: 10,000 USERS INACTIVE* UNIT: 10,000 USERS NEW ZEALAND COUNTRIES WITH LESS THAN 1,000 REGISTRATIONS

*»ACTIVE« REFERS TO UTILISATION OF NETWORK, I.E. IF A USER HAS NOT USED THE NETWORK FOR THE RUNNING YEAR (2008), HE IS CLASSIFIED AS »INACTIVE«.

THEME LAST.FM

Which artist is a »one-hit wonder«, and which has got a constant fan-community? Last.fm is a worldwide social online-network along with a personalized webradio station. Are the supporters of the band Radiohead more receptive forwards various fashions of music The network-user receives recommendations as to music and concerts. These recommendations than the fans of the hip-hop artist Nelly? In which towns do certain genres of music cumulate, are based on data generated by the user's pattern of listening. Within this setting he has the and in which countries is a newly brought-out album heard first? These and other questions can chance of listening to various radio stations, of placing tags for bands listened to as well as of be investigated with the help of Last.fm, which is a social music-network with a personalized communicating recommendations for music. He can also inscribe concert-data and biographies radio station. By listening to it, the member continuously generates new data, which are saved of artists as well as lay out his own artist's or label's profile. in the public database being pertinent to it. We observed part of these data for a period of four A further particularity is the so called »scrobbling« of music. The playing of each track on the months and evaluated them in order to find answers to the above questions, for the audience of Last.fm player or with an external Last.fm plug-in is saved in a database. This database is Last.fm. Our project consists of four parts: 01 »Comparing Fan-Groups«, 02 »Fluctuation of always freely accessible through an API and has served as a source of data for our visualiza- Fans«, 03 »Album-Release« and 04 »Cumulation of Genres«. tions. Within these scopes, we can present most interesting results gained from our observations. All visualizations were realized by using the programming language and integrated development environment »Processing«.

MONITORING AND VISUALIZING LAST.FM FINAL YEAR PROJECT OF CHRISTOPHER ADJEI AND NILS HOLLAND-CUNZ SUPERVISED BY PROF. EVA VITTING AND PROF. PHILIPP PAPE UNIVERSITY OF APPLIED SCIENCES MAINZ 2008 Design Topic Scope 1: 28 | 29 04 and Visualizing Comparing Fan-Groups

Comparing Fan-Groups #01 Question/Idea

First there were two questions: How can we describe the band- related music preference of a user group? What have the various user groups in common and what are the differences?

The Approach Last.fm offers weekly rankings of 1,000 users out of the top- listeners of a certain band. From these rankings we raise random samples of 100 listeners each, who represent the community of listeners to the musicians we selected. Every user is visually displayed by one graphic each. The result of this is a series of 100 different graphics generated by the same algorithm, and as a group can be compared with user groups of other artists. The graphics allow conclusions about the music preference of an individual as well as of the whole group.

Realisation By the API method »user.getTopArtists« we first gather in a text-file those 50 artists a user has mostly listened to with the Last.fm Player during the last three months. Also, for each artist the playcount* is being saved in a seperate text-file. We use the artists’ names for saving the three most frequent tags** of an artist by the method »artist.getTopTags«. These tags are then being sorted according to groups and quantities, not depending on an artist, in order to find out which tag a user listens to mostly. The result of this sorting would be saved in a text-file, as well. The five biggest genre groups together with the playcount would be used as a parameter for visualizing. This description refers to the last version of a series with several visualizing-attempts being shown on the following pages.

* Playcount: The playcount says how often an artist has been played on the player of a user's proﬁle.

** Tags: The network users can share one or several tags for each artist, describing the respective artist. For this they normally use the genre group the artist’s music belongs to.

These scribbles on notepads of the »Bitburger« brewery are silent witnesses for long nights that we spent in a pub during our diploma project. They testify part of our search for a visual solution to this topic scope. Design Topic Scope 1: 30 | 31 04 and Visualizing Comparing Fan-Groups

#02 First Visualisations

Formfindung

Graphic 1 Graphic 2

Searching for a suitable use of form for the illustration of a user contents. During the first steps we instinctively tried to express following attributes and parameters of a user-profile were being Graphic 1: This graphic was the ﬁrst of more than 40 different group and their music profile there were several approaches. It the difference between individual music genres by colour or by at our disposal: The frequency of its played music genres, the visualizing approaches. The upper part visualizes a fan group of was a challange to us to compare groups of users to each other classical forms as the circular diagram. Finally we abandoned playcount of individual artists and the combination of genres Radiohead the lower one a fan group of Moby. You can see that the and to filter out how a music profile of an individual user is built this and concentrated on more precise shapes. These (shapes) resulting from all artists played. fans of Radiohead predominantly have a one-way music preference. and how it differs in comparison to others. Starting from the first were to characterize a certain music genre of a user and to make programmed visualisations again and again we were confronted each user individual and comparible by their very different Graphic 2: : This graphic shows our ﬁrst trial to visualize the with the difficulty of finding a solution that is graphically convin- arrangments and shifting of parameters. By this we tried to relationship of genres to each other. cing as well as clear and understandable with regard to the »stage« a user group again and again in different ways. The Design Topic Scope 1: 32 | 33 04 and Visualizing Comparing Fan-Groups

#03 Approach Circular Diagram

hip-hop

freak folk freak

techno trip-hop

psychobilly

world

garage rock garage

celtic ambient

alternative rock alternative

bossa nova bossa

british

folk gothic

progressive metal progressive maminkovsky folk maminkovsky

progressive rock progressive

british

singer-songwriter

industrial screamo

straight edge straight

metal

new zealand new

irish

shoegaze

acoustic industrial

ebm

post-rock

electro

rhythm and blues and rhythm

techno

ska canadian indie

female vocalists emo indie rock

big beat big chillout

krautrock alternative

ninja tune ninja

trip-hop oi experimental

symphonic metal symphonic soundtrack post-punk

singer-songwriter

grunge hardcore

instrumental

rock blues rock

funk

alternative britpop punk rock

electronica downtempo electronic

ambient

noise rock noise -

new age new

electronica lo-fi post-hardcore

czech 80s new wave

indie

math rock math rap

drum and bass and drum

dance punk hardcore hip hop french

pop grindcore hip-hop

noisecore psychedelic

alternative rock alternative

idm goth instrumental metal post-rock

classic rock classic soul jazz

For this approach we developed an algorithm by which we were able to display any number of genres of a user with their relationship to other genres. The genres that are found most frequently have the most connections to other genres. This way it can be determined if the user listens to only one style of music or to several different ones. Design Topic Scope 1: 34 | 35 04 and Visualizing Comparing Fan-Groups

taxmantom obc_spike Fedra_JL ahau jcgutier korryfey s t

l e e v v i i

vocalis

t t t na na r r r k

e e

t t

e emale

industria al al v f pun chillou

i e

roc trip-hop al singer-songwrite nerdcor tranc

electronic - indie rock indie electronic

f roc al al po eb

emale

t t

p m

e e

r r na na

t t vocalis

i i

v v e e

- indi singer-songwrite ele roc roc t s

ct k k e roni

gsusy falconoc4um kicherErbs shmel735 Beat-m sam1a

n l l

alia e

p str

e e e

v v po v danc au minima trip-hop industria

i i i

t t t

na na na

r r r

k ct

e e e

t t t

roc al al ele meta al

indie rock rock electronic electronic rock

roc al hous f ele singer-songwrite

ct k

i s na e

roni

i v c e

indie roc po indi - - meta p r e

ninja_gecko r00tyfruit kennyboy019 djcopio ANTIGONEO5 PhilipHand

k t t p en

i e

p l

c v

ra amb dan v britpo chillou indie roc

i l i

t t

roni

na t

e e

t t

minima al gothic meta gothic elec hip-ho al

rock electronic electronic rock electronic indie

techn trip-hop al electronic roc indie roc indie

v k e a

meta - roc indi meta al

t e k e r na l l

t i v e

Dalfar lolapay BiZKeeD BlackthoRnX bjam70 jsystem

This approach is based on the same algorithm as in the preceding graphic, but here only the five most frequent genres are shown. The coloured genre is named »Independent«. With this approach

the shape is »closed«.

