The Audio File and The Audiophile: Understanding How The MP3 Has Impacted Music Production From 2010-2016

(1)

The Audio File and The Audiophile:

Understanding How The MP3 Has Impacted Music Production From

2010-2016

By

(2)

Abstract

The MP3 has become a ubiquitous format in the realms of digital music thanks, in part, to the ease with which it can be distributed and consumed. While a great deal of research has been conducted into the distribution and consumption practices involved with digital audio, particularly the MP3, very little research has been undertaken seeking to investigate how the MP3 audio file has affected processes of musical production. This thesis attempts to confront this shortcoming by conducting a content analysis of articles from four prominent UK based music magazines published between the years 2010 to 2016. By taking on this research I hope to discover who might be using MP3s to produce music, what are the purposes of MP3s in music production, and how are MP3s discussed in comparison to other audio formats.

(4)

Introduction

By using a content analysis of four UK based music magazines, this study attempts to discover how the MP3 audio file is being discussed in relation to the production of music. I regard the production of music to not only be a technological aspect of the process of recording, engineering, or performing of audio - but also musical form, compositional processes, song writing practice, geographically displaced collaboration, and the development of new styles of music.

While a great deal of research and study has been undertaken in regards to how consumption and distribution habits have changed in recent years due to the technological advances of compressed audio files and the Internet, very little research has been undertaken to see what effect this technology has had on the production of music.

Introduction To Technological Processes

MP3 Encoding

MP3 Definition & History

The MPEG 1 was a compressed format for both video and audio, and what is now the MP3 is the audio half of the MPEG 1 standard, or more specifically, it is layer 3 of MPEG 1 audio. The MPEG 1 sought to replace an earlier form of digital recording achieved via an Analogue-to-Digital (A-D) converter. This technique is more commonly referred to as Pulse Code Modulation (PCM), and was used for the encoding CD quality audio. PCM data is comprised of accurate measurements of audio waveforms’ amplitude at specific and regular intervals of time. There are two variables that affect this process: sampling frequency and word length (also referred to as bit depth). Sampling frequency is the amount of measurements that occur per second – as CD-quality audio consists of 44,100

measurements of a waveform’s amplitude per second, it has a sampling frequency of 44.1kHz. Each sample is formed as a series of 1s and 0s. If more 1s and 0s are used in the formation of a sample, then the measurement is more accurate. This is word length. As PCM word length consist of 16 bits, which could be either a 0 or a 1, “there are no less than 65,536 possible values” (216_{) for each}

measurement.

Bit depth should not be confused with bit rate, which is the measurement of the amount of data required to represent one second of audio. CD-quality stereo audio has: 16-bit word length multiplied

(5)

by a sample frequency of 44,100 multiplied by the two channels for a stereo signal, resulting in 1411 kilobits per second (kbps). PCM audio quality, therefore, is such that it “is practically

indistinguishable from its source” (Sellars, 2000). However, as PCM is an uncompressed format normally stored as WAV or AIFF files, its file size is relatively large. In the process of compressing any data file, information is removed. Compressed formats are commonly referred to as lossy such as MP3, AAC, MWA, and Ogg Vorbis; or lossless such as FLAC, Apple's ALAC and MP3 HD. Lossy formats employ “perceptual audio encoding” in the process of their compression, allowing massive reductions of files’ sizes, sometimes as low as 64kbps, but normally 128kbps or 256kbps, at the expense of audio information that would not normally be perceptible. Though there are many lossy audio file formats available and widely used, such as the ones listed above, none have proven to be quite as popular as the MP3.

Perceptual Audio Encoding

MP3 files are compressed via “perceptual audio encoding” in which the psychoacoustics of human hearing is exploited (Hepworth-Sawyer, 2014; Corbett, 2012). Assumptions are made about what audio may or may not be perceivable, and “the less-than-ideal conditions” (Diduck, 2012) in which the music being consumed can take place, considering busy areas, traffic sound pollution, and using poor hardware in order to determine what data can be removed (Corbett, 2012). It is impossible for listeners to perceive all the data being heard, and is therefore unnecessary to store and then reproduce all of the original data. In the process of deciding which elements are lost, various psychoacoustic principles are taken advantage of in different ways by different perceptual audio encoding algorithms. The human ear is most sensitive to frequencies between 1 and 5 kHz, and is substantially less

sensitive beyond the thresholds of 20Hz at the lowest, and 20kHz at the highest. These minimum audition thresholds become closer with age and exposure to damaging frequencies and volumes. As such by middle age few people can hear above 16kHz. Quieter audio data within frequency ranges to which we are less sensitive could therefore be considered imperceptible and therefore encoding them would not be necessary. Additionally, perception of frequency content is done in narrow divisions of bandwidth within the 20Hz to 20kHz range. If two sounds occur simultaneously within the same division, referred to as a “critical band” (Corbett, 2012; Hepworth-Sawyer, 2014), then the louder sound will mask the quieter one rendering it potentially imperceptible, and therefore could plausibly be erased in the encoding process.

Much like the masking of sounds in the same critical band, quiet sonic events can be masked by loud sonic events that occur within a small interval of time. We can seemingly ignore quiet sounds, even if the quieter event precedes the louder one (Corbett, 2012; Hepworth-Sawyer, 2014).

(6)

Audio is split into sections called windows, which are then analysed. The size of the window is determined by its surroundings, due to the necessity to cross-fade windows after they have been analysed. If content of two adjacent windows is similar the windows could be longer as the cross-fade is less noticeable. If adjacent content is less similar, windows could be shorter due to the cross-fades being more noticeable. While shorter windows increase accuracy of the encoding in terms of timing, it is at the expense of spectral resolution, and vice versa for longer windows. Though the encoding process splits the range of frequencies into “around 32 bands” (REF) – similar to, but not exactly, the critical bands. Content within these frequency bands, and within certain sections of time is analysed to ascertain what content is or is not perceivable. The algorithm then negotiates between audio content and bit-rate, depending on the required file size. Content that can be masked by other events or is below the minimum audition threshold is therefore not encoded (Corbett, 2012).

Lossy formats will always display some quality loss, whether this loss is perceivable, however, is debatable. The lower the quality of audio, the easier it is to hear the manifestations of overly compressed audio files. These manifestations, these side effects, are referred to as sonic artefacts.

Sonic Artefacts

Considering uncompressed CD-quality audio files are coded at 1411kbps, this reduction of over 90 percent of the audio data, without it being overtly noticeable, is an impressive achievement. This negotiation of audio quality for bit-rate, though perhaps not a concern for most listeners, can still result in many audible side effects.

As stated earlier, high-frequency content above 16kHz is usually removed when encoding to MP3 of 128kbps due to it likely being imperceptible. Though these frequencies may be imperceptible, they can still impact other frequencies, and therefore timbres may be affected. As compression relies on the listener’s inability to perceive sonic events that can be obscured by other content, a certain amount of by-product noise can be generated before we are capable of hearing it. Some levels of fluctuating noise can be heard, as musical content is replaced by it (Corbett, 2012).

