Harder Better Faster Stronger; The Digitalisation of Music

As we know, digitalisation has changed the concept of social interactions and communication. Media has become social, being accessible to many, as well as a place to which many can add. Within this digital landscape, music has flourished from live chamber music, to readily available streaming, to auditory memes and a form of communicating and connecting.

This essay will explore the different ways in which digitalisation changed music, and its social relevance. I will delve into four categories that I call Harder, Better, Faster, and Stronger, after the 2001 song by Daft Punk. In order, these will guide us through the music itself, accessibility, categorisation, and sampling. 

Harder – Music itself

To understand music in the digital world, we need to first understand what it means for sounds to be “digital”. Nowadays we understand the Digital as the world that we enter when we open our phones, computers, or other social electronic devices, but the reason this is all possible is through systematic processing of information. The world is a dynamic, and uncertain, fluid place which is not easily compressed into a rapidly reproducible documentation. The fluid and infinite reality is what we call the analogue, which we can document in for example analogue photographs, or in the case of music vinyl records, which will play the smooth waves when played on a record player, following the grooves in the material. Though this type of media is relatively big and stationary, as well as more susceptible to damage. Structuralizing the media would make it more easily documented in other forms, and could be saved on electronic devices. The audio waves are divided into steps, with different frequencies having their own level. The electronic device can understand the different steps of the sound waves which all together will approximate the full analogue wave. The steps are numbered with a binary system in which every 1 or 0 is called a bit. The more bits used to describe a level or a “byte”, the more variation there can be, and the more it will approach an analogue wave. 

fig.1) Graham Mitchell, CD Audio Sampling.

In 1980, companies Sony and Phillips set the standard for the formal qualities of the “Compact Disc” which is now commonly known as CD. After some rivalry and discussion, the two companies established a 16-bit system for the Digital-to-Audio conversion of these binary codes.1 This means that there are 2^16 different combinations, giving 65 636 different levels within the possible volume. 16 ones in a byte indicate the highest wave, while 16 zeros indicate no amplitude, and therefore no sound. The different tones come from the frequency of the wave, which are measured through taking samples over time. Waves in music are highly complex, and constantly change, so to get a representation of analogue sound, we have to take as many samples as possible, and link them to the closest level and code of one of those 65 636 levels. For a CD, the frequency of sampling is 44 100 times per second, so that the maximum frequency is 22kHz, just beyond the maximum human hearing ability.2 Later, when information could be stored even smaller, MP3 players and mobile phones became the new standard to listen to music. More about that later.

So the digital sound waves are not smooth like analogue waves, but instead a kind of blocked representation, in- and decreasing like a set of stairs. What does this do to the perceived sound of the music? Well, not that much actually. To the average listener the general digital sound sounds just as smooth as analogue music, the same way that current day movies look as smooth as analogue visuals, even though those are a bunch of still frames put together. 

It does, however, change how we make music. Over the course of a couple of decades instruments became electronically processed and easier to manipulate while playing, to becoming multifaceted and having different programmed sounds and instruments linked to the same hardware, to being freely accessible and editable on a computer(, to being generated by AI). 

With all these new tools sounds are played with in all kinds of ways. An example is Daft Punk’s use of a vocoder to change the vocal sound to a rough robotic sound. A vocoder is originally a device to analyze sounds to encrypt messages, but in music it is used connected to a synthesizer. The input of the synthesizer gets combined with the vocal input to create new sounds. The use of synthesizers in general increased a lot in the 1980’s, likely because the first digital synthesizers were produced in the 70’s. According to Dean Wallraff, developer of the dmx-1000 Signal Processing Computer, the digital synthesizers were way more flexible and had more variety in what sounds it could produce. Wallraff notes “It was the most flexible real-time synthesizer you could buy at the time, and it allowed composers to do things they couldn’t do with any other affordable system.”3 Later, the more affordable Yamaha synthesizers took over the market, making the use of synth sounds more available to the public. Fast forward to the current day and age, the use of synthesized sound has become ingrained in our music culture. Almost all mainstream songs use it, and many instruments we hear are adapted digitally. The sounds made are incredibly diverse, with umbrella-genres like house, techno, trance, and hardcore.4 I will talk more about genres later, but this illustrates the change of sounds that the digital synthesizers have enabled.