singer-songwrite

ele

roni c Design Topic Scope 1: 36 | 37 04 and Visualizing Comparing Fan-Groups

#04 Approach Circular Diagram without Connections

starcarly14 Christel82 strikerg danjahandz peibolm Errshey

apowers69 12yosami smplystephanie lullabum staddo G-unitGs

motoro11er lhh2009 anni_bus PiroGom paulie69 Blade555

With this approach we forgot about the connecting lines between the individual genres and decided to display them with larger areas. cozitWe arrangeda them in a way that would give a clearerOzij s mayonnes danjahandz vikingpopcorn neverbefore shape to the circular diagram.

streekz xlondoncallingx aleVillac discodadznori dance7 Design Topic Scope 1: 38 | 39 04 and Visualizing Comparing Fan-Groups

#05 Approach Cochlea-Shape

By this approach we tried to leave the more familiar graphics by aranging triangels mirror-inverted. Each triangle is representing one genre. Design Topic Scope 1: 40 | 41 04 and Visualizing Comparing Fan-Groups

#06 Approach with Individual Forms of Genres

This approach arranges individual genres in a horizontal line. Here, each genre is displayed in an individual form. Design Topic Scope 1: 42 | 43 04 and Visualizing Comparing Fan-Groups

#07 Approach with Mirrored Form

comedownnine mhkeez pubekas stealthislastfm skraggy Jiggins23 yurish bluesroad skraggy vuitton_playa art-i-choke trbcrp soapmachine lenlenaq

syrehue motyR godgough cburrhead stealthislastfm mgomezivaldi atomicus_pl George_85 boogiehoney amne- nedaleva El_Rabino xsteadfastx Emzadi613

bastrion zerohour269 TommyNovember7 Newborn_Bantam er_ay jord2006 Rbraz gttcasey gloomytheo ImCoolOnLastFM maarye Aarontown renasaurusowns pengu-

kariikarii hypnoses yekey JMoneyFresh mikedann cheeks11 scaretactix funwithlastfm JamieandKyrill lithium_head felipers RoryMack Carrionly goodbarryan

star_suckers SilentChicken signal74 tooweighmirror sototallykody ymmoTTommy brantlymurphy vboechat chirimar klavir ellis_shop vwjake sbrookss JussiA

This approach aranges the individual genres vertically. Each genre is here displayed by an individual form. Afterwards it is being inverted.

Mampfi varyavarya Xenostar Deutlich coltbang john_jack numar28 flexx55 Discolabirinto Yhaa Homesick99 ocmerius dodgenick RadioactiveDark

fallore mjwald Mork0vka brantlymurphy untitled37 scaup agfay Grey_Gh lxyt dannytoffee rabidasp kisssis t_seijiro_o irjman Design Topic Scope 1: 44 | 45 04 and Visualizing Comparing Fan-Groups

#08 Approach with Mirrored Form and Overlay

This approach is very similar to the preceding one, but the arrangment of genres lies in a horizontal level again. 01 RADIOHEAD STICHPROBE 2/3 MOBY STICHPROBE 1/3 NELLY STICHPROBE 1/3 FANGRUPPENVERGLEICH bluesroad skraggy vuitton_playa art-i-choke trbcrp soapmachine lenlenaq George_85 boogiehoney amne- nedaleva El_Rabino xsteadfastx DiREKtor taqx9 jacata 3e7o dmind1 Fetalij iwakawa_73 lomarmo chriskeay snilard luka_teixeira princevihtor pmati B-Duarte PhenomZ ainsleejane pepemarbo phuture84 newbilized AgeBlade Andromeda_409 fresh_freak vitaminbomb jrod Lynn_Luxurious bmsilva

BEOBACHTETE KÜNSTLERUng RADIOHEAD/Moby/Nelly

ROCK INDIE IR ALT BRIT MET ROCK SSW INDIE ALT INDIE ALT IR ROCK IP INDIE ROCK ALT IR EC INDIE IR ROCK ALT CAN ROCK INDIE IR ALT POP ROCK ALT AR INDIE ELEC ROCK ALT INDIE ELEC IR ROCK ALT FV AR ELEC INST PR ROCK SL EXP INDIE SSW ROCK FOLK ALT ROCK PR PUNK INDIE ALT ROCK ELEC INDIE ALT CR INDIE ROCK ALT ELEC IR POP FV ELEC DAN ROCK ELEC IDM AMB ROCK ALT HH RAP HH ELEC ROCK JAP ELEC - DAN INST FV POP 80s ROCK NW ROCK INDIE POP DAN BP ST - ELEC ROCK FOLK POP FV ELEC - ST POP FV ROCK RnB DAN MET ROCK HM PM TM POP - RnB HH RAP ROCK MET ALT GM SM ROCK FV POP HH ALT HH RAP HH JAZZ GR HH RAP ROCK HH - ROCK POP HH RAP EMO MET PM HH RAP ROCK - RAP HH HH RnB ROCK ALT AR CR GRUN MET TM MDM HM DM POP FV HH RAP ROCK ROCK ALT INDIE ELEC IR LITHUANIA DATENQUELLE/API-METHODEN ZEITPUNKTE DER DATENAUFZEICHNUNG LITHUANIA MEXICO lastfm.de/music/Artist*/+listeners/ 03.08.2008/10.08.2008/15.09.2008

User.getArtists/Artist.getTopTags MEXICO MEXICO BELARUS MEXICO MEXICO MEXICO MEXICO MEXICO MEXICO INDIE ROCK ALT IR ELEC ROCK CR SSW ALT - POP DAN ROCK - FV

FANGRUPPENVERGLEICH DER KÜNSTLER NELLY, RADIOHEAD UND MOBY Aus den wöchentlichen 1000 “Top-Hörern” der Musik von Radiohead, Moby und Nelly, ermittelten wir zufällig jeweils 3 Stichproben mit je 100 “Top-Hörern”. Von diesen ermittelten wir die jeweils fünf am meisten gehörten Musik-Genres und visualisierten sie in Form eines Amplifiers. Jedem Genre haben wir eine eigene Farbe zugewiesen. Bei der Untersuchung fiel auf, das sich das Hörverhalten von Fangruppe zu Fangruppe stark unterscheidet. So lässt sich beispielsweise aus der gezeigten Grafik ableiten, daß die Tophörer der Band Radio- imcoolonlastfm maarye Aarontown renasaurusowns pengu- funwithlastfm JamieandKyrill lithium_head felipers RoryMack Carrionly goodbarryan vboechat Jinguro ValentineJo dust0r k_rainbow JacksonSU Sky-in-my-handS c0sta ansiska sebasandoval edomatus nick_hunter_mvp JohnDSommers Biku-no-kiba bushkica Jouen pri_c_ll VVick3d abigbiglove jaaniitc re_douda linezn cky11 aninteger SusseKinD bchee Raya_13 head überwiegend Musik aus dem Bereich Rock hören. Die Tophörer des Künstlers Moby hingegen haben einen deutlich breiteren Musikgeschmack.