Encoded audio content can also become “time-smeared” when critical bands are processed. The result is that a sound signal can occur before or after the wanted sonic event, resulting in pre- and post-echoes. If these “smeared” sounds occur beyond the thresholds of perception, then they become audible. This can result in the attack and the release of a sound being longer, softening its character. If this effects perception of a sound in which a fast attack is necessary, such as the sound of a snare drum, then that would undo much time, effort, and money that went into its recording.

Time-smearing can also effect other instruments, such as vocals, creating the impression of a double-track. Reverb tails can also lose detail and texture; they can become less smooth, grainier, and more

(7)

transparent. Many details of sounds can be lost resulting in unwanted timbres, making sounds “coarse and hollow, missing body, depth” (Corbett, 2012).

Just as frequencies at the high end of the spectrum are cut, so those at the bass end are edited. Low frequencies are harder to be analysed because their long durations. Analysis windows often do not contain an entire cycle of a low frequency, sometimes “less than half a cycle of frequencies bellow 114Hz” (Corbett, 2012). Therefore analyses are inconsistent, resulting a loss of bass solidity.

The dynamic range of sounds can also become reduced in the encoding process. However, due to the removal of frequency content our perceptions of amplitudes of remaining content can be altered. Content can appear to be louder simply because content that surroundings it has been reduced. Sounds’ timbres can be affected by phase shifts, but so can one’s perception of dynamics. “When frequency content is smeared over time, its average level over time increases slightly. This can result in the listener perceiving it as louder”. Another common artefact that can occur is referred to as swirlies. These are “rapid coming and going of lower-level frequency content”. Swirlies sound unstable, and sibilance can sound harsh, detached, and less clear. There is a general loss of definition and clarity (Corbett, 2012).

As many consumers download music files from unregulated sources, there is little control over the quality of the file. Some files may have been converted from one format to another more than once, bringing about even greater losses in quality. These sonic-artefacts may be found not only in personal listening, but also in professional and public contexts, such as DJs using digital systems, karaoke bars, and Internet streamed broadcasts (Corbett, 2012).

Various MP3 Sizes

Though much of what is consumed is unregulated, what quality we can be sure of is in regards to legal downloading and streaming services. Audio files and their kilobit usage vary between streaming and downloading services. Amazon and iTunes are arguably the largest downloading services

available both of which offer MP3s and AACs, respectively at 256kbps – though in the past both used to only offer files at 128kbps. There is more variation among streaming services, depending on the device being used. Services tend to supply audio files with lower bit rates for mobile devices and higher bit rates for “desktop use” i.e. computers – but even these “higher” bit rates are quite low. Pandora, for example, streams 64kbps AAC files for desktop use. Spotify's standard free service uses the Ogg Vorbis format at 160kbps for computers and 96kbps for mobile devices, while their Premium subscription service can stream at 320kbps. Slacker streams MP3s at 128kbps for computers, and can go as low as 40kbps and 64kbps AACs for its mobile formats. Youtube is also a streaming service worth considering, though quality is inconsistent, at its very lowest quality YouTube audio was

(8)

encoded as mono 22.05kHz 64kbps MP3 files. This is quite a substantial drop compared to CD-quality audio of 44.1 kHz 1411kbps audio files (Corbett, 2012).

As consumers listen to music on a vast array of different devices, from MP3 players and phones to high-quality recording studios, data compressed audio files must sound as good as possible on as many devices as possible (MusicTech, 2016). This quality is achieved when mixing and mastering, but also when the audio file is being encoded. Therefore new technologies have been developed to aid in this process helping producers, both professional and amateur, to negotiate between

data-compression and audio quality.

Mixing and Mastering for MP3

In sound recording once all of the individual parts of a multi-track recording have been recorded and edited, they are then mixed together. The mixing process attempts to create a cohesive and consistent sound that blends the various recorded tracks together. This is achieved by using different processes affecting dynamics, frequency colour, stereo positioning; as well as effects such as reverb, delay, and distortion, among many others. After the mixing process is complete, the track is mastered for

reproduction of the recording on various formats.

The purpose of the MP3 format has shifted from being something used for demos to one that is now not just a viable a final format, but a preferable one (BPI, 2015). When considering which type of file to export audio as from a Digital Audio Workstation (DAW) project, MP3s were considered useful for rough mixes, which could be easily emailed as a demo “or an outline to someone” (Jones, 2010). Additionally, exporting as MP3 was good as it “obviate[d] the step between exporting … and then using something like iTunes to convert to MP3 format”(Jones, 2010). Ultimately, however, the MP3 was considered to have a limited range of uses since “you wouldn’t want to work with compressed files” (Jones, 2010). This attitude appears to have changed. Tracks are being mixed and mastered, and old records are being re-mixed and re-mastered for MP3, suiting the new listening contexts in which music is consumed.

Mixing

The ubiquity of the MP3, a resurging popularity in vinyl and cassettes, and the development of hi-fidelity digital downloading and streaming music services show how music producers must consider many different mediums of distribution and consumption when mixing and mastering. The scope of media fidelity has grown to the point that now recordings must sound “great within the confines of your own studio … on your expensive monitors, they should also sound good through the headphones

(9)

of your phone, compressed down to MP3; they should also sound great on the CD player in your car; and they should also sound great on your mate’s hi-fi” (MusicTech, 2016)

Mastering and Re-Mastering

When a shift in music distribution technology occurs it inevitably necessitates revision of the

techniques and technology used in the process of mastering tracks. Music that was once mastered for vinyl had to be remastered for “CD due to the very different qualities of the two mediums”.

The shift of medium popularity from CD to audio file, particularly MP3 and AAC, was not as drastic as vinyl to CD, but there is a shift nonetheless (MusicTech, 2012). The Red Hot Chili Peppers’ 2011 album I’m With You was one of the first albums to be marketed as “mastered for iTunes”

(MusicTech, 2012). Mastering engineer Vlado Meller therefore had the responsibility of mastering the album for vinyl, CD, and also produced an AAC master from the original mix rather than simply compressing the CD master. This process, of mastering and mixing tracks specifically for lossy audio files such as the MP3 has been made easier by the development of new technologies.

DAW Plug-ins and apps

Audio data compression has become such an important process in music production that producers no longer have to render their mixes and masters, before rendering a data-compressed file and importing and exporting audio files between DAWs for comparison. Compressed audio data files, particularly MP3s have become the most popular form of music consumption, and as such, their compression, without perceptible loss, is paramount (Senior, 2011).

Data compression is typically a process that occurs after mixing or during the mastering of a recording. New software plug-ins for DAWs have been developed in order to give producers the ability to data-compress while still mid-recording session or in the mixing process, giving the

producer a greater understanding of how the track will sound once compressed without the necessity of mixing, mastering, and exporting before data-compressing.