“Harder” here, indicates the technically rougher waves, as they are generated in steps rather than smoothly by analogue means. Most people don’t hear this in the sound itself, and generally it doesn’t affect the experience that much. “Harder” can also mean that the sounds that are now possible are rougher. This is not true for all genres, but there are definitely some very weird electronic twists made possible by digitalisation. Some genres lean into the electronic sound, and some seem to test the limits of sound waves as they create the most chaotic pieces. They make sounds that are only producible with digital instruments, and are in that way also “harder” to reproduce analogously, establishing the digital as a unique new tool for innovative sound.

Better – Accessibility

I just talked about the sound waves which don’t change much for the experience of music, but influence the ways we make, document, and spread music quite heavily. Because the music can be digitally stored, not needing the analogue waves, but a string of binary code, distributing the music to the public is much easier. I just talked about Compact Discs, made to hold digital wave-representations rather than analogue ones. They are an early example of digitally stored music and created the biggest economic boom in the music industry so far.5

Personal computers at the time did not have hard drives big enough for even one song, so the CD was a revolutionary discovery (pun intended), with about 70 times more data stored than a computer hard drive. It is conveniently compact with more than an hour of play time on a flat disc of only 12 cm, can be listened to on the go with a car radio with CD-player or a walkman, and is cheaper to produce in larger quantities and varieties than vinyl. Generally, CD’s are also more durable, and don’t distort with playback, like vinyls do.6 But of course, CD’s are still a physical item, and take up space when not inside a player. Especially when you want to listen to multiple different artists and albums, and have to take a whole box of different CD’s with you. Already in 1982 part of this problem was being solved. Frank Zappa distributed a multitude of songs over cable, and people who paid a monthly fee would be able to record any of those on tape as they wished.7 There have been attempts to distribute audio files over the internet, but it took longer than 90 minutes to download only one song. 

To solve this problem, and have audio files take up less storage space on computers, algorithms were designed to compress files. The files that were adapted by these algorithms had a compressed code, which also meant that the quality of the audio would have to be given up a little bit. This process is called “lossy”.8 The eventual product of the research into lossy algorithms led to the MP3 format. In the early 90’s, the Motion Picture Experts Group (MPEG) worked on the MPEG-1 Layer 3 algorithm, which soon became publicly known as MP3.9 In 1993 it became a standard for audio downloads. MP3 did approximate “lossless” audio, which means that it came close to a compression format without the loss of quality. In 1999 a database called Napster was launched. This database was filled with files that were illegally uploaded by its users.10 At the height of Napsters popularity the network counted around 80 million users. Due to the illegal methods used, the network was forced to shut down in 2001, but it had already set the scene of open source music.11

In the following years, new applications started to emerge, providing a similar kind of database. Napster got rebranded into a legal PressPlay music store by media company Roxio in 2002. Apple ITunes, the music store that accompanied the IPod, Apple’s famous MP3-player, launched in 2003. Pandora in 2005, with similar algorithms, but more intended as a reinvention of radio, had a “freemium” set-up, where users can choose between a free version with ads, or a subscription version without ads, similar to current-day Spotify.12 Spotify itself started in 2008, combining aspects from its predecessors, and providing a wide range of music to its users. The more they listen, the better recommendations Spotify provides. The most used audio formats for these platforms in the early days were AAC and MP3, until in 2014 streaming platform TIDAL was able to offer music in the lossless FLAC format. Spotify nowadays uses a couple of different formats, but the base file in the process is FLAC.13

All these developments bring better audio quality to more people. “Better” here does not only mean audio quality, but most definitely the better distribution. Spotify currently has over half a billion active world-wide users,14 who all can fetch a nice song for any occasion, make playlists of their favorite songs, and discover new music. Although there are issues with the current, consumer-central system, such as the low monetary credit artists get for their music, the direct access to millions of songs has become a part of our daily lives. 