AMPLIFIER AUFBAU

Die Ausschlagshöhe ergibt sich aus der Anzahl der Künstler, die mit dem entsprechenden Genres ausgezeichnt sind. ROCK ALT INDIE AR IR ROCK ALT INDIE 80s AR INDIE IR ALT IP SSW ROCK ALT MET BRA LAT INDIE IR CAN FV ALT INDIE IR ROCK CAN FV ROCK ALT GER AR INDIE ROCK ALT INDIE MET AR ELEC DnB PST PSD DnB ROCK INDIE IR PUNK ALT ELEC ROCK GM PUNK FFM INDIE IR EXP FOLK SSW - ELEC ROCK FV EA FV POP INDIE - SSW ELEC TH ROCK CHILL EA ROCK ALT INDIE IR EMO ROCK ALT CR INDIE BRIT ELEC FV INDIE ALT DAN ROCK ELEC INDIE PUNK CZE ROCK ALT POP CZE FV ROCK - CR ALT SSW ROCK INST CR POP ELEC INDIE ROCK ALT ELEC IR ROCK CR ALT ELEC MET ELEC DAN TRA PM CHILL - POP FV RnB ROCK ROCK ELEC CR FV 60s POP FV LAT INDIE SPAN ROCK ALT CR INDIE SSW HH INDIE ROCK ALT EMO RnB HH SOUL TRA POP ROCK ALT MET AR HH HH POP RnB FV SOUL RnB TRA ELEC HH SOUL HH RAP HH ELEC CHILL - HH FV POP RAP - RnB HH SOUL POP ROCK - POP ALT RnB

Eine Einheit repräsentiert einen Künstler. Die Dicke ergibt sich daraus , wie oft ein Künstler auf dem entsprecheden User-Profil abgespielt worden ist. ROCK ALT INDIE IR AC vw jake Username

Die einzelnen Genre-Kathegorien sind Farblich kodiert.

Die 5 am meisten vorkommenden Genres des ME DR GE FW DT Userprofils werden der Reihe nach abgebildet. Bezeichnung eines Genres bzw. Sub-Genres

vwjake JussiA flexx55 Discolabirinto Yhaa Homesick99 ocmerius dodgenick RadioactiveDark Grey_Gh lxyt dannytoffee rabidasp iceworld Nyzek martman89 ggmimura voleibol69 sisa989 ShiningNSilent Tatiss manwh0re dejavuu ttian evgen-ja Nacce Ronskibonski Volscio bocian4u jole45 matiaz_ kristin91188 asthoreinarss termites zimekzimny s0mnambulist GusR9

KATHEGORISIERUNG DER GENRES

HH HIP HOP AR ALTERNATIVE ROCK AJ ACID JAZZ ROCK ALT INDIE CR AR ROCK CR FIN JAZZ FUNK BM MET - AMB NOR ROCK CR ELEC ALT - ROCK ELEC ALT INDIE FV ROCK ELEC - ALT INDIE INDIE IR ROCK ALT ELEC ROCK INDIE ALT IR ELEC ROCK ALT ELEC AR - ROCK INDIE - IR FV - INDIE IR ALT ROCK INDIE ELEC ROCK ALT BP INDIE IR IP ALT ROCK ELEC JAZZ - ELEC FUNK ELEC ROCK JAP AMB ST ELEC ROCK INDIE ALT IR ROCK CR ALT HH RAP ELEC FV ROCK INDIE POP INDIE ROCK IR ALT SPAN ROCK ALT TR AR CR ELEC ROCK INDIE ALT EC MET FV - HR ELEC ROCK MET ALT HH CR ELEC ABM TH EXP IDM ROCK FV POP ALT CR - ROCK ELEC RUSS INDIE RAP HH HH ROCK - HH RAP MIN ELEC TECH HH ROCK ALT RAP ELEC - TRAN ROCK ELEC HH ROCK HH RAP ALT FIN RnB - HH SOUL POP HH RAP HH - ROCK ROCK ALT POP PR PP RAP HH HH ROCK ED RAP RAP HR HARD ROCK R RUMBA DS DIRTY SOUTH NT NINJA TUNE ROCK ELEC HH ALT RAP ROCK ALT HH RAP ELEC CR CHICANO RAP PUNK PUNK NJ NU-JAZZ OS OLD SKOOL PR PUNK ROCK TT TRU THOUGHTS POP RnB HH FV OPM

IHH INSTRUMENTAL HIP-HOP PP POP PUNK MEXICO MEXICO MEXICO BELARUS MEXICO GR GANGSTA RAP PP POST-PUNK CT CHIPTUNE UHH UNDERGROUND HIP-HOP HCP HARDCORE PUNK EP ELECTROPOP RnB R&B G GABBER SOUL SOUL FP FINNISH PUNK B BEATS RR RUSSIAN RAP SP SKA PUNK BB BREAKBEAT

Design PH PUNK HARDCORE DnBTopic Scope DRUM 1: AND BASS comedownnine mhkeez pubekas stealthislastfm skraggy Jiggins23 yurish syrehue motyR godgough cburrhead stealthislastfm mgomezivaldi Wojtuuu Meea geenidee48 Tatiss dmind1 6ofTen saenochka victor_mov Mimouna benticarlos Wes22 ntbnnt blueclouds rafinio85 muiman62 Proceed_Ger timber125 seryck VladV7 ragasman trogdor007 crakerjac xYAlucardYx eurichar Agonyya MET METAL 04 and VisualizingDP DEATH PUNK JUNComparing JUNGLE Fan-Groups HM HEAVY METAL FP FOLK PUNK EP ELECTRO-PUNK GC GRINDCORE DP DANCE PUNK MC METALCORE #09 Final Approach »Amplifier« TH TRIPHOP - HC HARDCORE CHILL CHILL INDIE ROCK ALT IR ELEC ROCK ALT ELEC INDIE FV ROCK CR ALT FV HR MET ROCK SSW INDIE ALT ROCK INDIE ALT CR IR CHILL LOUN - DT ELEC - ROCK INDIE ALT ELEC INDIE ROCK IR IP ALT ROCK CR ALT FV HR ROCK POP 80s CR SOUL POL FV ALT ROCK ELEC ROCK FIN FV POP ALT FV INDIE SSW FOLK NA ELEC ROCK INDIE ALT EC ELEC IDM AMB ROCK ALT - ELEC DAN ROCK POP ROCK ALT POP AR IR ROCK ALT FV AR NM ROCK ALT POP FV ELEC ROCK ALT AR INDIE EMO ELEC TRA DAN ROCK - FOLK INDIE SSW ALT ELEC HH RnB RAP HH SOUL RAP HH PUNK ROCK ALT ROCK ALT - ELEC AR SJ JAZZ RnB SOUL SAX ROCK CR HH RAP HH REGG HH RAP - DUB HH RAP - HH RnB ROCK ALT HH INDIE RAP RAP HH HH DS - ROCK ALT AR PP HH HH RAP HH - ROCK ROCK ALT POL MET PUNK DM DEATH METAL GRUN GRUNGE LOUN LOUNGE

INDIE ELEC IR ALT ROCK ROCK SSW CR INDIE FOLK TDM TECHNICAL DEATHCONSTRUCTION METAL OF AMPLIFIERPG POST-GRUNGE DT DOWNTEMPO GER HH HH REGG RAP MDM MELODIC DEATH METAL IR INDUSTRIAL ROCK CO CHILLOUT The amplitude indicates the number of artists who INDIE IR ROCK PUNK ALT PDM PROGRESSIVEhave DEATH been tagged METAL with genresGRR associated GRRLS with them. EL EASY LISTENING DDM DEATH-DOOM METAL MH MELODIC HARDCORE R RELAX

One unit is representing one artists. The thickness GM GOTHIC METALresults from how often an artistGR has been GLAM played ROCK by the T TURNTABLISM corresponding user-profil. GR GOTHIC ROCK PR POP ROCK EC ELECTROCLASH vw jake BM BLACK METALUsername BC BREAKCORE The individual categories of genres are encoded by colours. MBM MELODIC BLACK METAL SR STONER ROCK B BREAKZ GBM GERMAN BLACK METAL PR PSYCHEDELIC ROCK E EBM

ABM ATMOSPHERIC BLACK METAL SR SPACE ROCK tommynovember7 Newborn_Bantam er_ay jord2006 Rbraz kariikarii hypnoses yekey JMoneyFresh mikedann cheeks11 scaretactix star_suckers iriha-n r_deckard enry7515 voloko ROCK ALT INDIE IR AR maccalvin pop_art_v linu5 dogangul lubakk Jeroen210 HHCastan GuYGgi pumpkinyourface Jacabo StreetDream Geeeorge swingingfists shandrobedum Irish5 youneedahug iDashX CoolnessBears damionblack simonbee RBM RAW BLACK METAL GR GARAGE ROCK DAN DANCE