One such plug-in is the Fraunhofer Pro-Codec developed by Sonnox and Fraunhofer IIS, the latter being the research institute where the MP3 was first developed. It is a real-time data-compression plug-in, providing the producer the ability to compress data while still within an audio project of a DAW “streamlining this tediously iterative quality-control exercise” (Senior, 2011). Producers can encode for mono, stereo, and 5.1 surround sound; in real-time and also offline; and can also decode audio files, giving the producer a wide variety of options to easily toggle between. Mixes can therefore be auditioned through MP3 and AAC codecs in real time, allowing producers to make

(10)

adjustments and hear their effect on the encoding process instantly. This therefore gives producers the ability to tailor mixes and masters to sound better in compressed forms (Hepworth-Sawyer, 2014; MusicTech, 2012).

Though the Fraunhofer Pro-Codec is a mastering plug-in that can be used on various DAWs, the Ozone 7, made by iZotope, can be used as both a plug-in and as a standalone application. It gives a variety of mastering options, including processing modules based on “vintage analogue hardware”, and a bank of presets allowing producer to “achieve authentic sounds for any genre and style”. More importantly, however, the advanced version of the software includes a “Codec Preview” feature, which allows producers to audition the effect of various “industry standard” codecs upon their master, such as MP3 and AAC. Like the Fraunhofer this gives producers the ability to make adjustments to the mix and master before exporting the track (2015, Sound on Sound).

Plug-ins such as the Fraunhofer Pro-Codec and the Ozone 7 illustrate the popularity and importance of the MP3 by the fact that they are designed for, and used by, both professionals and amateurs alike. Aimed at a “wide-open” market, amateurs are able to monitor the data compression process allowing them to achieve a more preferable balance between audio fidelity and file size. On the other hand, mixing and mastering engineers stand to learn from what the plug-in reveals about the audio they are compressing – as well as saving them time and effort (Senior, 2011).

The Fraunhofer Pro-Codec plug-in was first developed in 2011, more recently, however apps and web services have been developed to aid in the production of audio files. MasteringBOX is a “free” app (or “online mastering utility”) (Future Music 2015) that allows producers to master MP3s online. It claims to have a “sophisticated algorithm that can detect the dynamic and spectral characteristics of the track being mastered and apply appropriate adjustments” (ibid).

The app is free to producers who only wish to master MP3s at 320kbps. The process is relatively simple: a producer uploads a WAV or AIFF file onto the MasteringBOX’s website or app wherein the file is processed via their “sophisticated algorithm” (ibid) for mastering and data-compression. Though the service does not achieve “what a good standalone mastering application can accomplish” (ibid), it shows again how non-professionals are able to achieve better audio results with

data-compressed files. The fact that producers are able to master tracks – once considered a highly

specialised process – on their phones, shows how much consumers and producers favour convenience over quality.

Hardware

It is not only software that has been developed to fit the needs of producers of MP3s. As DJing using MP3s has become more prominent, so new pieces of hardware have emerged. The Numark NDX900

(11)

DJ controller allows DJs to mix from “virtually any digital source including a computer, CDs, MP3 CDs, or even USB flash drives” (Future Music, 2011). It should be noted here that the term mix in the context of DJing is different from mix in the context of producing music in a multi-track recording. A producer may change different elements of a recording, such as volume, stereo placing, equalisation, and other effects and processes in order to create a coherent overall sound of a recording. A DJ, however, will play series of completed songs, attempting to merge their beginnings and endings to create a consistent sound – though they also can effect and process elements of the track being played, the NDX900, for example gives the DJ the ability to use filter, echo, flanger, and phaser effects (Future Music, 2011).

Prior Research

The effects of the MP3 on the music industry have been researched and written about a great deal over the past decade or so. While some have considered how the MP3 has affected music production, most studies have looked into how the MP3 has affected distribution and consumption. I have given outlines of these studies below and considered what issues they have drawn in the development

Individualisation

Michael Bull has for many years been researching mobile communication technologies; music and sound in urban culture; and has applied critical theories particularly form The Frankfurt School to his findings (University of Sussex, 2016). Over the past decade he has paid particular attention to the effect of MP3 players, conducting empirical studies into how it has changed consumption and listening practices. Bull argues that MP3 players give users unprecedented control over their experience of time and space by managing their mood and orientation to space through the micro-managing of personalised music. He uses some concepts and terminology established by Adorno to describe processes of mediating social interactions with music (Bull, 2005).

Privately listening to music is argued by Bull as being a form of “mediation” – attempting to establish a degree of power over one’s spatial and temporal surroundings (Bull, 2005; Bull, 2013, p.495). According to Bull the listener attempts to achieve autonomy over his surroundings in response to the “repressive environment” in which he lives. This “repressive environment” is described by Adorno as the “realm of the eversame” (Adorno, 1976, in Bull, 2005, p.344) and by Walter Benjamin as “ever-always-the-same” (Benjamin, 1973, in Bull, 2005, p.344). The MP3 player gives the listener vast choice and allows him withdraw from the routine of the situation, the realm of the eversame, and much more accurately control their “thoughts, feelings and observations” (Bull, 2005, p.344). The MP3 player therefore also acts as an archive, made possible thanks to its large storage capacity, allowing the listener to carry his whole music collection with him (Bull 2005; Bull, 2009, p.87).

(12)

Bull often posits that mobile listening experiences prioritize the individual over the social, similarly Heike Weber states that mobile musics have “helped to foster and maintain a sense of emotional and cultural identity, which appears to have drifted from a group to a more individual basis” (Heike Weber). Sonia Livingstone states that the home is “increasingly becom[ing] the site for individualised media consumption with children spending the majority of their home media use alone in their bedrooms” (Livingstone 2002 in Bull, 2005, p.345).

What autonomy the listener might gain from using music to mediate his space is not necessarily synonymous with individualism. Shuhei Hosokawa posits that listeners are not detached but rather “unified in the autonomous and singular moment” (Hosokawa, 1984, p.170).

Socialisation

David Beer, who has also been researching mobile musicking devices, music scenes, and digital sociology for many years (University of York, 2016) criticises the MP3 players individualisation, and contests Bull’s argument that MP3 players mediate and screen out the listener’s environment. He instead argues that a more layered sonic experience is true. Pervasive city sounds interact with mobile music to create new imbricated social-sonic experiences. Beer considers the popularity of the iPod and MP3 players in general to have had a transformative effect on how consumers collect and archive music, and considers its mobility as a tool for listeners to transcend boundaries between public and private zones (Beer, 2007; Beer, 2008).

David Beer and Raphael Nowak also consider the use of algorithms by streaming services and iTunes to extend mobile music’s more sociable perspective. Music services use algorithms to gather and compare users’ listening habits allowing the musical recommendations to be made (Nowak, 2016, p.5). Beer states that this creates new social listening experiences, rather than the individualised ones posited by Bull. Nowak reconciles these perspectives by considering the iPod as a hybrid object enabling both positions to be valid. As technologies and media are developed, individual and social musical experiences become increasingly merged. Previous cultural practices are extended, allowing users to relate to music, content, and other listeners in new ways (Nowak, 2016).

