Since a couple of years the concept of “post-digitalism” is shaking the heads of scientists, artists and philosophers. I stumbled upon this when reading a paper of Kim Cascone about the potential which lies in the limits of software-based systems for generating and analysing music. I tried to dig deeper into this topic over the past 2-3 years and then I put it to the side. But very recently some intimate debates at barcamps (e.g the Palomar 5 summit and the FMCs in Mannheim and the a2n in Berlin and .. and … and …) popped up again around this theme. Further brainfood was and is delivered on a daily basis by the very mundane fact that I still use Facebook .. too much … and the heavy interaction between the FB network and the real social network is driving me nuts …
This Saturday I met Jonathan Imme at FMC2010, a bright self-acclaimed digital native and major driving force behing Palomar5. When exchanging a little bit of our latest experiences on facebook and foursquare we agreed: it’s not a matter of the name of the game, the post-digital-neo-analogue movement is already lurking around the corner…
FMC2009: I presented a sketch about my very personal need for semantic music recommenders including automatic storytelling. In 2009 this raised very controversial discussions.
FMC2010: 1 year later we have already intermediate results and a first demo. Furthermore we are already cooperating on a conceptual level and exchanging first datasets with leading German-based startups (AUPEO, MUSICLOAD). At this year’s FMC I had interesting conversatons with SIMFY and EMI, further action ahead …
(Image taken from VisualizingLastFm)
Our call at SMC2009 is still open for inspiring submissions beyond the state-of-the-art of “Visualisation of Music”. Don’t hesitate, put your wildest ideas into a 1-page summary, submit until the 8th of June.
Meet all the dedicated addicts in Porto at SMC2009. Further details are just 1-click away.
The line-up of this year’s German Blogger Event #1 is incredible. Lawrence Lessig and Cory Doctorow are included and for sure you will meet all the famous A-blogger, B-blogger, pop scientists, celebrities and hopefully some forethinkers. I am really looking forward to join the event!
Wolfram Alpha entfacht einiges an Erwartung und Neugierde. Freunde, Kollegen, Presse, Funk und Fernsehen fragen an, was es wohl damit auf sich hat. Zahlreiche Blogpostings wurden bereits verfasst, eine Suche mit dem Memetracker von tiqqer hat viele Links herausgefischt. Natürlich habe auch ich einen Beta-Testaccount unter WolframAlpha.com beantragt … und muss weiter auf Antwort warten.
Andere auserwählte KI-Forscher und Unternehmensgründer im Umfeld des Semantic Web wurden aber bereits von Stephen Wolfram versorgt. Die Einschätzung derzeit scheint: Wolfram Alpha ist eine spannende Frage-Antwort-Engine, die vorwiegend sehr gute Ergebnisse zu faktischem Wissen liefert. Die zugrundeliegende Wissensbasis wurde von mehr als 100 Mitarbeitern auf Qualität überprüft und in einen Formalismus überführt, der speziell abgestimmte Reasoning Engines unterstützt. Beides ist motiviert von der Theorie der zellulären Automaten. Und jetzt?
Abwarten. Die Experten und Mitbewerber sind durchaus angetan, aber sehen auch die Limitationen, die dem Erfinder aber sehr wohl bewusst sind. Die Journalisten finden es einfach toll, also die Geschichte drumherum. Ein perfektes Meme, das derzeit große Runden dreht.
Weiterführende Links, die noch mehr Links enthalten anbei:
9 min about recommender systems for music and in general. Including statements of Last.fm and Pandora founders, Kathrin Passig, Forrester Research … and a few bits by Stephan Baumann.
(German only)
After all those years in basic research I finally have to accept that I live in a kind of bubble. It is the “10-years-ahead” bubble. From time to time I tend to loose contact to earth when thinking about “must-have apps” for mainstream customers. You develop a cool technology, you have a great community idea already in place but … it takes time to get the things out there and the needs and required technology and social patterns have to be in place.
Sometimes minor improvements may lead to greater user acceptance of such a “blue-sky” community. I hope that the spherical tag cloud which we have now in our open-source project OPENEER will help to increase the numer of users contributing details and links about their best moments in life with music.
Oscar Celma has been working for a couple of years on music recommendations. I know him since ISMIR2004 and over the years we met regularly and exchanged ideas, sketches, prototypes and more “philosophically influenced thoughts about the future of music in general”. Finally he finished his Ph.D about this topic. The written dissertation gives insights into the top-notch state-of-the-art in recommender systems fusing content-based, collaborative and innovative network-centric features.
Starting in November we will investigate the matter of Social Media Mining in Berlin. Therefore Kaiserslautern and the new Berlin office of DFKI will be the sites where we conduct this research. The move to Berlin helps a lot to get into closer personal contact to other scientific and commercial players over there.