PBM PROGRESSIVEThe BLACK five genres METAL most frequentlyPB appearing PSYCHOBILLY in the ME DR GE FWED DT EURODANCE user-profile are shown in a row. Abbreviation and names of a genres resp. sub-genres WR WROCK RB ROCKABILLY S SURF IND INDUSTRIAL SL SLUDGE G GYPSY

FM FINNISH METAL B BALKAN REGG REGGAE ROCK ELEC JAZZ JAP PIAN ROCK ALT INDIE EMO AR INDIE IR ALT ROCK ELEC INDIE SSW ROCK ALT IR POP ROCK INDIE FV SSW SCR PHC HC EMO ROCK ALT ROCK INDIE FV ELEC INDIE ALT POL ROCK FV INDIE ROCK IR HH PUNK ROCK INDIE ALT IR BRIT ROCK MET ELEC ALT IND EC - ALT ROCK EMO ELEC INDIE ROCK FV IR ROCK CR 80s HR ROCK ELEC ROCK ALT CR AMB ROCK ALT ELEC URU TANG ROCK CHR CHR - ELEC ELEC DAN ELEC TH HS ROCK CR ELEC - ALT ROCK ALT PR HH CZE ROCK ALT AR MET ELEC ROCK ALT MET AR CR HH RAP ROCK HH ALT ROCK - ALT ELEC RAP INDIE IR ELEC HH ROCK HH RAP HH UHH DS INDIE IR BRIT ROCK ALT ROCK EMO INDIE ALT PUNK HH RAP HH POP - ROCK AR COUN ALT HH ROCK EMO INDIE ALT SSW - HH RnB RAP POP FV POP INDIE ROCK ELEC HH RAP RnB HH SOUL GP GYPSY PUNK DH DANCEHALL PM PROGRESSIVE METAL R ROMANIAN R ROOTS FV POP RnB HH SOUL P PROGRESSIVEMusic-Genres TR TURKISCH-ROCK RT REGGAETON ROCK ALT INDIE EMO CR ROCK INDIE ALT IR EXP ROCK INDIE ALT IR SSW PM POWER METAL PR PROGRESSIVE ROCK SKA SKA For our topic scope »Comparing Fan-Groups« we formulated SM SYMPHONIC categoriesMETAL for all appearing styles of music in order to faster DUB DUB0/0/0/100 0/48/96/0 0/0/0/70 relate the user to a certain music preference. For choosing the ROCK POP OTHER AM ALTERNATIVE METAL PR POST ROCK RAGG RAGGA colours we orientated ourselves to the above categories as well as their proportions within a fan-group. Within the topic scope SM SPEED METAL MB MANGUEBEAT 92/76/5/0 0/14/91/0 »Cumulation of Genres« we proceeded in the same way, with the HIP-HOP METAL TM THRASH METALdifference that the shade of colours had to be lightly modiﬁed, CAN CANADIAN due to the dark background of the interactive application. PM PROGRESSIVE METAL ELEC ELECTRONIC BRA BRAZILIAN75/23/0/0 0/80/0/89/0 ELECTRONIC REGGAE ELA ELECTRONICA LAT LATIN

POP POP ELO ELECTRO GER GERMAN sototallynotty ymmoTTommy brantlymurphy Mampfi varyavarya Xenostar Deutlich coltbang john_jack numar28 fallore mjwald Mork0vka maruskaj KaRtMaN24 ExoticRoseSyrup eseneka pukemide zdena anjin_san SlowlyWasADemon 1337atron carlitos1213 pmati alesiks Flashpix legallyrouge Jennifer_the_G adamlon theanswer90 bradjh ekushoy brzykaj hiroshikotake Nezram colinlohenry peterlikesmusic andrehhh F FEMALE IDM IDM SPAN SPANISH

FV FEMALE VOCALISTS AMB AMBIENT CZE CZECH

SSW SINGER-SONGWRITER GL GLITCH FR FRENCH JP J-POP POL POLISH FOLK FOLK TECH TECHNO JAP JAPANESE

FR FOLK ROCK NOR NORWEGIAN ROCK MET DM DC MC FOLK ROCK INDIE POP SSW ROCK ALT AR ELEC CR ROCK ALT FV POP GER ALT FV ROCK INDIE ELEC MET ROCK PR PM ALT FV SSW ROCK POP ELEC ROCK ALT ELEC INDIE NM ROCK EXP INDIE ALT PR INDIE IR ROCK HH ALT ROCK INDIE ALT IR SSW ROCK ALT INDIE IR AR ELEC INDIE ROCK EA FV - RAP ELEC TRA FR ELEC - FV ELEC MIN ROCK HH - RAP HH INDIE ROCK IR ALT SSW CHILL JAZZ ROCK CR DT ELEC GER INDIE ROCK - ROCK FV POP ALT INDIE FV SSW ROCK INDIE ALT ROCK - CR JAZZ INDIE MET ROCK HM PM TM - ELEC TRA DAN POP ROCK ALT INDIE ROCK FV ELEC DAN INDIE - FV ELEC DAN POP FV - HH HH RAP GER POP HC MC PP HP MH - HH RAP HH HC RnB - SOUL HH POP ROCK ALT AR PR PUNK HH RAP HH RnB ROCK HH ROCK RAP ALT HH HH GER HH ELEC DAN ROCK HH GER - HH HH RAP INDIE HH ALT ROCK ALT HH RAP HS PR POP ROCK MIN MINIMAL FIN FINNISH 80S 80S MT MINIMAL TECHNO RUS RUSSIAN

70S 70S ME MINIMAL ELECTRO ROCK CR INDIE SSW ALT 60S 60S ET EXPERIMENTAL TECHNO SCR SCREAMO OPM OPM D DISCO HS HOUSE DAN DANCE PH PROGRESSIVE HOUSE JAZZ JAZZ IP ITALIAN POP DH DEEP HOUSE AJ ACID JAZZ VH VOCAL HOUSE SJ SMOOTH JAZZ BRIT BRITISH SH SOULFUL HOUSE SAX SAXOPHON

BP BRITPOP EH ELECTRO HOUSE CLAS CLASSICAL

IP INDIE POP TH TECH HOUSE PIAN PIANO agfay gloomytheo ellis_shop irjman zerohour269 tooweighmirror scaup gttcasey klavir t_seijiro_o bastrion signal74 untitled37 miamucho Welniuxte k-loo Raulitito caitie16 CaptainTails shitbone DontCallMeSurly ladyluck2008 heiko456 ElakNov mathewskyle SzalonyAlien DoubleG19 comradeb14ck pro100bog Kaptiv8 puffcita vamosjan91 BeliasIlKornuto teskoner pri_c_ll colinlohenry shawti01 nichdiemama Rogal91 S SCOTTISH

NW NEW WAVE TRA TRANCE AC ACOUSTIC PT PROGRESSIVE TRANCE INST INSTRUMENTAL G GUITAR VTRA VOCAL TRANCE EXP EXPERIMENTAL HTRA HARD TRANCE COUN COUNTRY ROCK INDIE ALT EMO IR ROCK ELEC ALT AR FV INDIE ROCK ALT SSW FOLK INDIE ROCK ALT FV IR INDIE IR FOLK ALT SSW ROCK ALT CR ELEC INDIE ROCK INDIE ALT CR - ROCK CLAS - ELEC SSW ROCK INDIE ALT CR IR INDIE ROCK ELEC IR ALT ELEC INDIE ALT ROCK HH ROCK ALT INDIE EMO IR FV E LEC - ROCK EA ROCK ALT AR NM ELEC ELEC FR ROCK CHILL FV ROCK ALT INDIE IR POP SOUL - FV ROCK POP ELEC ROCK ALT TRA DAN MET MDM MC DM HC ROCK HH CR ALT ELEC - HS ELEC DAN ELEC ROCK ST ALT INST CLAS ELEC ROCK FV DAN POP ELEC - CLAS ELEC DAN ELEC AMB IDM FV EA HH RAP HH DS RnB HH RAP HH RnB DS HH RAP HH RNB RR RnB FV POP SOUL ROCK ELEC FV ROCK ALT INDIE HH HH GER RAP ROCK MET ROCK MC ALT TM RAP HH - HH DS POP FV LAT INDIE SPAN ROCK HH GER - HH POP - FV HH RnB HH DAN POP RAP ROCK PP POWER POP T TIESTO ST SOUNDTRACK ROCK - ALT INDIE ELEC SP SYNTH POP MTRA MELODIC TRANCE

