Sunday, July 09, 2006

The limitations of genre

After a few years of denial, the broadcasting industry seems to have finally woken up to the irrevocable impact that the digital, on-demand world will have on traditional broadcast models. It has acknowledged that linear broadcast channels are destined to lose market share as the audience fragments in a world of potentially infinite choice. It has also accepted that it needs to respond to changing audience expectations about when, where and how they will be able to access content. However, it is in danger of lurching down another blind alley and its name is genre.

Not so long ago, genre was a word rarely heard outside of academic circles. Now it seems to be popping up all over the place. Apple must accept a fair amount of the credit for this shift, bringing the concept of genre to the masses via the iTunes application. Of course, record shops have been sorting music by genre for decades but I suspect the majority of consumers didn't engage with the term in the way that they are now encouraged to.

Genre has shifted from being a way for experts to categorise for research and retail purposes, to a way for Joe Public to organise and access all manner of things. It is difficult to underestimate the power of naming in creating this kind of cultural phenomena (witness the incredible rise of the word podcast, which far outstrips the uptake of the actual technology).

In the broadcasting sphere, the launch of the BBC Radio Player in June 2002 made a significant contribution to the rise of the concept of genre, affording it parity with the radio stations as a means of navigating on demand radio content. For the first time, BBC radio content was defined equally by its genre as its original broadcast channel, so lovers of Jazz music could find the BBC's rich and diverse Jazz output all in one place - a real boon to lovers of a particular genre of music (or speech).

Four years on from the launch of the BBC Radio Player and genre is becoming increasingly ubiquitous as a means of navigating content that would once have been defined solely by its broadcast channel. For example, the recently launched proffers genre as its primary means of navigation.

So what's the problem, I hear you ask? The problem is that some parts of the broadcast industry are seeing genre as the solution to the challenges of the digital world - a convenient, one-stop replacement for the channel brands that they are told are likely to struggle in a multi-channel, on-demand environment - rather than as just one navigational aid amongst a bewildering array of tools, many of which haven't even been conceived of yet.

Yes, genre is likely to play an important role in future navigation but it is looking like an increasingly blunt tool in comparison with some of the more sophisticated navigational technologies emerging from the latest generation of web apps (e.g. How many music lovers conceive of their musical tastes primarily in terms of genre? I know that I don't. Artists and tracks (and the recommendations they facilitate) are a more meaningful currency in my discovery of new music.

My plea to the broadcasting industry is that they try to embrace the diversity of future navigational models and avoid the temptation to try and impose a top-down taxonomy of navigation, as monolithic as the now deposed hegemony of channel brands.


Anonymous said...


I like this. I particularly like the historical reference to the 'significant contribution' of Radio Player to the rise of genre-based output: nothing to do with the author of FoF of course!

But I wonder who you are speaking to with this cautionary tone: 'some parts of the broadcast industry' getting hooked on 'genre' being who, other than the BBC and other megaliths (indeed perhaps just the BBC)?

The very fact of Last FM and indeed iTunes which is hardly dominated by assumptions of genre-based engagement suggests that 'genre' is only a preoccupation for megaliths, e.g. Auntie Beeb, looking for a convenient way to rebrand their former output channels. R4 becomes 'spoken word' etc.

I would also make further points that develop and somewhat mitigate your caution. As a starting point, it looks like we have three options: channel-output (selected by editors), genre output (classed by content itself), and search (by whatever metadata is available and interesting to the user). Your argument seems to be that channel is dead, genre is overhyped, and search is ever-more powerful. I agree on search, but I think genre is going to be redfined, and channel will have a major resurgence. In reverse order then, and in detail:

