Book title image (GIF 6K) About the book Updates Feedback Home

A sample from
Web Radio. Radio Production for Internet Streaming

This is the first half of the book's final summary chapter, which references the earlier chapters in which the detail is contained ...

The copyright for this extract is owned by Focal Press. All rights are reserved.

Chapter 12 So how is web radio different? A checklist.

Part I

(Click here for Part II.)

The aim of this book has been to provide an overview of what is involved in putting together a web radio station:

In each of these three areas the book is a guide rather than a set of definitive instructions. It is an invitation to explore as a listener and experiment as a producer, because we do not know where this road is taking us: an invitation to research the possible routes ahead. In the process of this research we may find out almost as much about 'old fashioned' analogue radio as about the nature of its upstart offspring. For those of us interested in the study of radio the Web provides the most comprehensive toolkit:

For all the differences I have put forward in this book, web radio is still a continuation of the 80 years or so of radio tradition that preceded it. Many fundamental techniques, tricks and shortcuts have been learned along the way about what works in sound for the listener. Those lessons learned are not restrictions, more of a kind of liberation in the hands of the skilled programme maker. They cover a huge variety of styles of radio and they shift around according to the ways listeners use it: there is no one right way of making radio programmes. But what that accumulated experience does tell us is that there are plenty of wrong ways to make any kind of radio: uninterested presentation; false intonations; badly researched content; convoluted scripting; lack of sign-posting what's happening or what's coming next; obvious, jarring editing ... the list goes on. Because it now seems technically quite easy to get 'on air' (i.e. online), there is a the danger that we don't have to think enough about what we put out there - as testified by the wealth of stations on the Web who can count their weekly reach on the fingers of one hand. We don't need to relearn these lessons for web radio.

So there are the essential aural qualities and production techniques that both forms of radio share in common. And then, superimposed in broadcasting, are institutional customs and practices, regulations, assumptions about audiences: limitations on the medium which have evolved over time and through very different eras than our own. These facets of established radio we can largely forget about in web radio. Time to rethink. Of course we can decide to run a web station along the same institutional lines as a conventional broadcaster - and those also broadcasting their web output on the airwaves have no choice - but the potential of web radio is diminished if we treat it only as an imitation of what mainstream radio broadcasting has become. The mass audience characteristics of DRM, DAB and satellite are better suited to that role, if we are imminently to leave analogue transmission behind. Indeed there is the real danger that the more we treat web radio were broadcasting the more it will become exactly like another (rather inefficient) form of broadcasting - regulated and organised along the same vertical lines. And for a second time in history we will have passed up the opportunity to find out what radio in its horizontal, interactive form could be for us.

In this brief chapter then I'll draw together the main themes that have emerged from the book and attempt to summarise these distinctive ways in which web radio differs from terrestrial radio and from which its real value derives. In Chapter 1 I suggested two questions to use in order to do this:

I'll apply those across the general headings of :

Each one of the tentative conclusions - or theories, call them what you will - on the following checklist seems to me to warrant its own research project.

Review of the characteristics the Web adds to radio

Web radio works best as a narrow-cast or niche medium. This is its advantage. Analogue radio found its most efficient use as a broadcast medium - even though it can be used to target small geographical areas. Web radio can be used in a 'broadcast' mode, but in the context of this comparison it cannot compete with the real thing., for so long as big enough concentrations of audience remain to justify free to air broadcast radio. (Chapters 3 & 11)

Web radio is a 'non-zero sum' medium. This is shorthand for saying that the Internet's infrastructure is an elastic resource, whereas the spectrum of radio frequencies is a fixed resource within each geographical area. The Internet can continue to accommodate new stations and each station can continue to add new streams to make more than 24 hours of programming available in a day without taking away any existing services to make space. Analogue radio became broadcast and increasingly vertically organised because it is, in principle, 'zero sum' and hence requires some form of regulated system to function: for each new station added to a full radio spectrum, one must be taken away and for each new programme added to a 24 hour schedule the same length must be subtracted to make room for it. (Chapter 4)

Web radio aggregates listeners' specialist interests. Combine the fact that web radio is non-zero sum medium with its global reach and this further strengthens its advantage as a narrow-cast medium. There may not be enough fans of Celtic folk music in your part of the world to justify a local broadcast station for it, but if you add together listeners scattered around the world who are missing that kind of entertainment the potential audience looks a lot healthier. Each new niche is an addition to the sum of radio. Some niches may be well filled by simulcasts of stations who do have broadcast licences elsewhere, but many more can come from web only stations. (Chapters 3, 5 & 11)

The scalability of web radio cuts both ways. It makes all the difference at the entry level of radio where audiences, hence bandwidth, hence costs are small, but becomes increasingly burdensome the more (thousands of) concurrent listeners a station attracts. Multicast can be a way around this but excludes on demand interactivity. The sums in this equation will continue to change with the technology, but in the longer term scalability will continue to go hand in hand with narrow-casting. (Chapter 4 & 5)

Web radio is inherently interactive. Whereas terrestrial radio stations have to increase their identification with their broadcast by other means - especially now through the station web site. But email forums or request lines or chat rooms don't look after themselves and they require the investment of time and money to maintain. Ironically the more portable web radio becomes and the further removed from a reasonable sized screen and keyboard the less of this visually interactive edge it will retain. There is no reason though why such interactivity should be confined to text. Listeners can become increasingly more involved as contributors or makers of programmes if a web station creates that opportunity. The tools to make and transfer high quality programmes are now widely available to anyone with the interest and the outlet. (Chapter 6 & 10)