NA NEW AGE HH RAP HH RnB GR ROCK ROCK NF NEUROFUNK HIN HINDIE SL SEEN LIVE LF LIQUID FUNK CR CLASSIC ROCK PSD PSYCHEDELIC HR HARD ROCK FO FULL ON PST PSYTRANCE INDIE INDIPENDENT GOA GOA IR INDIE ROCK PT PSYCHEDELIC TRANCE EMZADI613 chirimar kisssis atomicus_pl SilentChicken brantlymurphy bse27 d3d-design Vincentrou unicorn_penis alpinestars1 poikao zollko Terantula ingomarr xaver vagavundo lrdcossity VisitKwoon Ric0_Suave ALT ALTERNATIVE 8-B 8-BIT

ROCK INDIE ALT IR AR - HH RnB RAP HIN ELEC INDIE IDM IR FOLK ROCK CR ALT INDIE AR ROCK GER MET ELEC IND FR ROCK ALT FV SSW ROCK ALT AR IR - ROCK ALT INDIE IR FOLK ROCK ELEC ALT DAN CR ROCK FV ALT INDIE SSW - RAP HH HH RnB HH RnB - RAP SOUL - HH RAP HH REGG REGG HH SKA DH RAP HH RAP HH ROCK CR ROCK CR HH RAP - HH RAP COUN HH -

ROCK ALT FV AR POL MET PM DMM PR GM ROCK ALT AR ELEC CR MONITORING AND VISUALIZING LAST.FM Diplomarbeit von Christopher Adjei und Nils Holland-Cunz / Betreut von Prof. Eva Vitting und Prof. Philipp Pape / Fachhochschule Mainz 2008 HÖRERFLUKTUATION Design Topic Scope 2: 49 | 49 04 and Visualizing Fluctuation of Fans

Fluctuation of Fans #01 Idea Development

Step 1 Step 3

WOCHEWEEK 1 1 WOCHEWEEK 22 WOCHEWEEK 3 3 WOCHEWEEK 4 4

HÖRERFLUKTUATION Poland

Poland Germany UK Germany

UK Step 2 Step 4

GROUPGRUPPE WEEK WOCHE 1 1 GRUPPEGROUP WEEK WOCHE 2 2 GRUPPEGROUP WOCHEWEEK 3 3

In this scope we tried to prove that there is a relationship listeners captured weekly. From each user's country and ranking Step 1: This graphic shows what actually happens among Step 3: Individual users are visualized by a line. If a user is no between an artist's tour and the network users mostly listening to we determined a ranking of countries together with a correspon- others with the investigated issue. The differently coloured more to be found among the 1,000 »top-listeners« the user-line his music. We presumed a growing share of users in a country ding proportion of users. groups reduce from week to week. ends at the mean line. within the 1,000 »top-listeners« of an artist, after a performance in that country. As a starting point for the visual approach we Step 2: We then interpreted this observation as a sinus curve Step 4: Lines that stopped before now »leave« the graphic. took among others the principle of the sinus curve, which in a slowly flowing out. Moreover, the newly arriving users »flow« into the diagram by widest sense visualizes a circular movement. This seemed to be the lines coming by from above. very suitable for visualizing the fluctuation within this group of Design Topic Scope 2: 50 | 51 04 and Visualizing Fluctuation of Fans

#02 First Realisation

It was one of the ﬁrst attempts to visualize each user by an individual line. But this did not produce a clear picture of the issue. Design Topic Scope 2: 52 | 53 04 and Visualizing Fluctuation of Fans

#03 Finalizing the Graphic

Graphic 1 Graphic 3

Graphic 2 In the algorithm, producing the ﬁnal graphic, the 1,000 users are and Japan are coloured. The changes of the postions can be being co-ordinated to countries. The vertical position of a stripe noted clearly. representing a country is obtained from the positions of the listeners pertinent to it. Thus it can be that there is a relatively thin Graphic 2: This graphic consists of two »knots«, and thereby it stripe in an upper position because the pertinent users are found in shows the fluctuation within 3 data captures. the ﬁrst places within the1,000 »top-listeners«. Graphic 3: Differentiation of all countries by colour, using the Graphic 1: In this graphic the user-groups from Russia, Iceland complete HSB colour spectrum. Design Topic Scope 2: 54 | 55 04 and Visualizing Fluctuation of Fans

#04 Final Poster and Colour-Coding

92/100/7/0 0/90/55/0 11/26/95/0

100/91/2/0 0/80/94/0 48/0/97/0

80/5/10/0

For our topic scopes »Fluctuation of Fans« and »Album-Release« We tried to tune the colours complementary to each other and give we developed a colour-coding according to continents and divided a »warm« shade to continents with a warm climate. Europe into East and West. Russia within this context belongs to Eastern Europe. DAZUKOMMENDE HÖRER 221 231 244 SINGAPOR URU BULGARI SWEDE PO FRANC NEW ZEALAN AUSTRALI BRAZI INDONESI UN MEXIC NORWA HONG KON FI G NETHERLAND UKRAIN UNITED KINGDO CHIL JAPA B CYPRU C NORWA G ECUAD S ES NETHERLAND TURKE IRAN, ISLAMIC REPUBLIC O UN MAURITIU ARGENTIN CHIL PORTUGA ICELAN FI BRAZI NEW ZEALAN PO BELGIU ITA LEBANO AUSTRALI SAUDI ARABI UNITED KINGDO UKRAIN CZECH REPUBLI C MEXIC SWEDE RUSS GREEC FRANC R LITHUANI ISRAE VENEZUEL PER SERBI EGYP UZBEKISTA TAIWA ES RUSS TURKE SERBI C BELARU QATA JAPA LA MAURITIU ARGENTIN CHIN N KAZAKHSTA JORDA C S ITA LA C BELGIU BELARU HONG KON C C SOUTH AFRIC BRAZI PO AUSTRI AUSTRALI ITA ISRAE PORTUGA S TAIWA UN UNITED KINGDO ARGENTIN MEXIC O KO NETHERLAND PHILIPPINE SWEDE FI G GREEC RUSS NORWA T UKRAIN FRANC MA CHIL VENEZUEL MONAC PARAGUA SINGAPOR S PER O VENEZUEL BULGARI REUNIO BELGIU UR OSTA O O O L O O MA MA I CARA OVAKI TO TO TVI TVI L L L R L MAN A A A LD P P P ERMAN ERMAN ERMAN U U KE OMBI OMBI OMBI EA, I G N N E V R T N N A NL NL NL N N U A A I IVE A A L L E E S Y I I O E E N N E LAN LAN LAN A A A A N N N N R A O A A A ITE ITE ITE Y D G I M M M N R L L L E R I S S Y A N A C L L U A A A A A EP S Y A A A E E S A S A A I I I A A A G G A A A A Y Y Y N E E U S N N N A A A N A B A I I I D D D D D L D D D AN AN AN O O O L L L C N N N I Y Y C D D D A A A O N N N STATE STATE STATE Y Y D D D E E F Y Y Y F F F A A A EDERA EDERA EDERA F S S S S S S M M M TIO TIO TIO N N 02 N

FANFLUKTUATION MEXICO MEXICO MEXICO BELARUS MEXICO MEXICO LITHUANIA LITHUANIA

BEOBACHTETE KÜNSTLERUng TOKIO HOTEL

DATENQUELLE ZEITRAUM DER DATENAUFZEICHNUNG lastfm.de/music/tokio+hotel/+listeners 04.August 2008 bis 24. August 2008

