A simple take on video streaming

The article talks about how Spotify CEO, Daniel Ek, is “Eating Google’s Lunch,” both economically and literally in Google’s restaurant, and “Loving it.” I thought it was interesting that some small company is actually giving Google a run for its money in an Internet business. The “Jack of All Trade” business model that Google follows […]

The article talks about how Spotify CEO, Daniel Ek, is “Eating Google’s Lunch,” both economically and literally in Google’s restaurant, and “Loving it.” I thought it was interesting that some small company is actually giving Google a run for its money in an Internet business. The “Jack of All Trade” business model that Google follows is obviously taking a hit here.
Some “distance” into today’s original article, I realized that readers may actually not understand all this video-streaming stuff. So, I decided to apply the brakes, back-pedal, and change direction, to talk instead, of some intro stuff on audio/video streaming.
The technical details of digital sound “recording” and playback, although quite interesting, are somewhat complicated and are usually beyond the purview of all but mere psychoacoustic mortals. However you dice it, music or sound starts and ends with the way that pressure fluctuations (over a base reference level) vary with time. The amount of pressure fluctuation is sometimes expressed in what is known as the sound pressure level (SPL). Thus, sound/music data can be stored in SPL as a function of time.
The default SPL-versus-time data is obviously analog, not digital, since, in digital form, it would have to be stored in zeroes and ones, as any other data that is processed through the computer. Data stored in 0’s and 1’s are said to be in binary form. Because sound is produced by pressure fluctuations and is also heard as pressure fluctuations, there is no advantage to digital technology at the beginning and the end of the process of recording and playback. So, the advantage of digital technology can be found in the efficiency of storage of the music data and in robustness. This storage efficiency comes in quite handy when we move sound through telecommunication media (fibre optics, waves, etc.). The smaller the size of the data, the faster it can be moved across telecom media, and hence the more modest the bandwidth requirements for transporting the data. Needless to say that digital data can be stored on different types of media, such as (digital) audio tape.
To be able to pass sound (music) through digital equipment (computers, smartphones, etc.), the sound must first be converted from analog (pressure-time) data to binary (pressure-time) data, using a software that is sometimes called a ripper. The psychoacoustic cut-off limits of hearing are used here to reduce the size of the stored, converted data. These limits are known, and are usually expressed in terms of the frequencies below which humans cannot hear sound and the frequencies above which sound damages your ear drums and becomes hazardous to your health.
The more stuff you remove from the original sound data the smaller the size of the resulting data and the faster it can be moved through digital media. Data compression techniques can be “lossy” or “lossless,” depending on whether or not the quality of the original sound is reduced significantly during the compression process. Most music formats in use, such as MP3, are lossy, and can potentially annoy sound aficionados who must hear every note in a sound. Such people still prefer to listen to music CDs, as opposed to digital contents in smartphones, which have been generated from CDs.
Because sound data depends very sensitively on time, it is inherently very huge in size. So, the question of how often (in time) you should sample the pressure becomes critical for digital sound processing. Whatever pressure is recorded at any time is easily converted into digital format via the standard secondary school-level procedure of decimal-to-binary conversion.
With your sound data now in digital format and can be posted on the Internet or moved through digital equipment (smartphones), I can talk about video streaming.
So, what is video streaming? Sounds and music are everyday component of the Internet, and through the Internet you can listen to radio stations, interviews, church services, mosque procedures, sound clips, and so on. You could do this by first downloading the (digital) audio files in their entirety and then play the files to listen to them. Alternatively, you can listen to music from the Internet without first downloading the music files to your PC or smartphone. You do this by streaming audio.
Mat McConaughey, in “Internet Tutorial by Butterscotch.com,” has a particularly simple, albeit wordy, description of data streaming, which is written here verbatim: “A way to transfer data, in a steady stream, like a river, such that it goes out, and the user gets to listen or watch in real time as the data comes in. The movie is compressed down by a Codec that makes the movie smaller in size. Then it gets streamed. As it comes to the program that is receiving it, the program takes in the data, puts it in a buffer, then the buffer sends it to the user. If everything works correctly with the streaming, there is always extra in the buffer, so from the user perspective, it is like seeing everything live. The video, for example, keeps flowing in compressed form, gets decompressed, comes in, goes to the buffer and goes from the buffer to the user. As long as things go nicely, the buffer stays full and the user sees no delay. So, you’re streaming the information in compressed form, then decompressing it, keeping it in a buffer, so that the stream of data stays continuous.”
The main purpose of this article is to describe audio/video streaming, using fairly simple everyday terminologies.