What's Next for Music Search?

In his recently published memoir, Idea Man” (Penguin Group New York 2011),  Microsoft Co-founder Paul Allen stated, “A database is only as good as its interactive search functions.  If people can’t find and download what they’re looking for, it’s like a vast library with no call numbers.”

While that assertion is most certainly accurate, and possibly obvious, it is also something that needed to be said.  The art of the search, while greatly improved in recent years, is not yet as intuitive or capable as it should or could be.  This is true for the majority of Web search tasks, and also true for application-specific objectives like precision music search.

Moving In the Right Direction

Consider recent moves by leading Web vendors in the area of digital music sales and marketing.  Microsoft®, Google®, and Amazon® have all announced major initiatives in this market sector in the past year.  While each company has focused resources on various aspects of music classification, marketing presentation, search formulation, and customer access—it is probably fair to postulate that, as of now, no single company provides a search and delivery system that can be shown to be truly optimized for a unique customer.

For instance, Amazon.com now offers "music in the cloud".  This service permits shoppers to search for their music, purchase as desired, and then store the music in the "cloud"—i.e., on remote Amazon servers.  To many this will be a valuable concept.  With music acquisitions stored in the cloud, there is no need to carry the digital files from place to place.  Instead, portability and instant access require a broadband-class Internet connection.

Amazon proclaims that users can "Shop 16 million songs...", and, of course, reproduce them on compatible players.  Song diversity is a good thing.  Amazon's efforts should be praised.  But is there an easy and efficient way for any single person to consistently identify and select his 40 or 50 favorites of the moment—from among such a "vast" collection of 16 million songs?

The difficulty of the challenge is emphasized when it is understood that, for most music lovers, their "high-affinity" song shopping list will need to be identified and extracted from a multi-faceted creative canvas; a popular music repository formed one song at a time over a period of 6 decades, and comprised of more than one-hundred music genres, and thousands of artists.  The possible combinations of artists, genres and eras is almost limitless—and every individual will have developed a different set of likes and dislikes.

A U.S. teenager born in 1994 may like, in 2011, Beatles songs from 1964.  And, a U.S. born adult, a "baby-boomer" minted in 1955, may not like Beatles hits, but may react favorably to Rolling Stones tracks.  Both Beatles and Stones hits can be broadly classified as "British Invasion" music from the mid-1960s, however, to a certain individual, the Rolling Stones might represent ROCK while the Beatles typify POP.  An individual may enjoy 1960s ROCK genre music, but disdain 1960s POP music.  A customer-configured filter mechanism could help by hiding unacceptable genres—if the music profile could be uniquely connected with an individual every time they shop for music!

From our early research, it was unclear how ANY vendor's search tools could quickly narrow a potential 16 million song universe to a more manageable number such as 1,000 or 2,000 target songs—while persistently preserving a particular customer's unique song-affinities.

Overwhelmed by the Numbers

By the math: 16 million songs, averaging, for example, 3 minutes per song equates to 48 million minutes of music.  Dividing 48 million minutes by 60 (the number of minutes in 1 hour), produces a total of 800,000 hours of music.  Knowing there are about 8,760 hours in 1 calendar year, and dividing 800,000 hours by 8,760 hours, demonstrates that for any single person to listen to each of the 16 million on-line songs 1 time only—would take 91.3 years* (*This projection allows no time off for sleeping—or any other activity.)

FYI:   To put this another way, imagine that an individual music shopper could spend 8 hours per day—7 days per week for his or her entire life, from infancy to old age—listening to only the first half of each of 16 million songs, just 1 time.  That assignment would take 137 years.  This scenario demonstrates a mathematical impossibility—the 137 year old music shopper!

In reality, while the "great haystack" of available recordings may include 5, 10 or 15 plus million music tracks—most music shoppers probably want to identify and purchase from TWO to FIVE HUNDRED of their favorite songs.  Going forward, the challenge must necessarily be HOW to quickly identify a customer-relevant and MANAGEABLE music list.

We are confident in our assertion that, without an improved method of song identification, most currently deployed search systems cannot demonstrate hit-music affinity prediction that is engineered to be efficient, accurate, and consistently precise for a specific person.

The "advanced tools" now operational on most music search sites include the old standbys of collaborative filtering (if you like artist A, you might also like artist B), genre, artist and title specification (but the searcher must know what or who he is hoping to find), and keyword inspection (a broad, random parameter).  These commonly-used methods do not provide a mechanism that persistently connects a unique user with his or her configurable music profile; and therefore, they cannot equal the precise music search efficiency proposed by our patent-pending Codentity®  LLC system.

So, as innovative and exciting as Web-supplied music might be, closer inspection of the potential customer "search experience" reveals a perhaps imperfect implementation—for now.  Undoubtedly, enhancements are constantly being devised, but Codentity has disclosed a blueprint for a greatly improved music search strategy, and it is available today.   Yes, Codentity has imagined and outlined a way to bring the immediacy of "customer-specific" call numbers to music libraries.