FANFLUKTUATION EINES KÜNSTLERS/EINER BAND Diese Grafik zeigt die Fluktuation der Fans der Band Tokio Hotel in einem wöchentlichen Rythmus. Zu sehen ist ein Ausschnitt von drei Wochen. Auf den zugehörigen Html-Verzeichnisseiten haben wir 1000 "Top-Hörer" Woche für Woche abgerufen und lokal gespeichert. Jeder Strang in der Grafik repräsentiert einen Staat. In jedem Strang sind also alle "Top-Hörer" eines Staates zusammengefasst. Die vertikale Position des Stranges wurde AUFZEICHNUNGEN KW32 KW33WEEK 33 KW34 KW 35 mit dem Ranking der jeweiligen Hörer ermittelt, da jeder Hörer auf einer Position zwischen 1 und 1000 zu finden ist. Aus Platzgründen konnten wir nur die ersten 500 "Top-Hörer" zur Darstellung nutzen. In dieser Grafik ist auffällig, daß Tokio Hotel die größte Fan-Gemeinde in Brasilien und den USA hat, obwohl Sie eine deutsche Band ist. Brasilien hat ca. 72.000 Last.fm Hörer weniger als Deutschland. LÄNDERRANKING DENMARK OMAN OMAN JAPAN IRAN, ISLAMIC REPUBLIC OF DENMARK IRAN, ISLAMIC REPUBLIC OF BOLIVIA LATVIA CYPRUS FARBKODIERUNG 01 ISLAMIC REPUBLIC OF IRAN PORTUGAL VENEZUELA PERU COLOMBIA CANADA BULGARIA 02 DENMARK REUNION NORWAY ESTONIA BELGIUM BELGIUM NEW ZEALAND 03 INDONESIA BULGARIA GERMANY BELARUS SWEDEN HONG KONG COSTA RICA 04 CHINA ECUADOR COLOMBIA SOUTH AFRICA 05 UKRAINE POLAND SPAIN 06 PHILIPPINES COLOMBIA 07 MACAO ESTONIA 08 HONG KONG SPAIN NETHERLANDS BRAZIL 09 COLOMBIA TURKEY IRAN, ISLAMIC REPUBLIC OF SINGAPORE 10 LITHUANIA ITALY URUGUAY BULGARIA FRANCE NEW ZEALAND WESTERN-EUROPE NORTH AMERICA AFRICA 11 SWEDEN EASTERN-EUROPE SOUTH AMERICA AUSTRALIA/NEW-ZEALAND AUSTRALIA ASIA 12 SPAIN POLAND 13 PERU AUSTRIA 14 NORWAY AUSTRALIA

UNITED STATES ITALY 15 UNITED KINGDOM BRAZIL ISRAEL 16 ISRAEL PORTUGAL 17 NETHERLANDS SPAIN 18 RUSSIAN FEDERATION INDONESIA LITHUANIA

19 BELGIUM TAIWAN MAURITIUS ARGENTINA 20 SERBIA MACAO CHILE

21 FRANCE PORTUGAL ICELAND 22 BRAZIL 23 GERMANY FINLAND UNITED STATES 24 UNITED STATES UNITED STATES 25 AUSTRALIA 26 CHILE

27 POLAND BRAZIL 28 SWITZERLAND MEXICO 29 FINLAND ISRAEL NORWAY UNITED KINGDOM HONG KONG 30 ITALY NEW ZEALAND

ARGENTINA POLAND 31 ESTONIA FINLAND MEXICO SERBIA BELGIUM OMAN 32 BULGARIA KOREA, REPUBLIC OF ITALY NETHERLANDS 33 TURKEY LEBANON

GERMANY PHILIPPINES 34 MEXICO AUSTRALIA SAUDI ARABIA SWEDEN 35 AUSTRIA NETHERLANDS UNITED KINGDOM

36 BOLIVIA UKRAINE UKRAINE CZECH REPUBLIC FINLAND 37 CANADA CANADA UNITED KINGDOM 38 PORTUGAL MEXICO CHILE GERMANY 39 ANGOLA SWEDEN ESTONIA GREECE 40 MALAYSIA RUSSIAN FEDERATION RUSSIAN FEDERATION RUSSIAN FEDERATION 41 PUERTO RICO TURKEY SERBIA GREECE NORWAY FRANCE 42 SOUTH AFRICA ROMANIA LITHUANIA CANADA ISRAEL TURKEY 43 CZECH REPUBLIC VENEZUELA UKRAINE BELARUS FRANCE QATAR PERU MALDIVES JAPAN MACAO MACAO SERBIA PHILIPPINES EGYPT LATVIA LATVIA 44 EGYPT MAURITIUS UZBEKISTAN CHILE ARGENTINA TAIWAN CZECH REPUBLIC VENEZUELA PUERTO RICO MONACO CHINA PARAGUAY NICARAGUA SINGAPORE 45 ROMANIA KAZAKHSTAN JORDAN SLOVAKIA PERU 46 ARGENTINA 47 NEW ZEALAND 48 BELARUS RUSS RUSS RUSS UNITED KINGDO UNITED KINGDO UNITED KINGDO I I I AN AN AN UN UN UN NETHERLAND NETHERLAND NETHERLAND F F F ITE ITE ITE IRAN, ISLAMIC REPUBLIC O AUSTRALI EDERA EDERA AUSTRALI EDERA BULGARI G PORTUGA G G D D UKRAIN D FI FI FI ERMAN SWEDE ERMAN NORWA SWEDE NORWA SWEDE ERMAN CANAD CANAD CANAD STATE STATE P STATE FRANC P P TURKE MEXIC BRAZI TURKE BRAZI MEXIC BRAZI TURKE MEXIC NL NL NL CZECH REPUBLI CZECH REPUBLI O O O SPAI NEW ZEALAN SPAI CHIL CHIL NEW ZEALAN SPAI SAUDI ARABI SWITZERLAN SOUTH AFRIC ITA ITA ITA KAZAKHSTA PHILIPPI LAN LAN UZBEKISTA LAN PER PUERTO RIC VENE VENE N HONG KON ARGENTIN ARGENTIN MAURITIU HONG KON INDONESI MAURITIU ARGENTIN LITHUANI SINGAPOR LITHUANI TIO AUSTRALI LITHUANI TIO A TIO A A MA C BULGARI C D I C PORTUGA BULGARI PORTUGA LEBANO CARA R BELGIU BELARU R BELGIU URU REUNIO BELGIU O UKRAIN BELARU O ECUAD UKRAIN O E ICELAN ES NORWA ES AUSTRI ES O JORDA TAIWA O B FRANC L ANGOL B L SERBI N L CYPRU GREEC ISRAE ISRAE SERBI L N N JAPA JAPA N MAN Z OMBI O MAN Z OMBI CHIN O OMBI L QATA L CHIN MA TO L AYSI TO EGYP CHIL TO EGYP UEL UEL G L L M G N M I UA M I A A A A N O N E E A N E D V N N U N N Y Y Y A L O D D E Y L V N N S N D D Y Y Y Y U S N N S L L O S D S Y E R Y Y O S I I I I M I I I M M N N N N A A A A A A A A A A G O R N N N N D D E A A A A A A A A S S S R K E E E L L C S S F D D Y Y L L A A A A A A A A A A A A A A G E E C S T T AUSSCHEINDENDE HÖRER 224 240 247

KONZERTE STADT/DATUM/VERANSTALTUNGSORT SAYREVILLE/07.08.2008/STARLAND BALLROOM CLEVELAND/11.08.2008/HOUSE OF BLUES // DETROIT/12.08.2008/FILLMORE DETROIT // CHICAGO/15.08.2008/HOUSE OF BLUES SAN FRANCISCO/19.08.2008/THE FILLMORE // LAS VEGAS/22.08.2008/HOUSE OF BLUES KW32/04.08.2008–10.08.2008 KW33/11.08.2008–17.08.2008 KW34/18.08.2008–24.08.2008

MONITORING AND VISUALIZING LAST.FM Diplomarbeit von Christopher Adjei und Nils Holland-Cunz / Betreut von Prof. Eva Vitting und Prof. Philipp Pape / Fachhochschule Mainz 2008 Design Topic Scope 3: 58 | 59 04 and Visualizing Album-Release

