One of the problems that the field of Music Information Retrieval tries to address is music recommendation. Much of the technology draws from techniques developed in textual information retrieval systems, while treating the ‘document’as the encoded sound itself. While the technology for audio recognition is being developed (speech, music, noise, etc.), MIR research relies heavily on human tagging to generate music recommendations and genre-based classifications.

What is entirely missing to me is, well, everything else. The research is necessarily focused and narrow, but if you conceive the full spectrum of music as it manifests throughout history and across the globe, the “music as product” concept applies to mostly the international economic model of music. This model drives research into developing more nuanced genres in more popular types of music, and more token acknowledgement of less popular genres that still have substantial market share.

Amidst all this, what is obviously missing to me are the broader music information behaviour of individuals beyond the desire to find music. Consumers of popular music don’t just consume the song, they consume the culture, the history, the fan base, a whole ecology of information and resources that they have an interest in. It’s a wicked problem, at a broad conceptual level, and even in traditional settings such as music libraries, or novel knoweldge exchange platforms that facilitates “direct collaboration rather than a series of directed monologues acting as a makeshift conversation”.

The conceptual picture is coming together for me, now it’s time to scale back and look at what the milestones needs to me. Systems are not built on an idea after all, time to take it back to the “line by line” level.

But first: lots of sleep and relaxation!

In addition to proposal writing, some old thread are getting picked up. After meeting Kiku Day at the Music Knowledge Conference, we co-submitted a proposal to the DIY Citizenship conference and got accepted! This is exciting, except for the fact that it’s not a paper we need to submit, but an exhibition space. The opportunity to be creative is always exciting, and I am looking forward to what we’ll come up with. Greg Adams and I have also been in more conversation about his Banjo project. He was about the one person who could read my music knowledge organization model paper and actually go through it with genuine interest and provide insightful feedback.

I also made some new contacts in my own city that are pursuing research-based enterprises relating to music. One of them is Sina Fallah, who brings a cognitive science perspective to address the lack of music education programs for students with reduced aural, visual and motor abilities due to different reasons. Another is an MIR researcher whom my brother introduced me to, who recognizes the complimentary nature of a community development and software development. His team has a weekly that might serve as a great place for me to not just bounce my ideas around, but apply them in their development projects.

While these certainly have been great highlights of the month, I am also taking the time to slow down a bit and step out of the constant motion for a bit. I discovered a great park nearby that’s perfect for a picnic or a lazy day in nature. Found a new cafe with a frozen yogurt that I am addicted to. Life is a marathon, and recently I was reminded that the price of trying to stick with the pack may actually cost you the race. So I’m dropping back a bit from the pack, pat myself on the back on the gap I have closed between my starting point and the finish line I envisioned, and adjust my perspective to recognize that this is just one of many milestones to keep me going.

Life is too full of wonderful things to let them slip by unnoticed, even if you are in the middle of a race.

As a case study, Metaweb (recently acquired by Google) is a form of knowledge organization systems, that aggregates existing online content providers around topics of interests. Freebase, their open-source database, organized in the form of an ontology, with a page dedicated to the discussion and development of their Music Schema. Music entities as they describe it could be reviews, concerts, album releases, and there are in-depth discussion on the kinds of ongoing issues that are still opened and unresolved. Music and the performing arts at large already has its own cataloguing and classification challenges and rules within the traditional library setting. In the context of emerging ICTs, the design of online knowledge organization models has a host of new challenges as well as opportunities to capitalize on. (My thoughts on this in the domain of music is summarized here.)

Values are reflected in any kind of design decisions that are made. Forms of technology in the past embodied the values of an era. Think of the materials that ancient manuscripts were produced with, the advent of the industrial age, the invention of the light bulb. The difference today is the speed at which new technologies are being introduced, as well as the phenomenal growth and adaptation that characterize many successful social media enterprises today. With the increasing amount of noise in the shape of new applications and online tools, the ability to qualify and quantify the fundamental nature of the communication and social processes within which a project is situated, distinguishes it from those that are simply imitating the hype of the moment.

As I develop a knowledge organization model that facilitates the discovery of music education content, it becomes clear that it requires a unique conceptualization of community. The needs of a domain-specific community is beyond the scope and capabilities of existing open-source social media tools design for the facilitation of “online communities”. It also becomes clear that the creative process cannot begin with a review of existing tools, but an in-depth analysis and understanding of community processes and user behaviour of the target audience. It doesn’t have to take forever, and it is not implied that this step has to be ‘completed’before you begin to design, but it is certainly the first step in an iterative and reflexive process that combines research methodology, community and user engagement.

The Reading in First Nations multi-node video-conferencing event was a success. We connected with people in Toronto, Thunder Bay, Sioux Lookout and Keewaywin (a first nations community in keewaytinook Okimakanak), with local facilitators (Nadia, Mark, Daniel & Marta, and myself respectively) to bring out the different perspectives. The introductions and keynote is already up on the website, and the rest of the feed will become available and archived. It was amazing to see all the technical, virtual, physical and communal elements come together, and the experience of working with an inspiring team of individuals, without whom none of it would have been possible.