The Way Forward with Precision

For the benefit of all music collectors, no matter what their generation, Codentity has envisioned a superior, customer-connected music search system:

SCALABLE SYSTEM AND METHOD FOR PREDICTING HIT MUSIC PREFERENCES FOR AN INDIVIDUAL;  United States Patent Application (document) number: 20100063975; by inventor:  Thomas J. Hayes

Our invention has many valuable aspects including a method for identifying the time period in a person's life when they would have been most likely to form strong music associations, and a system of creating a personal music profile that can be uniquely associated with a particular music customer.

The Primary Exposure Window Filter

Our invention discloses a method of designing a smart filter which includes a user-customized Primary Exposure Window (PEW).  The boundaries of the PEW could be determined by a time period calculated from information such as the user's birth year.

In one embodiment, but not necessarily each embodiment, a person might reasonably be presumed to have developed his or her well-formed music affinity associations between the ages of 12 and 34—a time window that approximates the start of the teen years, and ranges through young adulthood.

After a certain point in an individual's life, other considerations, such as marriage and family commitments, or career plans, will generally cause a person to spend less time immersed in music and other aspects of the popular culture.  For most persons, this 22 year "window" is the period when they would have been most likely to devote time and interest to absorbing music made prominent in the popular culture.  Accordingly, the PEW correlates with the period in their lives when the majority of users formed their natural affinities for "hit-music".

The first objective of the PEW component might be to identify, and give priority sorting to, a selection of music recordings that were released and made popular within the defined PEW.  Using this technique, highly-ranked PEW songs would appear near the top of the list for corresponding user query results, and songs that are not in the Primary Exposure Window would be naturally sorted so as to be listed following the PEW songs.

For example, a person born in the United States in 1955 might have a personal PEW beginning in 1967 and ending in 1989.  This person's Primary Exposure Window query type (i.e., search and filter formula) would give increased priority to songs released and made popular between the '67 and '89.  Starting with 1955 as the PEW year of birth, and factoring in filter controls for user genre voting—emphasizing POP and ROCK—music examples from this period include:

1967: "Light My Fire" by The Doors and "Happy Together" by The Turtles

1968: "Revolution" by The Beatles and "Susie Q" by Creedance Clearwater Revival

1969: "Lay Lady Lay" by Bob Dylan and "Sugar Sugar" by The Archies

1970: "Evil Ways" by Santanna and "Kentucky Rain" by Elvis Presley

1971: "American Pie" by Don Mclean and "Gypsies, Tramps, and Thieves" by Cher

1972: "Heart Of Gold" by Neil Young and "Take It Easy" by The Eagles

1973: "Crocodile Rock" by Elton John and "The Right Thing To Do" by Carly Simon

1974: "Come And Get Your Love" by Redbone and "Hooked On A Feeling" by Blue Swede

1975: ""Evil Woman" by ELO and "Sister Golden Hair" by America"

1976: "If You Leave Me Now" by Chicago and "More Than A Feeling" by Boston

1977: "Go Your Own Way" by Fleetwood Mac and "Night Moves" by Bob Seger and the Silver Bullet Band

1978: "Just The Way You Are" by Billy Joel and "Two Out Of Three Ain't Bad" by Meatloaf

1979: "Crazy Love" by Poco and "Here Comes My Girl" by Tom Petty and the Heartbreakers

1980: "Hit Me With Your Best Shot" by Pat Benetar and "Ride Like The Wind" by Christopher Cross

1981: "Hey Nineteen" by Steely Dan and "Take It On The Run" by REO Speedwagon

1982: "Don't Fight It" by Kenny Loggins with Steve Perry and "Hurt So Good" by John Cougar Mellencamp

1983: "Owner Of A Lonely Heart" by Yes and "Tell Her About It" by Billy Joel

1984: "Girls Just Want To Have Fun" by Cyndi Lauper and "Karma Chameleon" by Culture Club

1985: "Material Girl" by Madonna and "Smooth Operator" by Sade

1986: "Danger Zone" by Kenny Loggins and "Invisible Touch" by Genesis

1987: "I Heard A Rumor" by Bananarama and "With Or Without You" by U2

1988: "Man In The Mirror" by Michael Jackson and "Never Gonna Give You Up" by Rick Astley

NOTE:   The examples above were identified with a PEW based on the birth year of "1955", and placing additional priority on POP and ROCK genre songs.  If the genre voting filters were expanded to include other music types, such as SOUL and DISCO, the summary could include hundreds of additional hit songs within the time window; songs like "Let's Get It On" by Marvin Gaye (1973) and "Dancing Queen" by Abba (1977).

The PEW prediction based on 1955 as the user's year of birth is only one example.  The same methodology could be used to quickly determine a song shopping list for users born in 1965 or 1975 or 1985.  The important point to remember is that application of a Primary Exposure Window query filter can produce—in only one or two steps—a music summary that is highly likely to be favored by the user.