'On demand' transfers control to the listener. The ability of web radio to offer both live and on demand listening streams is a unique asset. On demand streams need to be used and organised with care by the webcaster, but both for simulcast and web only stations they can build the total listening audience for a programme both in the short term (in a 24 hour or weekly cycle) or the longer term. This also has the effect of spreading demand, so audience figures for a programme accumulate over time. On demand archives are also starting to create an invaluable, collective, global memory for all forms of radio. (Chapter 11)

Web radio can measure audience preferences directly. A unicast server can collect data about most requests it receives for a stream. Correctly used, that data can be converted into the kind of listening statistics broadcast audience researchers can only dream of - even with allowances for streams left connected but unattended and the fact that IP numbers can be 'concealed' by listeners who want. As with any set of statistics these can be used positively or negatively. They can allow a station, for example, to respond to listening patterns, to test interest in a new programme, to relate airplay to click through record sales or to provide advertisers with instant feedback on a particular campaign. On the other hand, their precision can encourage a relentless, instantaneous, morale-sapping measurement culture - to a much greater degree than is possible for terrestrial radio with its less exact and retrospective audience measurement methods. Both makers of and listeners to new programmes need nurturing and time to develop their affections. (Chapter 4)

Managing demand. The Internet's version of 'word of mouth' can spread implausibly quickly. So it is always possible that today's a quiet backwater of web radio will tomorrow be deluged, should one visitor decides to email 20 friends at work, each of whom email 20 more, and so on. These powerful surges of interest can be positive or negative in terms of reputation (and a server's ability to cope). Whichever, the station will find out about it very soon - another dimension of instantaneous feedback that broadcasters never experience with such direct force. In broadcasting swings in popularity happen over much longer periods and, perhaps more importantly, sudden surges will almost certainly never be registered accurately on broadcast audience listening figures. (Chapters 3 & 4)

Web radio as part of the mosaic of radio platforms

The evidence at the moment is that the above characteristics mean that web radio can be complementary to broadcast radio in several ways, just as it cannot compete with it in others. There are areas of radio transmission it can cover more effectively than any other digital of analogue route.

Adding a global audience. A terrestrial station can reach beyond its transmission area to listeners in other regions and other countries. A web only station can define its international appeal from the outset. While this capability is not unique to the technology of the Web (think of the highly resourced external services at one extreme or ham radio at the other) the way the industry is now structured web radio is proving to be the most practical and accessible way of achieving it. (Chapters 3, 5, 6 & 11)

Adding highly targeted audiences. Similarly, at the other extreme, web radio can get itself into individualised spaces not easily reached by terrestrial broadcasting (although of course micro-power transmitters can cover very small patches). Thus the Web can add audiences in particular locations within a terrestrial stationÕs transmission area, notably office workers for whom FM or AM reception is poor within particular buildings. (Chapters 6 & 11)

Simulcasting a terrestrial output on the Web. For an established station this is a simple but effective way of achieving the above incremental, additional reach. However this tactic is not without potential problems: the cost of additional bandwidth sufficient to cope with very large simultaneous demand can be prohibitive; tensions are likely to arise between local and global targeting (e.g. over news and advertising); similarity with other stations' playlists may make it a pointless exercise; music copyright issues need thorough investigation. (Chapters 5, 8 & 10)

Adding side channels. Terrestrial broadcasters are experimenting with adding side channels accessible from their websites alongside their simulcast. Some are narrowly targeted music channels, some are archive channels for special interest content. This is a smart way of integrating the unique niche and on demand capabilities of web radio and at the same time reinforcing the interactive attraction of their websites. (Chapter 5)

Integration of tuners. For the listener the transmission route is usually immaterial so long as they can tune in to a clear signal. On this basis the pioneers of web radio receivers are demonstrating models that integrate an Internet band alongside FM and AM in the same tuner. Similar integration is being tried with DAB and digital direct satellite delivery. (Chapter 3)

In house radio stations. Again in house or internal radio stations are not unique to web radio (store chains have used satellite distribution of their in house stations for years, designed to build the brandÕs customer relationship). But the Web now provides an easier way for any suitable institution to do this without occupying valuable transmission spectrum. Even geographically dispersed organisations can run their own flexible internal radio station on their Intranet. It may, for example, only want to transmit for an hour a two a day. (Chapters 3 & 4)

Web radio as a proving ground. Web radio has the potential to be a rich resource for broadcasters as a proving ground for new presenters and, we hope, daring and imaginative programme makers, prepared to think beyond the mindset that has evolved in the process of chasing median tastes to maximise audience. Claire Condra (2000) suggests for example that a broadcaster with a website, who is looking to create some niche content for a side channel might be able to find just the programmes or presenters already finding the target listeners on their own small sites. This approach could enrich the total mix of radio programming. (Chapter 10)

Automated music channels. The global mass market in formatted music stations can now be most efficiently served by automated music channels - whether they appear on the Web or on satellite, digital or analogue terrestrial stations. The Web has at least three clear advantages in this trend:

The economics of automation make it almost inevitable that music channels will come to occupy a very substantial section of the total listening market. I have been keen to stress that this form of listening is very much on the margins of what we know as radio. However I have little doubt that on whichever platform it will transform the pattern of radio listening. (Chapter 2)

Part I ends. Click here for Part II.

Return to the top of the page

Chris Priestman, December 2001
Taylor & Francis Group LLC, 2002.
All rights reserved. No part of this extract may be reproduced in any material form (including storing in any medium by electronic means and whether or not transiently or incidentally to some other use of this publication) without the written premission of the copyright holder except in accordance with the provisions of the Copyright, Designs and Patents Act of 1988 or under terms of a licence issued by by the Copyright Licensing Agency Ltd, 90 Tottenham Court Road, London, England W1P 9HE. Applications for the copyright holder's written permission to reproduce any part of this extract should be addressed to the publishers.