Album-Release #01 First Scribbles #02 First Realisation

Graphic 1 Graphic 2 United Kingdom United States United Germany

Netherlands

United States United Canada United Kingdom

Germany

Brazil

Canada Netherlands Mexico

Finland

Argentina Austria

Mexico

Brazil Austria Finland Argentina Japan

Norway Norway

Japan Sweden

Sweden Kong Hong

Hong Kong Hong

Alice Cooper of Republic Korea, Korea, Republic of Republic Korea, The Verve

Portugal Portugal

Turkey Along came Turkey Forth

Spain

Lithuania Spain

a spider Lithuania

Italy Belarus

Switzerland Belarus

Estonia Ireland

Latvia Ireland Switzerland

Estonia Latvia

Slovakia Italy

Ukraine Czech Republic Czech

Poland

Russian Federation Ukraine

Slovakia Czech Republic Czech

Russian Federation Poland

How successfully does an album spread after its release on countries after each week the largest increase of listeners took Dispersion: The system of the circular diagram holds the advan- Graphic 2: Last.fm? How many users are being reached within a certain place. During the period of realisation we tried to visually describe tage of using sectors allowing a lot to arrange. These graphics, The new album »Forth« by The Verve succeeded a lot in period? Starting from these questions, we tried to get conclusions the phenomenon of dispersion that applies to this issue (album- however, do not allow to see the weekly intervalls of dispersion. Western and Eastern Europe. as to where (which countries) an artist is or is not successful with relaese). In the final result we cut a circle into individual sectors, The further examples (see next pages) do allow this. We also clearly his new album. For shortly after its official release the new album each standing for a country. The weeks that have past since the see, as well, the difference between the units of area and several would be published on Last.fm as far as the artist has got a Last. publication on Last.fm »run« as rings from the center outwards. rectangles that represent the share of new users for each country. fm profile. Here, too, we are making use of the weekly ranking of After each ring the new number of users is indicated by little 1,000 users having heard the album mostly and appearing on rectangles, and now we can see how many new album listeners Graphic 1: index-sites by Last.fm. After the publication every week we captured come by in each country every week. Hence it can be concluded The dispersion of the new album »Along came a spider« by Alice the ranking of the 1,000 users and sorted them according to the if the album is spreading faster in one country than in another. Cooper. The weeks are not yet comparable here. In comparison to countries they belonged to. From this you can conclude in which others this album does not seem to be particularly successful. Design Topic Scope 3: 60 | 61 04 and Visualizing Album-Release

#03 Finalizing the Graphic

Japan

Australia

Israel

Brazil

Turkey Canada

Estonia United States

Czech Republic

KW 36 KW 37 KW 38 KW 39

United Kingdom Poland

Netherlands

Russian Federation

Germany

Finland

Graphic 1 Graphic 2 Italy

Sweden Spain

Norway

Graphic 1 and Graphic 2: Those graphics are based on the same principle but an additional line in graphic 1 emphazises the weekly splitting. Additionally, the areas of graphic 2 are being divided into units of rectangles: one unit is one user. 03 ALBUM-RELEASE

BEOBACHTETE KÜNSTLERUng OASIS/AMY MACDONALD/GABRIELLA CILMI/ NELLY/HADOUKEN!/ALICE cooper

API-METHODE ZEITRAUM DER DATENAUFZEICHNUNG lastfm.de/music/Artist/+listeners 28.Juli 2008 bis 02.November 2008

ÖRTLICHE AUSBREITUNG DER HÖRER NACH DER NEURSCHEINUNG EINES ALBUMS Bei dieser Grafik beobachteten wir die 1000 “Top-Hörer” eines neu erschienenen Albums über einen Zeitraum von jeweils vier Wochen. Wir wollten wissen, wie sich die Hörer eines neuen Albums auf die Staaten verteilen. Ebenso beobachteten wir für jeden Staat den wöchentlichen Hörerzuwachs. In einer Kreisanordnung stellen wir diese Entwicklung für 19 verschiedene Staaten dar, die sich in jeder Grafik immer an der gleichen Position 15 16 15 16 15 16 befinden. Die farbliche Kodierung basiert auf dem gleichen Prinzip wie in der Grafik “Fanfluktuation”. 17 17 17 14 14 14 18 18 18 13 13 13

19 19 19 OASIS AMY MACDONALD GABRIELLA CILMI FARBKODIERUNGng 12 12 12 DIG OUT WÖCHENTLICHE DATENAUFZEICHNUNG VON 06.OKTOBER 2008 BIS 02.NOVEMBER 2008 THIS WÖCHENTLICHE DATENAUFZEICHNUNG VON 18.AUGUST 2008 BIS 14.SEPTEMBER 2008 LESSONS WÖCHENTLICHE DATENAUFZEICHNUNG VON 28.JULI 2008 BIS 24.AUGUST.2008 KW 41 KW 42 KW 43 KW 44 KW 34 KW 35 KW 36 KW 37 KW 31 KW 32 KW 33 KW 34 11 11 11 YOUR SOUL IS LIFE TO BE LEARNED

ASIA

1 1

RELEASE-DATUM 04.OKTOBER 2008 RELEASE-DATUM 31. JULI 2008 RELEASE-DATUM 29.APRIL 2008 1

10 LABEL BIG BROTHER LABEL MERCURY

SOUTH AMERIKA 10 LABEL UNIVERSAL ISLAND RECORDS 10

TAGS FEMALE-VOCALISTS/POP/JAZZ

JAPAN TAGS BRITPOP/ROCK/BRITISH TAGS FEMALE-VOCALISTS/ 2 2 2

9 9 SINGER-SONGWRITER/ 9 ISRAEL ACOUSTIC

AUSTRALIA

3 3

BRAZIL NORTH AMERIKA 3 TURKEY 8 8

15 16 8

4 4

4 4 17 4

7 7

14 CANADA 7

5 5 5 6 6 18 6 ESTONIA 13 19 USA

12 CZECH REP.

EASTERN EUROPE 1 UK

POLAND 10 2

9 NETHERL.

3 3

RUSSIAN FED. 4 7 GERMANY

5 6

ITALY FINLAND SPAIN

SWEDEN

NORWAY

WESTERN EUROPE

Design Topic Scope 3: 04 and Visualizing Album-Release 15 16 15 16 15 16 17 17 17 14 14 14 18 18 18 13 13 13 #04 Poster and Colour-Coding HADOUKEN! NELLY 19 19 19 12 ALICE COOPER 12 MUSIC FOR AN 12 BRASS WÖCHENTLICHE DATENAUFZEICHNUNG VON 06.OKTOBER 2008 BIS 02.NOVEMBER 2008 ACCELERATED WÖCHENTLICHE DATENAUFZEICHNUNG VON 18.AUGUST 2008 BIS 14.SEPTEMBER 2008 ALONG WÖCHENTLICHE DATENAUFZEICHNUNG VON 11.AUGUST 2008 BIS 07.SEPTEMBER 2008 KW 41 KW 42 KW 43 KW 44 KW 34 KW 35 KW 36 KW 37 KW 33 KW 34 KW 35 KW 36 11 11 KNUCKLES 11 CAME A SPIDER

CULTURE

1 1 RELEASE-DATUM 25.JULI 2008 RELEASE-DATUM 17. OKTOBER 2008 RELEASE-DATUM 24.APRIL 2008 1

LABEL SPV

LABEL Universal Music Australia 10 10

LABEL ATLANTIC RECORDS UK 10 Pty. Ltd.

TAGS HARD ROCK/CLASSIC ROCK/

2 2 TAGS NEW RAVE/ELECTRONIC/ 2

ROCK 9 9

TAGS RAP/HIP-HOP/RnB BRITISH 9

3 3

3 3 3

8 8 8

4 4

4 4 4 4

7 5 5 5 6 6 6

A C I R E M A ASIA H T U O S

A I

L J A A A

R P T I A S S L N I R U Z A A A E L R TURKEY B NORTH AMERIC 15 16 17 14 CANADA 1 8 E ST 13 ON A IA 1 US 9 12

CZ ECH REP . 1

N EUROPE EASTER 1 UK

POLAND 10 2

9 NE

THE . 3 R

ED L. N F 8

SIA

US 4

R 7 GERMANY

5 D 6 The colour-coding of each sector is based on the same principle as

I T A FINLAN L N in topic scope 2. In each diagram the different countries are placed Y E S P D Y A E A I N W W S R at the same position.