Faster – Categorization and Analytics

In the previous chapter I mentioned Spotify’s recommendation system, and the user-interaction in streaming, making and discovering new music and new playlists. This is a key point in Spotify’s model, and is often referred to as “the algorithm”, the same way we refer to the determination of interests and recommendations on other social media platforms. These algorithms used to be primarily developed to aid the user in finding their way in the chaos of a platform with so many different kinds of music. Titles of songs often don’t communicate the musical genre of the song, as content and genre are not inherently linked, plus the addition that many song titles are quite vague and poetic. There are trends within genres, of course, but when searching through such a big database, it wouldn’t be an effective way of finding what you like. So to navigate this chaos, the algorithm analyzes your preferences and filters all the songs for you. The music gets structured for you to consume.

Nick Seaver notices a shift in this paradigm. Where the makers of algorithms first aimed to help the user filter the chaos of the database, now the companies aim to “[entice] people into becoming users.”15 The algorithms are meant to capture the audience, and keep them listening. Seaver explores this in a story about a hypothetical streaming service Willow. “When users skip songs or turn Willow off , this is a sign that something has gone wrong.”16 The algorithm analyzes what you like; what makes you tick, and feeds you more of that to trap you into staying on the app. Of course, this sounds very evil and grim, but Seaver writes that these algorithmic traps might be a natural way of interacting with the world. The algorithms are more human than we usually understand them to be. Or in words that make more sense, we are more like algorithms than we think. “But the shape of a trap reflects a process of learning: torn nets, empty baskets, and abandoned playlists suggest changes in design. Traps are simultaneously devices for acting in and learning about the world.”17 With this, Seaver illustrates that the algorithm is not just about presenting a trap to capture the user and rope them in like a predator capturing its prey, but function as an interactive experience in which the user learns as well. The algorithm, of course, does not want to kill its user, but form a bond with them. 

One way that algorithms get to know its users is through genres. Genres are nothing new, but the way in which artists are assigned to genres has. I had mentioned before that the digitalisation of music caused a boom in the variety of music, and that many new sounds emerged. This adds greatly to the list of genres we know of, but is not the only cause of named genres. In The Pudding’s interactive essay, Matt Daniels and Michelle McGhee explain how the algorithm determines genres, through analyzing users’ listening habits approximating other listeners and their habits.18 They note that genres used to be very difficult to determine, and that the categorisation usually began when it had already grown beyond that state, but in the digital age and the age of algorithms this changes. By proximity, artists and genres are mapped and in turn recommended to those users who might be interested in them, according to the algorithm. With data up until the 9th of august 2023 the essay lists the amount of genres to be over 6000, many of which have been added in the last 7 years. The music has become so nuanced, and the algorithms are the tools that are used to reflect these nuances. 

So with the streaming services it is easier to link user’s listening habits to understand different movements and genres. I explained how Seaver understands the algorithm to work both ways, and in the last couple of years we can recognise that in the interest in the collected data. There are dozens of online applications that can generate a nice overview or listening habits. We of course know Spotify’s own yearly Wrapped, which dominates the social media every first few days of december. But the users don’t want to wait for December to see their juicy stats. Other analytical programs have emerged such as Receiptify, Spotify Pie Chart, or Stats.fm.19 For these applications, you can allow the site to access your spotify, and the program will read the info that spotify has stored. Some are more simple and just give a top 10 of your most listened songs or artists, while others are more silly, such as Icebergify, Obscurify, or How Bad Is Your Spotify, which judges your music taste. Their social media presence shows great interest in these kinds of data.

Besides these displays of data, the internet also creates other kinds of algorithms that filter the giant field of music. Examples of these are Forgotify, which filters songs based on having no streams, or “a clock where the time is a song title”, which searches songs with the current local time in its name.

“Faster” indicated the incredibly fast processing speed of data, and the near immediate categorisation of music. Initially it seems like a trap to lure the user into the service, but we see that it has a big interactive essence, and that users also like to see their own statistics and tastes. Besides that, the algorithms make finding music based on certain specifics way easier, and can be used in silly ways, but also in ways that benefit unknown artists, and broaden horizons of users.