Experiential Peripherals

Heidi Gerber has also considered how the MP3 has changed listening practices and other elements to consuming music, which she refers to as “experiential peripherals” (Gerber, 2011). Gerber asks whether these contextual elements, such as “coolness factor (brand status), ease of use and ease of music acquisition, psychological mood benefits, and the control of personal space” (ibid), which are

(13)

significant and influential factors in regard to MP3 technology, have been impacted by record production practices.

Multi-Channel Consumption

Between the 9th_{and 18}th_{of November 2015, the British Phonographic Industry (BPI) conducted an}

online survey of 1,000 music consumers via audience.net. The intention of the research was to ascertain through which channels music is consumed, focusing particularly on “multi-channel” music consumers (BPI, 2015).

The report found that:

“Two thirds of the total sample listened via streaming (free or paid) and either CD, vinyl or

downloads. The most common multi-channel scenarios were free streaming and CD listening (49%) and free streaming and downloads (44%). The age group most likely to be multi-channel listeners was the group aged 35-54 (80%). Many of these people are listening via free streaming and CD. The Millennial generation (aged 16-34) were the most likely to be in the paid streaming multi-channel groups and the free streaming and download group” (BPI, 2015, p.2)

Music consumption has now diversified, and listeners use multiple channels to listen to music. They stream and download MP3s as well as buying; collecting; and gifting physical formats such as CD and vinyl. This study clearly shows the importance of digital formats in the music industry.

Additionally, the IFPI Global Music Report from April 2016 has detailed that revenues of digital formats had overtaken physical ones for the first time, and that digital sales contribute 45% over physicals 39% to total revenue of the music industry in the USA (BPI, 2015).

The Effect on the Industry

There are various studies that have been undertaken looking into how the MP3 and the Internet have affected the wider music industry. In several chapters of his PhD, Ian Michael Dobie gave various descriptions and explanations concerning the impact of technology upon the music industry. He considered production, consumption, and distribution, as well as marketing, copyright law, piracy and counter piracy techniques. He gives descriptions of how technology has shifted power dynamics between consumers, producers, and distributors throughout the 20th_{century giving historical context,}

and very early 21st_{century, focusing particularly the impact of the Internet and MP3 between the}

(14)

Ashraf El Gamal considered the wide range of implications the Internet has had on the music industry in the 21st_{century. He assesses the rise of legal digital distribution, economic implications, general}

welfare effects, and changes in consumer preference and social phenomena in regard to recorded and live music (El Gamel, 2012).

Tracy Redhead looks at various ways in which digital music formats have impacted music. Though she considers issues inherent in consumption of music to be important, she argues that a research approach based on participation and interaction with digital formats is more appropriate, rather than simply considering it from a passive consumer perspective. By considering how digital formats have influenced artistic innovation and consumer participation, Redhead illustrates their impact upon collaboration, interactivity, audio manipulation, and the fostering of new musical cultures and processes - such as remixing.

Cyborg

Bruce Darlington has looked at various functions of music as a means of establishing power in certain contexts, such as muzak music in factories as a tool to increase work output, and over-spilling sound from MP3 player headphones creating primary (active) and secondary (passive) listening experiences. In the essay Train Listening: Music in the Spaces of the 21st_{Century Darlington introduces new ideas}

and concepts relating to interactions between listeners and technology. Rather than listeners

consuming music for personal reasons posited by Bull or Beer, they are creating new listening spaces and “experiencing modern life in the mode that technology has made the norm… machines contain far more information about people than is generally realised, and when consciousness is extended or distributed globally via social interaction over the internet such machines become even more potent extensions of the subject” (Lupton, 2001, Stone, 2000, In Bruce Darlington, 2013, p.255). Darlington begins to draw on theories posited by Donna Haraway particularly the Cyborg Manifesto in which human and non-human interactions create new spaces and paradigm shifts. He relates Haraway’s ideas to human-music interactions: “music breaching the outer skin of the human subject via the extension of human subjectivity beyond the boundary of the flesh, through technology” (Darlington, 2013, p.255).

Collaboration

Alvaro Barbosa has looked into how the Internet has led to new collaborative methods of musical composition and improvisation between geographically displaced producers. Drawing on the technological advantages and disadvantages of streaming MP3s globally between producers. (Barbosa, 2003)

(15)

Rolf Grossman’s article The tip of the iceberg: laptop music and the information-technological transformation of music gives an overview of musical transformations have occurred in the wake of technological developments. Grossman shows in this article provides many examples of how sound files and the Internet are used to create new musical styles within the realm of laptop music, such as glitch, click techno, microhouse, and others (Grossman, 2008).

John Shiga has also written on how the MP3 and easily accessible audio editing software has led to a rise in “mash-up culture” a music scene in which practitioners “splice and combine pop songs

encoded in MP3 format to produce hybrid recordings” (Shiga, 2007, p.93). He goes on to explain how the Internet has also had a big impact upon the growth of this style (Shiga, 2007).

Form

More recently, Jonathan Sharkhovskoy and Rob Toulson have undertaken an empirical study to understand the impact of the MP3 upon popular music forms – considering, in particular how the definition of an album has changed. The study outlines how consumption practices have shifted from being album oriented to listeners downloading individual digital files. This has had an impact upon the wider music industry and artists, whom still consider the album to be the primary form of music production, distribution, and consumption. New “album apps”, however, seek to “reverse these trends and become the definitive delivery format for music”. These apps would, according to the study, benefit the music industry and artists, allowing them to regain control, credit, and capital for their work, while also re-establishing the album as an important musical form (Sharhovskoy & Toulson, 2015).

Question

From this research I hope to gather information regarding how compressed audio files have affected the creation of music itself. Who is using the MP3 in music production? Is it older or younger generations; amateurs or professionals; mainstream or fringe music? What are MP3s used for? Is it only used for demo tracks; does it aid song writing; does it also have pedagogic purposes? How is the MP3 judged in relation to other formats? Does its lower fidelity overshadow what positives could be gained from its practicality; how are its specific audio characteristics discussed; does it devalue the experience one has with music?

(16)

A classic content analysis of popular music magazines was performed in order to discover what impact the MP3 audio file has had upon the production of music. I used this form of analysis due to it being well used within media and mass communications research over the past seventy years, and has become the “fastest growing technique over the past 20 years” (Macnamara, Neuendorf, 2002, p.1). I used a deductive approach in order to establish categories of analysis from research that has been conducted in areas of distribution and consumption of the MP3 audio file. I used these categories as a starting point, after which I applied them to my content and proceeded to revise where necessary by amending, deleting, and adding categories.