First, I think genre is going to morph, readily and quite naturally, into 'brand'. I can't see static genre stabilising as a long-term top-level organisational concept for content. Consumers will sooner or later be channelled down 'brands' of content that basically match their patina of preferences. So, rather than clunky, singular 'spoken word', 'hip hop', 'easy listening' (joke: can't imagine that monstrosity of genre-classing will survive the hyper-defined age) genre-categories, there will 'urban', '80s retro', 'DINKY', 'expat teleworker' etc brand-genres that have carefully fine tuned content - of all sorts - to cater to specific audiences across their whole range of appetites. The reason for this is obvious: once channels disappear, the fact remains that the majority of consumers will not have the time, tech savvy, or interest to build their own 'blend of genres', nor even to search for content. The supernova of channels - in effect genres - of cable TV, and genre-formatted output on the web, demonstrates the problem of on-demand genre-based (or search-reliant) programming: sub-niches of shite till the last IP address and fraction of spectrum is taken. Consumers, soon enough, will simply wish to purchase one or two 'brands' of mediatainment, but this will have to be damn well calibrated to their taste. Hallmark Channel? Martha Stewart Channel? Maybe brand channelling already exists.

Compared to your review of personalised homepages, this is rather like channels being replaced with 'customised portals' where, however, the cusomisation has been done for you.

Arguably, and ironically, this takes us back to the old channel idea: the point of a channel, since this is inherently an programming idea that has become confused with single-channel broadcast practice (you could easily have, and did, have different channels broadcast on the same frequency in early days of broadcast; the Third Programme was at one time a Programme not a channel, etc.), was to groom content into a 'blend of genres' that matched a certain 'audience demographic'. The only difference is that it will be massively more tuned to preference, since at any point, if users aren't getting what they want they can either get it for themselves or switch 'brand'.

What about 'genre' per se? What about that? The reason why genre has emerged over channel is that is is a selection-aid in the on-demand world; but if you follow my argument, 'brand' will resurface to map back onto the conceptual and operational space formerly occupied by 'channel' because self-selecting content from 'genres' will never work for most people. And two things about genre which are inherently rather dud in the first place are: a) it has been hyped as an on-demand feature largely in the absence of good search or sufficient meta-data tagging of content, but both both of these are now being solved (the former by computers, the later by users); b) it is an arbitrary way of classifying content (when do 70s/80s 'public information films' become 'comedy' as they surely do?).

So, while genre has been a temporary way of solving the organisation problem in the on-demand age, it will be replaced by 'brand channels' on the one hand, and good search on the other. But I think genre can survive too. Most of the genre classing in very modern content portals is not editorial or arbitrary (as it however still is on iTunes): i.e. the categories do not come from on-high (e.g. easy listening...what non-musician ever thought of that 'genre'?), nor selected or invented by editors (e.g. trip-hop, Balearic, acid). No, the 'genres' on technorati, delicious, YouTube, Last Fm, Flickr etc are 'dynamic genres' based on either tags or search-terms. I think is the likely future of genre, perhaps rooted in some fixed, crude structures (rock, pop, classical, world), but animated by dynamic sub-genre category features. Not only do genres emerge, grow, die back, compete, but each genre develops: while one of the problems I have with Last Fm is how bad the 'dynamic genres' are (in the Ennio Moricone 'genre' I got a Black Sabbath-alike, a White Lace-alike, and some spoken word, in order, with no Ennio M), if such a concept works better, it can evolve to include unexpected things, and be the better for it.

Finally, for search, the classic problem with search, which is why genres will persist even when search works properly, and indeed why 'brand channels' will become more popular, is that..people don't know what they are looking for. If I didn't know that The Concretes existed, I couldn't search for it; and even if I know some categories by which it might have been tagged, what if I don't know them? If channel is all push (and runs the risk of not meeting expectation), and search is all pull (and runs the risk of offering too much), I think we can see the importance of genre, defined dynamically as above, as a meeting point between the two. Not genre as pre-ordained silos, but genre as the most popular band search this week, or most downloaded track, or most watched video etc. These are the genres of the future: as such, search feeds back to genre.

Anyway, I think all three will survive, but in new forms.

Dan Taylor said...

Incisive comments - thanks John. Don't suppose I can persuade you to come work in the media?

Anonymous said...

Not likely. My hope is that I can somehow, one day, find a way of being a content provider of whatever sort! And conversely, I would myself like to convince minds of your calibre to work in the sustainability field!