E S TE

R N E

U R O

P E

Cumulation of Genres #01 Question/Idea

We wanted to find out where in the world e.g. »Metal« or countries and continents. For this we chose a projection based on this to an interactive level, and by means of an application make Graphic: By the first mapping of the coordinates, using the »Electronic« are in trend. So, for a period of four months, we first the so-called Mercator formula, as this appeared to us most it empirically perceptible for the user. He is to autonomously Mercator formula, we could see how our concert data acted. collected all worldwide concert data being registered with Last. suitable because of its extremely low perspective distortion of the regulate distance and perspective in the 3-dimensional space and Besides figuring the european continent, we could already fm. Through the Last.fm application programming interface (API) relevant parts of the world. In order to be able to more precisely to have the possibility of finding information about concerts or realize where areas of concentration of certain genres of music and by the command »geo.getEvents« we got to xml-data-files, localize particular concerts at the same place of performance we genres of music himself. appeared, as well as the volume of particular concert perfor- containing, besides the artists as headliners, the places and tipped over our projection in a 3-dimensional space, and »piled« mances. Yet, still there was the problem of exactly localizing dates, the titles and coordinates as well as the complete lineups the number of the concerts one upon another in the form of cubes. countries, towns and concerts. of concert performances. So, using the coordinates, we then By this, the new concert scenery receives the impression of a projected all performances on a map indicating the borders of skyline inviting for further exploration. We decided to elaborate Design Topic 4: 66 | 67 04 and Visualizing Cumulation of Genres

#02 2D and 3D

joëâ~ì hê~ãÑoêë lëäo ioåÇoå dÉåí háÉï mÉêìÖá~ poÑá~

First test for a typographic solution

First test of showing the proportions of concerts. In this example the precise localizing has failed.

Transferring the projection into a 3-dimensional graphic makes it of the connecting lines. The alternating switch between the now easier to localize individual places of events. The size of the 3-dimensional projection of concerts and the 2-dimensional »piles« of concerts in the form of little cubes allows conclusions circular arrangement of their data inclusive their connecting lines as to the number of concerts, at one point of the coordinates. enables the user to keep the view over his collected information The locations may now be compared with each other, and in an and their origin. Apart from this, the user has the possibility of interactive way the needed information can be indicated in a having indicated the tour-routes of particular artists and of tracing 2-dimensional form of a circle. Since this is a matter of interac- them back, as well. This enables the user to trace up the countries tive application we were at this point facing the question of the and towns that are or have been preferred by bands or artists on operability for the user. While, looking for music genres he keeps their tours. Interactive test of combining the world map with connecting lines of an circular chart the view over all his collected data and their locations by means Design Topic 4: 68 | 69 04 and Visualizing Cumulation of Genres

#03 Realizing Application

One column is built by several On »Roll-Over« over the cubes On »Roll-Over« we can see The navigation gives you the Zoom-Button Options of Selection of several months cubes. One cube is representing we can see above the approp- event names inclusive their possibility of selecting different different views of which concerts are a concert, and one column a riate columns banners labelled performing artists in an circular genres. indicated certain city. by particular city names. arrangment.

Colour-Coding

Rock/Pop Metal Electronic Other Design Topic 4: 70 | 71 04 and Visualizing Cumulation of Genres

#04 Realizing Poster

The picture is a cutout from the exhibition poster. We can see Europe inclusive all the concerts of »Rock« and »Pop«, which were taking place in July 2008. Furthermore the performances of the band »Blondie« are connected by lines, which are following the tour-routes chronological. Formative Elements Poster-Label 72 | 73 05 Typeface

Formative Elements #01 Composition of the Poster-Label #02 Typeface

02 lutz headline Topic Scope FLUcTUATION OF FANS GRAVUR CONDENSED OBSERVED ARTIST TOKIO HOTEL Artists, User-Groups or Genres we observed

DATA SOURCE PERIOD OF DATA CAPTURING GRAVUR CONDENSED lastfm.de/music/tokio+hotel/+listeners 04. August 2008 - 31. August 2008

THE FLUCTUATION OF FANS OF AN ARTIST/OF A BAND This graphic shows the fluctuation of the fans of the band Tokio Hotel in weekly intervals. We can see a cutout of four weeks. On the Last.fm top listeners index sites we captured and saved 1,000 top listeners each week. Each stripe in the graphic is representing a country. So, within each stripe all top listeners of a country are GRAVUR CONDENSED being comprised. The vertical position of a stripe was found by its top listeners, as each listener is ranging in Explaining Text a position between 1 and 1,000. For shortage of space, it was not possible to utilize more than the first 500

top listeners for presentation. Noticeably this graphic shows that the largest fan groups of Tokio Hotel, though

being a german band, are found in Brazil and the USA. Brazil has got approx. 72,000 Last.fm listeners less than germany (status: august 2008).

COLOUR-CODING Gravur Condensed Gravur Condensed

Explaining Graphic for the Colour-Coding or for the Composition of a Graphic Gravur Condensed

WESTERN EUROPE NORTH AMERICA AFRICA EASTERN EUROPE SOUTH AMERICA AUSTRALIA/NEW ZEALAND ASIA

For legends, labels and descriptions of our visualisations and graphics we inclusively used majuscules being still well legible posters we chose »Gravur«. This seemed to be particularly even in a very small size. »Lutz« in its function as a headline suitable for these purposes, as its round shapes adapt themselves font was meant to be a real contrary and contrast to »Gravur«. On each poster the label is applied in the same place. It gives very well to the graphics. It is not concurrent to them (the information about what is to be seen on the poster. graphics) and yet remains strong in character. For labelling the Resumé What have we learned? 74 | 75 06 Bibliography

What have we learned? Impressum

In pursuit of data-visualizing, which within our project includes data-visualizing. This hurdle had to be overcome by either trying Monitoring and Visualizing Last.fm the comprehension and utilisation of data plus the conceptional to elaborate and to combine forms of illustration or looking for Documentation of the Final Year Project of elaboration and programmed visualisation of certain issues, we original visualisations and use of form. Thereby we met with a Nils Holland-Cunz und Christopher Adjei gained a deeper inside into information-design and its multiplicity. number of tasks that do not directly have anything to do with We noticed that several sets of data are interesting only if there design as such but were absolutly necessary for the realisation, Superviser: Prof. Eva Vitting, Prof. Philipp Pape is an interdependence and if they allow statements or conclusions starting from the advance and improvement of our programming University of Applied Sciences Mainz 2008 that would not be possible in case of considering each set for skills up to the research for more complex mathematic formula or Course of Studies: Design itself. It is only the kind of visualisation plus the appropriate methods of stochastics. Even if all the visualized data gathered by performance that allow to make clear to the viewer what is Last.fm can finally only describe what is happening within this happening among the data and which new correlations are hiding network an impression of general trends and tastes of music can behind. During the visualizing-process we often came across some be obtained. difficulties concerning the psychological perception. We often saw ourselves challanged to see to a balance between graphical aesthetics on one hand and a precise visual presentation of an issue on the other hand. The working-out and testing of various solutions over and over again was in our focus and finally let us to the actual results. In the beginning we found ourselves with typical forms of illustration that often appear in relationship to

Bibliography

Jacque Bertin Sémiologie graphique, de Gruyter 1974

Edvard Tufte Envisioning Information, Graphics Press 2003 Beautiful evidence, Graphics Press 2006 The visual display of quantitative information, Graphics Press 2007

Philipp Oswalt Atlas of shrinking cities, Hatje Cantz 2006

Ben Fry Visualizing data, O’Reilly 2007

Ira Greenberg Processing, Creative Coding and Computational Art, Friends of Ed 2007