Stronger – Sampling and media

This chapter is named after the Kanye West song Stronger, which is a remix of the original Daft Punk song. Remixing, sampling, featuring, and covering are a direct effect of the digitalisation. Probably the most controversial way of collaborating with other artists is the concept of “sampling”. I talked about sampling earlier, when I discussed the measurement of a wave, and this is a little similar, but with a small portion of another song. Instead of using it to give a representation of the original as with the waves, this kind of sampling uses the small portion to create something new. Sampling itself is not a new thing, but when the synthesizers and other technology evolved and became commercially available, sampling songs became much more accessible. Within the hip-hop and electronic fields, sampling has had a great influence in the creation of sub-genres.20 Although artists have always been referencing and taking things from other artists, it is seen as something controversial. It brushes against the line of plagiarism, and even with very short samples this can lead to massive copyright claims. An article in the American Business Law Journal explains the difficulties around sampling within the copyright laws.21 Copyright was initially intended as encouragement for new content, as well as legal protection of the work of artists, and this would list sampling as infringement on this concept. On the other hand, people consider a sample to be “fair use”, and therefore accepted. This is not a paper about legal considerations of course, but the article does note the most important aspect in judging sampling cases to be the market effect. The paper concludes that, in many cases, the original song goes up in sales when referenced by another artist, which makes sampling despite its legal difficulties a strong asset in its success.

We also observe a trend of full remixes and covers. This is similar to the case of sampling, but instead of taking a small part and using it in new material, a song in its entirety is taken and adapted to fit the artist’s style, and reinvent the song. There often are networks of bands and artists who will remix each other’s songs and form a little society within their discography. As an artist you need people who will like your music to find and know your name, so to share the audience artists will collaborate with and remix songs of artists that fit their brand. This type of interaction with other artists’ work is more often done with artists that are of similar popularity, and have a more mutual essence than the sampling of short parts of a song.

Besides the musicians themselves collaborating and taking inspiration from other artists, the general public has also gained influence in the music industry. I am talking about the “aural turn” of social media, which Bojana S. Radovanović explains to be the shift that took place during the COVID-19 pandemic, during which social media platform TikTok emerged and rapidly grew in popularity.22 It is a platform on which no static posts, which is the main feature of platforms like Instagram and Facebook, are posted. Rather it revolves around short videos, usually of around 10 seconds. As it can have audio, it is a regular occurrence that songs are added as background noise, or as a front sound to which a short dance is performed. Memes that usually were static images combined with a short text to create a funny or relatable, are now moving images with sound, gestures, expressions, and sometimes also text. Radovanović’s main focus is on the change in how music is created and distributed. The videos on the platform are short, and can give the perfect snippet of a song. The songs that have a certain line that works well as a meme format get re-used in the videos of others. 

A lot of older songs gained a big audience of younger generations because of this, such as Kate Bush’ Running Up That Hill. The example of The Kiffness and his Alugalug Cat Symphony, is a great example of the new way of creating music. The Kiffness, a South-African musician who often makes songs from funny sounding animals, created a cong called “Alugalug Cat” in may 2021. This became a big hit on social media, and other people joined in, together creating an “International Symphonic Mashup” of this song.23 Here, 7 different people, and a cat, from all over the world work together to make music. Of course this is a silly example of people wanting to make something fun on a social media platform, but as indicated earlier collaborations happen across the entire field.

“Stronger” is a reference to the remix of the song I used to format this paper, and in itself becomes entangled with the social network of artists. This network, also featuring the listeners as source of feedback, and occasionally original creative input, becomes stronger as our inter-connectivity becomes stronger and more entangled. The general public also gets a stronger voice, as they have new tools which give them way more power in what will be popular, and even in what is made.

Conclusion

The music became Harder, rougher, and more difficult to produce analogously. The audience has Better access to the music as the digitalisation of sound allowed music distribution to evolve into streaming platforms. Thanks to these platforms, the categorisation and analytics of music are produced Faster with the help of algorithms, and lastly the social networks between artists, and the audience have become Stronger. I hope this paper has sketched an image of the different ways in which digitalisation has progressed the music industry.