I considered it to be expected that discourse on MP3 distribution and consumption was similar to MP3 production – especially when considering the increasingly blurred boundaries of these three areas since the development of the Internet, for example, live streaming services allow one to

produce, distribute, and consume content simultaneously. Additionally considering how the MP3 has changed discussion on consumption practices – for example by extending mobile musicking concepts and practices to give listeners greater degrees of autonomy over their context or allow their context to effect the listening experience ecologically – there may also have been similar shifts in how music production is discussed. As such I wanted to confront the content I have been analysing with the prior research in order to consider how they may interact, uncovering both similarities and disparities between the two areas of research.

History of Content Analysis

Though first introduced by Harold Laswell as a systematic method to study propaganda in mass media, the process has gone through various changes. Early usage of the process was focused on quantitative analysis, defined by Berelson as a “research technique for objective, systematic and quantitative description of the manifest content of communication” (Macnamara, 2005, p.2). This definition has been critiqued and revised due to its claim of objectivity and its restriction manifest content, ignoring latent content, which can also be analysed (ibid). Mostly, however, this early definition has been criticised because of its concentration on quantitative factors and the assumption that they alone indicated social impact (ibid).

More recent definitions describe the process as “any research technique for making inferences by systematically and objectively identifying specified characteristics within text” (Stone, Dunphy, Smith & Ogilvie, 1996, with credit given to Holsti, p. 5 in Macnamara, 2005, p.2). Additionally, Neuman (1997) considers content analysis to be an important non-intrusive research methodology “for gathering and analysing the content of text” (ibid).

(17)

Kimberly Neuendorf subscribes to a strict quantitative approach. She defines content analysis as “a summarizing, quantitative analysis of messages that relies on the scientific method ... and is not limited as to the types of variables that may be measured or the context in which the messages are created or presented” (ibid). Neuendorf advocates the use of scientific methods focusing on the importance of a study’s “objectivity-intersubjectivity, a priori design, reliability, validity, generalisability, replicability, and hypothesis testing” (Macnamara, 2005, p.3). She considers qualitative methods to be more appropriately described as “rhetorical analysis, narrative analysis, discourse analysis, structuralist or semiotic analysis, interpretive analysis or critical analysis” (ibid). Though Neuendorf makes this differentiation, she also posits an “integrative” model of content analysis, in which many qualitative methods can be appropriate “with only minor adjustments” (ibid). Both quantitative and qualitative approaches have benefits and faults. Generally, quantitative content analysis conforms to a “scientific method and produce[s] reliable findings”, whereas qualitative analysis is “maybe impossible” to do with scientific reliability – Krippendorff once described the process of coding content qualitatively as being “an art. Little is written about it” (Krippendorff, 1980, p.76). Qualitative analysis, however, is recognised as being necessary “to understand deeper meanings and likely interpretations by audiences – surely the ultimate goal of analysing media content” (Macnamara, 2005, p.5). As such, recent formulations of content analysis methodologies have sought to negotiate these two approaches and reconcile their weaknesses and strengths. This has not led to “a new methodology” but rather reconstructs “different steps of analysis with their different logics” (Mayring, 2014, p.8).

Deductive Approach

I attempted to use the deductive approach laid out by Neuendorf and expanded upon by Macnamara, while also using some additional explanations given by Mayring. As stated earlier by Neuendorf, there are certain steps that must be undertaken in order for the method to be “scientific”, those being: objectivity-intersubjectivity, a priori design, reliability, validity, generalisability, replicability, and hypothesis testing (Macnamara, 2005, p.14). Jim Macnamara, professor of Public Communication at University of Technology Sydney, seeks to combine the deductive and scientific steps proposed by Neuendorf, with qualitative analysis by, for example, analysing “positive and negative words and phrases … to identify the tone of text” (Macnamara, 2005, p.14).

While I attempted to subscribe to Neuendorf’s definition of a scientifically rigorous approach I found it difficult to follow some aspects concerned with reliability, due to a lack of experience and an inability to use two or more coders. The employment of inter-coder reliability was researched by Riffe and Freitag (Macnamara, 2005, p.10) in studies between 1971 to 1995 published in Journalism and Mass Communication Quarterly, where it was found that only “56% of these reported inter-coder reliability” (ibid). More recently, Lombard, Synder-Duch and Bracken found that only 69%

(18)

“discussed inter-coder reliability” (ibid). Though the lack of inter-coder information is a widespread issue, its absence in my research is my fault. I am aware of it and understand its importance in these studies. Acknowledging this flaw, and recognising changing attitudes towards methodology, if I were to conduct this research again I would address this issue, and do my utmost to employ a more reliable procedure.

Method for choosing sources

To ascertain a high degree of inter-subjectivity I attempted to make my sample is as representative as I could. Though my research topic was relatively niche, I could not consider all possible media outlets that might contain relevant content due to the vast amount of music and technology magazines that are published. Therefore I took a purposive method, selecting material from key music media sources within the UK. Macnamara also states that a purposive method is most appropriate for focusing on relevant media (Macnamara, 2005, p.14).

I considered content from 2010 to present day due to a development in UK consumption of digital music. Though combined digital and physical album sales declined by 7% overall in volume from 2009 to 2010, digital singles were the highest on record, and digital album sales were proficiently high enough to consider it to have entered into the “mainstream market” (BPI, 2011). The 2010 BPI report describes the digital album as “com[ing] of age” in that year, as it represented 17.5% of album sales (ibid). It also described the growing digital single sales as “no fluke” (ibid). The increased popularity of legal digital media reflects the fact that consumers choice for recorded music had “never been greater – both in depth of catalogue and the many ways to buy it” (BPI, 2011). As legal

distribution and consumption of digital media has become increasingly popular, and consumers have greater choice of medium, I was interested to see how might this effect or reflect music production.

The Magazines

I used four magazines that represented music production and consumption discourse in two distinct areas: technological and cultural. I wanted to do this in order to collect a broader scope of content, hopefully allowing me to see a wider impact of the MP3 upon music discourse. I chose four magazines that are based in the UK. By keeping my corpus selection to a single nationality and culture, it increases the probability that the language being used is standardized, and lowers the risk of various dialects or linguistic variances being used, which could have complex multilayered

meanings. The magazines I used were Sound On Sound, Music Tech, Fact, and The Wire. Sound On Sound and Music Tech were concerned with the technological processes of music production

particularly recording and engineering. Fact and The Wire tended to focus on the cultural impacts of music.

(19)

Sound on Sound is a monthly music recording technology magazine, which has been in print since 1985. The magazine’s website explains that it acts as an archive housing the magazines “highly informative articles that chart the impact of technology on the music production and recording processes”. The magazine supplies readers with product tests and reviews, interviews, tutorials, and advice. Sound On Sound boasts having “the most experienced editorial team to grace any music magazines staff list”, and attempts to appeal to a variety of readers. The magazine’s content is intended to help reader’s “whatever [their] level of expertise”, and according to its own website is “the 'bible' of the music recording industry”. I think that its experienced writing staff and broad readership would render it to be both a detailed and also comprehensive source of information regarding music technology production (Sound on Sound, 2016).