By contrast, developing a user-optimized song summary with most other commonly deployed search methods could take 10, 20, 30 or more steps—just to approximate the accuracy of a PEW-based query.  (In fact, with today's conventional methodology, it may NOT be possible to come close to the precision-crafted hit-music prediction capability of our Codentity® PEW design.)

Conclusion: A faster and more efficient music search and identification method will reduce "search fatigue".  This outcome will benefit BOTH the potential music search customer, by producing accurate results, while it also provides an advantage to participating music vendors.  A PEW-based query result will enhance the vendor's ability to bring extra value to shoppers, and thereby transform those search-satisfied shoppers into CUSTOMERS that initiate and complete digital music sales transactions.

[Read our Web essay "The Quest For The Ultimate Search Engine" here.]

Personal Music Profile

In order that everyone may benefit from faster, more relevant, more consistent search operations, shouldn’t someone build a demonstrably superior search product?  Could that superior product be based on a “Persistent and Unique User Profile” methodology?

In this posting, the term "persistent" is intended to mean a consistently reproducible process, while the term "unique" refers to an absolute identification method (such as that potentially embodied by the linking of a mathematically distinctive number with an individual’s true name, and possibly a corresponding user alias).  And, finally, the term "profile" indicates a compiled, stored data repository consisting of personal preference declarations and other biographical information which can be linked to the unique user—for recall on demand.

To elaborate further, the ultimate search engine should be able to:

Absolutely identify a unique user*

Accept and store that user’s biographical and demographic information*

Accept and store that user’s declared preferences on a broad range of tests*

*Offer user-controlled access permissions to said information and preferences as stored in a universally adaptable, cross-platform  profile

Provide pre-filtered search attribute values which are dynamically determined, in part, with the collaboration of said unique user

Provide true portability so the user may create, control and carry his PROFILE at will to any participating connected device or service

Would this methodology save time, and promote greater precision in search results?  We assert that, if properly implemented, it would produce immediate benefits.  And these methods are discussed in our disclosure.

Implementation Models

When effectively implemented with its associated business model, our patent-pending music search system (“the system”) can be expected to facilitate the capture of a larger share of the music search market for any participating music vendor.

Since music search execution is the crucial first step taken by each customer in the $8 billion dollar annual U.S. music market, the improvements envisioned by our system could provide a distinct advantage to licensees.

When the system is deployed, our company anticipates significant revenue (in the form of licensing and transaction fees) from motivated music sellers who wish to utilize the system to direct customers to their Web stores, where they can efficiently close music sales transactions.

The participation of interested parties as a development partners, or as investors, may lead to NEW revenue streams for such organizations.  Additionally, a Web vendor might choose to use the system’s design to augment its currently deployed search methodologies.

Certain aspects of the system also have the potential to act as a model for expansion into other retail search categories, and to possibly serve as an inspiration for improving the efficiency of ALL search products, including leading Web browsers.

Our disclosure envisions implementation via a Web services / ecommerce model that is likely to attract millions of potential customers who would benefit from its uniquely-customized search methods.  The system and methods in the invention collaborate with users to eliminate search fatigue because customers can intuitively locate desired content, based on calculations made from information stored in their “personal music profile.”

At the same time, music vendors profit from the system’s ability to efficiently direct user content focus.  This precision targeting means the user is more likely to accept presented selections (recommendations), and complete his purchase of recorded music products.

The system disclosed in the invention could be crafted as the interface that fundamentally determines customer behavior with respect to initial product discovery, and their subsequent desire to acquire specific selections.  The system might quickly achieve universal adoption because it benefits both parties to a potential “search and close” transaction.

By “universal adoption”, we mean that the system could be tailored to supplement the existing query methodologies of all leading music sellers (such as iTunes®, Wal-Mart®, Best Buy®, Rhapsody™, Yahoo® and Amazon®).  With the appropriate technical and marketing incentives, the system could appeal to a majority of their established customers—while it serves to invite new shoppers.

These objectives are admittedly ambitious.  However, we believe they can be achieved because our system overcomes observed deficiencies in at least 3 key areas of the current art: customer collaboration, music affinity calculation, and profile transportability.  Our system will help participating vendors deliver an improved customer experience.

Interested vendors, developers, and collaborators are encouraged to send a message to info@codentity.com.  Referenced trademarks are the property of their respective owners.

Learn how this technology may improve Web search here.

This informational essay is Copyright 2012 by Codentity® LLC member Thomas J. Hayes.  All rights are reserved.


Codentity® LLC is the developer, publisher, and exclusive licensor and distributor of the Music MegaBase® Catalog (MMBC) Media Edition—which includes the True Blend™ MPS.  FREE trial downloads will permit potential customers to "test drive" the software on their Microsoft® Windows® XP or Windows® 7 PCs.

Get an overview of the True Blend™ Media Player System here.  Learn more, including details about technical requirements, here.

NOTICE: All trademarks, including (where applicable) listed song titles and their corresponding artist names, are property of their respective owners.  The appearance of a song title or artist name on this page does not constitute an endorsement of Codentity designs, products or services by the named artists, or their publishers or material copyright owners.