It was an great opportunity to see a perspective on Canada’s great north, to be away from the constant bombardment of advertising campaigns and going at nature’s pace instead of my calendar’s pace. What was most rewarding were meeting the locals, albeit briefly. The things you learn in a face to face encounter is more than any paper or article can tell you, and it helps you better define and operationalize the scope of future initiatives and research. The most interesting question is how to move forward. I have a few ideas, but first I have Oxford to tackle.

In addition to presenting at the Music Knowledge Conference, I am also hoping to pay the Oxford Internet Institute a visit. It’s an opportunity to meet the specialists who are interested to the two key areas to my own focus; those that try to understand the nature of musical knowledge, and those that try to understand the dissemination of knowledge using ICTs. It is exciting and isolating at the same time to be bridging between different domain interests, and I am very grateful for the opportunity to address this gap during this visit. I hope to make the most of it, meet as many people as I can, and learn as much from them as possible. The hard work is done, the rest is up to serendipity and luck.

Lay out the key features of the research interest that calls out your name. Given that, how would a canonical sociologist (a traditional, quantitative-minded social scientist) design or “set up” this research question? What kind of data would be sampled or gathered? How would the hypotheses be tested? What kind of steps would be taken to guard against bias, and to make sure that the data were generalizable?

I would like to study the learning and information-seeking behaviour of individuals who are trying to learn a musical instrument, using guitar as an example. “How do guitar students learn how to play the guitar by using Information Communication Technologies (ICTs)?”

In order to achieve a ‘random’sample, controls for variables need to be considered by collecting a wide range of data about age (young prodigies to mature hobbyists), literacy or education level (ability to read or follow instructions, experience with music education), physical or mental capacity (small hands, visually impaired, etc.), physical location (city, suburb, small town), comfort level with ICTs (information addict, technology luddite), motivation for learning (love of a musical genre, pop icons, access to instrument and information, etc.), amount of time they have devoted to learning (regularity of individual and group lessons, time devoted to practicing, other opportunities to play with others) and the level of achievement (by standards defined by music schools and conservatories). Such a variety of perspective is desgigned to ensure that the data collected are not unreliable due to inadvertent bias in the selection of subjects.

In addition to data about who the subject is, one would also interested in how they use ICTs as part of their learning initiative. The frequency of use (hourly, daily, every other day, weekly) of particular ICTs (paid websites, online videos, free sheet music, instructional programs, how do you even define ICTs??) and for what learning objective (to learn a song, a technique, mechanics of the instrument, historical and cultural context, to just ‘browse’).

These factors represent different aspects of the subject in question, and provide multiple ways of measuring higher level data, such as their personal commitment to learning music. The survey would be designed to collect interval data, from which a hypothesis or generalization would be developed to the effect of, “This is how ICTs are being used to facilitate music education in this particular group.”, based on a statistically significant type of learning or information-seeking behaviour that is found in one or more identified groups. There really isn’t anything to test, as I don’t start off with a hypothesis.

Why is that way of studying your particular question not adequate for the question you want to ask?

Good question. There’s a lot of historical-cultural factors that are not accounted for. A literature and case study review of music education practices in theory and practice is lacking in the approach above. In particular, education as a practice is dynamic and constantly changing. If my interest is in the emerging phenomena related to music, learning, and technology, the quantitative approach will be highly contingent and thus highly unreliable, as valid as the data gathered may be. (I have a feeling this is where the exercise is going to turn the attention to Qualitative approaches…)

What question concerning the social world would you like to investigate if you were absolutely guaranteed you would not fail? Be as ambitious and wide-ranging in your thinking as you want.

I want to investigate the way music intersects the different dimensions of human experiences. This is an old rant and sounds perhaps a bit frivolous; more like a day dream than a research or investigative agenda. Yet, the musical dimension of our existence — history, culture, philosophy, art, sciences, engineering — gives me an entry point in exploring just about anything in this world. In the context of the social science, and perhaps more specifically in the tradition of information science and related fields, I want to investigate the nature musical knowledge. There may be some philosophical and critical reflections, but on a practical level how is music knowledge classified and in what context? How is musical knowledge generated, stored, transmitted, and forgotten? What role do Information Communication Technologies play in transforming the assumptions, or challenge the ideas of ‘tradition’, that we associate with music? What are, if such things exists, the universal and the particular ways in which humans interact with and respond to music (or perhaps simply sound in abstraction)? Are there design, development and implementation considerations for systems that handle the delivery of music-oriented products and services? How do differing world views affect our perspective on any of the questions above? (Perhaps I am getting philosophical afterall.)

What kind of questions might you want to spend several precious years of your life investigating? What kinds of things about the world worry and provoke you? What would you like to know that you don’t already know? Most important, what kinds of questions do you find interesting enough to get you out of bed in the morning with energy and excitement, at least on most days?

• What are the recent innovations in music education involving ICTs that serve long distance students? (For the purpose of mapping the individuals, groups and institutions that share similar or related interests and world views.)
• What is the information seeking behaviour of someone who wants to learn an instrument in the context of ICTs? (For the purpose of understanding emerging learning or information weeking behaviour in the context of music education.)
• How can you mine and reorganize the incredible amount of learning  resources online that relate to guitar? (For the purpose of improving the level of information access, as well as the quality of information accessed.)