Music Tech magazine was founded in 2002 and is also concerned with music technology production, “dedicated to the latest trends, techniques, gear, and software within the world of music production”. It’s physical publication and the website are integrated so that there is no difference in content

between either. The magazine is also written by “a panel of dedicated experts working in the fields of recording, mixing, and mastering across every major DAW”. Music Tech provides reviews of

hardware and software, technique features, and “round-ups of the best gear, what ever your budget”. As with Sound On Sound I consider this magazine to provide content that is concerned specifically with the technological aspects of music production, while also appealing to a broad readership (MusicTech, 2016).

Fact was founded in 2003 as a “bi-monthly music and youth culture” magazine. By 2009 the physical print of the magazine was discontinued and the publication became exclusively online. The magazine covers topics concerned with popular culture and popular music with contributors to the magazine consist of various musicians and producers such as Damon Albarn, Andrew Weatherall, and St Vincent. The magazine describes itself as being concerned with music and youth culture and its contributors are well-established musicians, this leads me to conclude that its intended readership is young people who listen to popular music. The magazine’s focus on mainstream culture and its contributors being established figures within popular music and mainstream culture give opportunity for its content to reflect trends in music production from the perspective of songwriters, performers, and cultural critics (Fact, 2016).

In contrast to Fact, The Wire concerns itself with “a wide range of alternative, underground and non-mainstream musics”. It describes the “musics” it is concerned with as “avant rock, electronica, hiphop, new jazz, modern composition, traditional musics and beyond”. The magazine has a small full time team, and a rosta of “over 60 freelance writers”, as well as having contributors including Simon Reynolds, David Toop, and Robert Fripp. These contributors are, like the contributors to Fact, established musicians and/or music writers. The difference being that they belong to academic

(20)

available. As with all of the other magazines I am considering, The Wire boasts a large readership, in Europe, the US, and Japan. The Wire’s interest in non-mainstream music, and its rosta of academic writers and musicians from the music it is writing about allow it to cover areas that the other

magazines I am considering might not. I hope that its appeal to a large readership is indicative of the magazine’s consideration of broader musical trends and cultural criticisms within its areas of concern (The Wire, 2016).

Sound On Sound and The Wire were both established in the 1980s (Sound on Sound 2016; The Wire, 2016), whereas Fact and Music Tech were established in the early 2000s (FACT, 2016; MusicTech, 2016). I am interested to see whether the ages of the magazines has any affect on how the MP3 is discussed. Perhaps the older magazines are predisposed to a more purist stance? Or perhaps as they have seen greater changes in musical consumption since their inception they tend to take a

descriptivist approach to contemporary shifts in production trends. Sound On Sound and Music Tech appealed to audiences that were both amateur and professional producers, made clear by both

magazines statements that they advise and write for readers with varying levels of expertise and budget. I hope that this will be apparent in the discourse and will depict how the MP3 has influenced professional and amateur areas of music production. Fact magazine is aimed at a younger readership and its content is focused upon popular culture, whereas The Wire is focused on non-mainstream musics. I hope to see how these two magazines reflect trends in both popular and alternative music, and whether there are differences or similarities between them.

Unit of Measurement

I used the online versions of these magazines rather than the hardcopies. I did this to make finding articles easier, and because content is not dissimilar between published and online editions.

Additionally Fact is an online-only magazine, making online use a necessity in that case. I used the online searching tools of the magazine websites to find articles that include the term MP3. I made a general searched for the term MP3 because I wanted to generate results that would include not only content concerned directly with music production, but also distribution and consumption. I wanted to include articles concerned with distribution and consumption to ascertain what, if any, effect they have had upon music production. I narrowed my search to articles published since 2010 because of consumption trends, I discussed earlier, that have been reported in the BPI.

Clipping and Coding

Having established my sources, I manually read, “clipped”, and downloaded relevant items. Though it is true that online editions often do not contain all content that may be in the printed edition of the magazines, I made sure that in this case the magazines websites stated that the online content and published editions were identical.

(21)

A Priori Design

In order to keep the categorization process more “science” (Macnamara, 2005, p.3) than “art” (Krippendorff, 1980, p.76), I have considered the prior research and used it to establish a list of issues, rather than inductively adding issues and messages that might be found in the articles as I code the content (Macnamara, 2005, p.8). Khun argues, however, that deductive processes can be self-limiting and do not “foster innovation” (Macnamara, 2005, p.9). Nonetheless this “dichotomy can be overcome” through the process of exploring the “literature of critical scholars” (Neuendorf 2002 in Macnamara, 2005, p.9). By undertaking preliminary readings I was able to establish what issues there might be within this topic in general, and therefore what should be considered when coding and analysing content. As stated earlier, I then confronted new content with these preliminary categories in order to better understand how the areas of distribution and consumption may influence music production.

Definition of Preliminary Categories

In my attempt to establish the categories that I used for the coding of my research I read and ascertained issues and messages that were prevalent in research that already exists on the impact of the MP3 upon music. Because very little research has been undertaken in regard to the MP3’s effect on the production of music, I have included research that considers the MP3’s impact on consumption and distribution practices as well. Arguably the ease with which MP3s are produced, distributed, and consumed has led to a blurring or simultaneity to occur between these three phenomena, making what issues that occur in listening practices perhaps just as relevant in production practices.

As Neuendorf suggests, I immersed myself “in the world of the message pool” (Neuendorf in Macnamara, 2005, p.9). I considered the texts I have mentioned in the prior research section. As the texts move from being listener oriented to producer oriented, so the themes, issues, and messages shift from being concerned with listening practices to production ones, however overlapping issues can also occur within texts.

Preliminary Categories

(22)

The first key issue was that of cultural prominence and ubiquity of the MP3. This issue has been discussed explicitly by Beer, Bull, and Sterne as well as researched in depth in the BPI report (BPI, 2011; BPI, 2015; Beer, 2007; Beer, 2008; Bull, 2005; Sterne, 2016).

Listening Practices ---Isolated

The second group of issues I have recognised are concerned with new listening practices and new listening contexts. Beer, Bull, Weber, Livingstone, Hosokawa, Gerber, and Darlington All discuss the effect the MP3 is having in creating isolated listening experiences, giving the consumer greater control over mood management, and regaining some autonomy over their environment (Beer, 2007; Beer, 2008; Darlington, 2013; Gerber, 2011; Hosokawa, 1984)

---Socialised

In contrast Beer, Bull, Weber, Hosokawa, Nowak, and Darlington also consider the MP3’s socialising effect, bringing listeners together, and also creating a cultural identity.

---Aural Ecology

Lastly in this group is the issue of the new listening contexts brought about by the mobility of MP3 players. These new contexts create new experiences with the “aural ecology of the city” (Beer, 2007). Other Interactions

---Collecting Archiving

The third section is concerned with other ways of interacting with or using the MP3. The size of the MP3 makes it easy to establish a collection or to archive and retrieve (Beer; Randall, 2016, p.127; Roessner, 2016, p.62).