  1.  Howie Singer, Key Changes: The Ten Times Technology Transformed the Music Industry, Oxford Scholarship Online (New York, NY: Oxford University Press, 2023), Chapter 7. https://doi.org/10.1093/oso/9780197656891.001.0001. ↩︎
  2. Singer, Ch.7. ↩︎
  3. ‘DMX-1000 Signal Processing Computer. Dean Wallraff, USA 1978’, 120 Years of Electronic Music (blog), 23 August 2014, https://120years.net/dmx-1000-signal-processing-computer-dean-wallraff-usa-1978/.
    ↩︎
  4.  Guide to Electronic Music Genres, 2015, https://www.youtube.com/watch?v=B4r0MdBQI6U. ↩︎
  5. Singer, Key Changes, Ch.7.
    ↩︎
  6. Stewart Adam, ‘The Myths and Reality of Vinyl Records vs. CDs.’, Creative Audio Works (blog), 31 January 2021, https://creativeaudioworks.com/audio-perception/the-myths-and-reality-of-vinyl-records-vs-cds/.
    ↩︎
  7. Singer, Key Changes, Ch.8. ↩︎
  8. ‘A History of Music Streaming’, accessed 21 February 2024, https://dynaudio.com/magazine/2018/may/a-history-of-music-streaming. ↩︎
  9.  Jonathan Sterne, MP3 the Meaning of a Format, SST: Sign, Storage, Transmission (Durham, N.C: Duke University Press, 2012), Ch.5. ↩︎
  10.  ‘A History of Music Streaming’. ↩︎
  11.  ‘A Short History of Napster’, Lifewire, accessed 21 February 2024, https://www.lifewire.com/history-of-napster-2438592. ↩︎
  12. ‘A History of Music Streaming’; ‘What Is Pandora Music? History, Plans & Features of the US Streaming Service’, 23 July 2022, https://promusicianhub.com/what-is-pandora-music/. ↩︎
  13.  ‘Audio File Formats for Spotify’, Spotify, accessed 21 February 2024, https://support.spotify.com/us/artists/article/audio-file-formats/. ↩︎
  14.  Rohit Shewale, ‘Spotify Stats for 2024 (Users, Artists, & Revenue)’, DemandSage (blog), 15 February 2024, https://www.demandsage.com/spotify-stats/. ↩︎
  15.  Nick Seaver, Computing Taste: Algorithms and the Makers of Music Recommendation, 1st ed. (Chicago: University of Chicago Press, 2022), 51. ↩︎
  16.  Seaver, 50.
    ↩︎
  17.  Seaver, 68.
    ↩︎
  18. Matt Daniels and Michelle McGhee, ‘You Should Look at This Chart about Music Genres’, The Pudding, accessed 21 February 2024, https://pudding.cool/2023/10/genre.
    ↩︎
  19.  For more of these examples, check out: Elena Cavender and Elizabeth de Luna, ‘13 Best Websites to Analyze Your Spotify Data’, Mashable, 7 November 2023, https://mashable.com/article/best-websites-to-analyze-your-spotify-data. ↩︎
  20.  Mason Youngblood, ‘Cultural Transmission Modes of Music Sampling Traditions Remain Stable despite Delocalization in the Digital Age’, PloS One 14, no. 2 (2019): 2, https://doi.org/10.1371/journal.pone.0211860. ↩︎
  21.  W Michael Schuster, David M Mitchell, and Ken W Brown, ‘Sampling Increases Music Sales: An Empirical Copyright Study: Sampling Increases Music Sales’, American Business Law Journal 56 (2019): 177–229, https://doi.org/10.1111/ablj.12137. ↩︎
  22. Bojana S. Radovanović, ‘TikTok and Sound: Changing the Ways of Creating, Promoting, Distributing and Listening to Music’, INSAM Journal of Contemporary Music, Art and Technology, no. 9 (2022): 51–73. ↩︎
  23.  The Kiffness – Alugalug Cat (International Symphonic Mashup), 2021, https://www.youtube.com/watch?v=S61ENc51Z1Q. ↩︎

Tags:

February 22, 2024 by Media Maestros 0

You may also like...

Leave a Reply

This site uses Akismet to reduce spam. Learn how your comment data is processed.