---Non-physicality

The change in format from physical to non-physical has also changed tactile and aesthetic experiences consumers may have with the format as well as extra musical hardware including branded headphones (Beer, Gerber and sharkhovskoy & Toulson).

---Cybord

Additionally, Beer and Darlington have raised the issue of how humans and technology more closely interact, considering Haraway’s Cyborg Manifesto (2006) and “networked bodies” (Beer) as other forms of interaction with MP3s and MP3 players.

The Internet ---File Sharing

The second and third groups are mostly concerned with interaction from a consumer perspective, however Beer, Barbosa, Grossman, Redhead, Gerber, sharkhovskoy & Toulson, Dobie, and El Gamal have considered how the MP3 in conjunction with the Internet has effected music production

constructively and positively. Collaboration between artists has become easier as the MP3 file is easily transferred around the globe. The MP3 and the Internet have also impacted the broader music

(23)

industry negatively. Gerber, sharkhovskoy and Toulson, Dobie, El Gamal, Redhead all raise issues concerned with the ease of acquisition, the legality of file sharing, and the subsequent loss of capital for artists and labels.

---Apps & Websites

Redhead and Sharkhovskey & Toulson have also looked into how online Apps are being used to both produce and distribute music.

Technical Processes

---Reduced Quality & Encoding Technicalities

The fifth group is concerned with more technical processes and issues such as the reduced quality of the sound of the MP3 and the technicalities of the MP3 encoding process (Beer, and sharkhovskoy & Toulson).

---Recording & Editing

Grossman, Shiga, and Redhead also consider the ease with which producers can edit and manipulate MP3s now, thanks also to easily acquired DAW software.

Trial run and revisions of categories

After establishing preliminary categories by extracting themes and issues from the prior research literature, I began a “trial run-through” (Mayring, 2014, p.95), referred to as “pilot coding” by Neuendorf (Macnamara, 2005, p.12) to make initial codings. Extracts were taken from the material to check whether the categories are at all applicable and to establish anchor examples (Ibid). I could then revise the appropriateness of my categories.

I resolved issues discovered in the trial run, and redefined these categories in a way as described by Mayring in his 7 step deductive analysis model illustrated below in figure 1. After revising the categories, I used a coding list to categorise data more precisely (Neuendorf).

Revisions and new categories

Though various listening practices were important subjects in much of the prior research, the material I am now researching considers similar listening practices in an indirect way. There are many articles that discuss the technical processes of audio production. At times these articles discuss these

processes in terms of catering to various listening practices and contexts. For example Bull and Beer discuss the importance of the MP3 in terms of isolating and/or socialising listeners within urban and domestic environments. The article 10 Ways To Mix A Track also underlines the importance of giving new experiences to listeners within new contexts and new devices. It does this by advising producers

(24)

“make [their] mixes sound good on any device, including MP3 players and tablets”. Additionally the article The Final Mixdown – Part Three says, “It can be useful … when checking your mix on other playback systems, such as on a hi-fi, MP3 players, or in the car”. I therefore deleted the categories concerned with listening practices and placed articles concerned with catering to listening practices into the category Encoding for Different Formats .

Additionally, many critics discuss the MP3 in terms of providing authentic musical experiences, usually in comparison to other formats such as PCM audio or vinyl. For example in the article Keeping Up With A Jones: Part 4 – Feeling Queasy it is stated: “I’ve had good monitors in the past, but have now become used to headphones so forgotten the joy of listening to proper music on proper speakers. And it’s proper music too, as in CDs not MP3s. We’ve all probably become used to MP3s a little too much and the combination of using CDs today and listening to them on proper speakers is raising the hairs on my arms. This is what listening to and making music is all about”. I therefore created a new category to cater to articles that discuss authentic listening experiences called Overall Experience.

Often in regards to listening practices, the prior research would discuss the MP3 audio file

synonymously with the MP3 player. This meant that the MP3 player was not discussed directly, or independently enough for me to consider establishing a category for it. After considering the content for my research, however, it is clear that the topic of discussion for some magazine articles is that of the MP3 player. The article From Start To Finish: Paul White’s Leader briefly discusses the

importance of the MP3 player as changing musical form due to the option to randomise song order, and therefore adds to the demise of the album over the individual MP3 audio files. In several articles the term “MP3 player” is used as a modifier or euphemism to denote poor quality. Consider the article Mix rescue: Motor Tapes “It sounded good through cheap generic MP3 player headphones”. I therefore thought it was necessary to add a category that catered to discussion of the MP3 player explicitly.

Another theme I had discerned from the prior research was the MP3’s increasing popularity and becoming a standard format for consumption. I wondered whether this would be reflected in the content I was analysing. After confronting the categories deduced from prior research with my content, however, I discovered that this issue was not addressed directly. The issue was discussed implicitly, as either the cause or effect of the articles’ topic at hand. It could be seen, for example in articles that discuss software and hardware that can import and support MP3 usage. The article Seven Wonders states “Reason [a DAW] now supports many more audio import formats, including MP3”, and the article IK Multimedia Amplitude 3: Amp and Effects Modelling Software states, “you can import WAVs, MP3s and other popular audio file formats”. I therefore deleted this category and placed content into categories that were concerned with the manifest issues such as Recording & Editing.

(25)

I changed the categories labelled “file sharing (positive)” and “file sharing (negative)” to more general titles of “the impact of the internet (positive)” and “the impact of the internet (negative)”. These new categories give a wide enough scope to include the issue of file sharing, as well as other themes I encountered when looking at the content such as downloading, emailing, social media, and blogs. These ways of interacting with music via the Internet may seem varied, but they have similar impacts. The consistent positive impact of the Internet is upon collaboration, whereas the negative points discussed are concerned with how the music industry has lost control over distribution and therefore capital.

I kept the category concerned with the discussion of applications separate due to it not being necessarily positive or negative, and not necessarily reliant upon the Internet. Instead discussion tended to be descriptive of new ways that applications are used by producers to sample and effect form. For example the article Play the MusicMappr sampling app states “MusicMappr is an app that splits up any mp3 or Soundcloud link into samples, then runs spectrographic analysis on the samples, and clusters them together into similar-sounding groups. The results are displayed in a playable interface that runs in a browser window”.

(26)

Macnamara states that an “a priori design is operationalised in a Coding System” (Mac p.9). Macnamara gives examples of Coding Systems, such as a Coding Book and Coding List. I also considered using Mayring’s Coding Guideline (Mayring, 2014, p.97) because it appeared to be an easily formulated table for use in deductive coding. Ultimately, however, I employed a coding list endorsed by Neuendorf that categorizes issues simply, first depicting domains – larger areas in which categories sit – and then categories themselves, with anchor examples beneath (Neuendorf, 2016). For Coding List see Appendix.

(27)

Results and Analysis

The study yielded 147 articles that used the term “MP3” at least once. These occurrences broken down across the separate magazines are: Sound on Sound had 63, Music Tech had 44, The Wire had 27, and Fact Magazine had 13. The years that the articles were published in ranged from 2010 to June of 2016. 2010 had 27 articles, 2011 had 15, 2012 had 21, 2013 had 29, 2014 had 17, 2015 had 22, and 2016 has so far had 16 articles published in these magazines.

Macnamara’s methodology proposes a mixture of various methodological steps in order to garner the scientific strength of deductive processes as well as the depth of qualitative analysis (Macnamara, 2005). By using prior research to establish categories deductively and confronting the content with these categories, I was able to keep my research embedded within pre-existing work while also drawing out new themes and issues that are pertinent to the new topic. Thus my final categories were produced both deductively and then partially inductively as they were revised. This has led to

interesting interactions and developments in the general area of research concerned with the impacts of the MP3, as well as new areas to be considered specifically within music production. I have approached the results and analysis in the order of the categories and themes on the Coding List.

Recording & Editing

Importing and MP3 Support

There was a lot of discussion concerning producers’ ability to record and edit MP3s in DAW software. Much of what was written was advisory and almost exclusively was within the magazines MusicTech and Sound on Sound.

The first point of discussion in the larger issue of editing MP3s in DAWs and recording them with both software and hardware was whether the technology being discussed was able to import and support MP3 files. I coded 13 articles that discuss this issue, 7 from Music Tech and 6 from Sound on Sound. Of these articles; 1 was published in 2010, 2 in 2012, 5 in 2013, 1 in 2014, and 4 in 2015. As was expected, magazines that were concerned with technological processes of music production published more articles discussing this issue. It is noteworthy that discussion surrounding this issue, generally, increased over the years.

The article IK Multimedia Amplitude 3: Amp & Effects Modelling Software, published in 2010, discussed usage of the MP3 as being for “knocking out demos”, and described it as a “popular audio file format” (White, 2010). Indicating that white considers MP3’s value to lay in its practicality rather than its audio quality. The two articles from 2012 concerned with this issue discuss the ease with

(28)

which producers are able to import and edit MP3s. For example, the article Apple Notes: Great Little Apps states, “Audio Hijack Pro outputs MP3 … so its easy to drop recordings it makes into your DAW” (Bigwood, 11/2012). Contextualise quotes better., but again The article Apple Notes: Audio Editor roundup acknowledge there is still an issue of lower audio quality stating, “[The DAW] can edit MP3… without having to transcode first and re-encode them afterwards, and thereby avoids any cumulative quality loss” (Bigwood, 12/2012).

Articles published in 2013 discuss increasing technological support for the MP3 - though there is still a sense that it is a novelty. Bigwood’s statement in Sonic Safari: Reason Tips and Techniques that “version 7 is more compatible than ever before, accepting MP3” (Bigwood, 2013) seems to illustrate that the MP3 was not a standard format at that time. Walden, however, does not just state that MP3 support is happening, but rather that it is “improving” (Walden, 2013) showing that MP3 support on software and hardware was perhaps becoming more widely used. The single article I gathered from 2014 Reason Tutorial: Become a Power User Part 4 – Audio Editing discusses the possibility of importing files into the DAW Reason, but only mentions the MP3 by name, referring to other

compressed audio files simply as “other compressed formats” (MusicTech, 1/2014). This could be an indicator of the MP3 was becoming more popular format for music production.

I gathered five articles from 2015 discussing this issue. The article Cakewalk Announce SONAR Cambridge Membership from Music Tech Magazine points out that MP3 import/export support has “improved” in the DAW Cakewalk (MusicTech, 4/2015), and the article The 6 Best Freeware DJ Tools goes as far as to describe the MP3 as “standard” (MusicTech, 10.2015).

There is a clear shift in how discussion over hardware and software import and support for MP3 audio files over these five years. A key change between 2010 and 2012 was from focusing on

whether a producer might use MP3s purely for “demo” (White, 2010) purposes, to it being discussed in terms of a unit of construction that can be imported and edited in DAWs. This could be an

indicator of a shift in general attitude toward the MP3 – moving from usage of it for demos due to its lower audio quality, to being an integrated tool for music production. The terms used to describe the MP3 in the more recent articles also suggest that it has developed from being a format that has been on the periphery of music production to being standardised.

It is interesting to note that there were no articles that appeared from 2016, resisting the apparent trend of increasing articles discussing this issue each year. However, this could be explained by the fact that these articles were gathered in May and June of 2016, therefore resulting in a smaller

window of time for articles to be published and gathered for this research. Nonetheless I posit that the increased discussion of the MP3 within this subject shows that the format has become a favoured medium among digital music producers.

(29)

Song Writing

There were three articles that discussed editing and recording with regards to song writing process, from 2012, 2014, and 2016.

The article from 2012 Noah ‘40’ Shebib: Recording Drake’s ‘Headlines’ discussed recording processes used when making the song Headlines by Drake. In particular, the article considered how different parts of the song were being written and edited simultaneously using MP3s. The MP3s were initially used as reference tracks and then became individual tracks within a recording project as the writing and recording progressed. As the article stated, “Drake was writing to the stereo MP3 … I then swapped those for the MP3 tracks, so by the time Drake had finished writing, he was working against the separated files” (Tingen, 2012).

The sentiment inferred earlier that MP3 usage was for making “demos” (White, 2010), is reiterated in the 2014 article Inside Track: Arcade Fire’s Reflektor – indicated by the use of the phrase “quick mix” (Tingen, 2014). In the article, producer Mark Lawson stated that after making a “quick mix … people have something to take home with them, which will be on memory sticks as WAV files or as MP3s” (Tingen, 2014). What differs here, however, is that both WAVs – a standard professional lossless audio file – and MP3s are both considered as options for a producer to create a demo, perhaps showing a growing acceptability toward MP3s as a standard professional format.

It appears that the advice that surrounds song writing processes, recording processes, and MP3s, has not changed a great deal between 2010 and 2016. The MP3 is still discussed primarily in terms of its functionality as a medium to review works-in-progress. The article published in 2016 Songwriting 2016: Introduction To Songwriting Using A DAW says “often, we’ve found it’s useful to bounce down a rough mix of your demo to MP3 and then go for a walk, listening back to the song on your headphones” (MusicTech, 7/6/2016).

What these articles illustrate, however, is that the MP3 has used by some professional producers while making records that went on to be commercially successful, while also advising and encouraging amateur producers to use this format.

The MP3 Becoming Standardised

Discussion concerning the MP3’s shift from a fringe format to being more accepted can also be seen in regards to default recording modes on MP3 recorders. A review of the DR-100 MKII recorder in Music Tech Magazine explains that the “default recording mode” is in MP3 format and that in order to record in WAV “you need to power down the device and restart, selecting an HD mode”

The Audio File and The Audiophile: Understanding How The MP3 Has Impacted Music Production From 2010-2016