Om Malik thinks online radio's time has come.  And gone.  He's wrong.

I've never met Om but his blog is one of my must-reads for all things broadband and I respect his opinions though I don't necessarily always agree with them.  His argument is 3-fold:

1. People won't spend money for online radio, especially with all of the subscription services around, which will make it unnecessary because you can mass customize what you want.
2. Most radio listening is done in the car and satellite provides a compelling alternative.  Also, radio content stored on ipods will be hooked up to cars.
3. It won't be offered at work due to bandwidth throttling by IT managers.

Actually, I think the issue is partially a semantic one.  Streaming, podcasting, satellite, analog and digital broadcdasting are all distribution technologies for audio content.   As a consumer, I don't really care what technology is behind the curtain, as long as the value it brings me is worth what I pay for it (whether it's via my time to listen to ads or my money to subscribe). 

So while Om thinks that streaming radio is separate from podcasting, I consider them to be separate faces of the same "IP-enabled Radio" coin, if you will, and not mutually exclusive.  If I were a radio brand, all of the distribution technologies above would be fair game for me to distribute my programming (though the business model may well vary depending on the economics and value prop of what I'm offering).  I consider the playlist streaming services offered by Rhapsody & Napster to also be IP-Radio.

I think Om would agree that there is a need for IP-Radio.  It allows for many more programming options than satellite and terrestrial (whether analog or digital) can offer, not to mention interactivity and flexibility.  Since network connectivity has lagged storage in terms of cost and availability, space shifting technologies like Podcasting (essentially RSS with enclosures), have become a viable mechanism to distribute audio content and will continue to gain currency.  So why stream?  Let me count the ways:

1. Unlike podcasting, streaming copyrighted music is completely legal with clear-cut licensing and royalty rates.

2. It's much easier to do ad-supported radio.  I can insert ads dynamically and replace any ads from a terrestrial simulcast.  I can also target the ads to demographic or geograhpy.  (and listeners can't skip the ads as they might with podcasts)

3. I can version based on geography.  For instance, listeners in my radio station's region would get the regular simulcast with local ads; those outside the region but in the US would get national ads; and those tuning in from the UK would need to pay.  Streaming supports these models.

4. Many, many office workers don't have a radio at their desks.  They do have a computer, which they use to stream.  If I don't stream, they will tune into one of my competitors.  Om is right that IT departments crack down on bandwidth-intensive applications.  But he is wrong in its impact on overall listening.  It's a well-known fact that the majority of streaming radio listeners do so during the workday, and presumably from the workplace where they have broadband connections.

5. I can create a web-based player to support advertisers (like a branded player), commerce, community, etc.  Yes people will minimize it while they're working, but they do interact with it to learn song titles, click "Buy" buttons, and email the DJ.

6. I can create many side channels.  The Long Tail, if you would.  I can even let my listeners create them and offer them to others via my site.

I'm sure there are more reasons, but these should suffice for now.

Om goes on to claim that online radio has not penetrated the mainstream consciousness.  Well it certainly hasn't garnered the ink that the satellite companies have (though raising and spending several billion dollars between the two of them might have had something to do with this).  Worse, much of the ink for online radio has been negative -- CARP royalty rates, AFTRA licensing, companies going under.  Yet despite all of that, Arbitron & Edison Media Research estimate that 37 million Americans listened to online radio in the past month.  Terrestrial radio is still much, much larger; but 37 million monthly consumers sounds like a mass medium to me.  And, currently, it has vastly more listeners than podcasts, though this will change (and notwithstanding Pew's dubious claim that 6 Million have downloaded podcasts).

From a business perspective, online radio’s failures have outweighed the modest successes out there.  That should be a strong argument for streaming royalty rates to be set lower in the next CARP, and to be more like the rates satellite pays (which in itself is fodder for another post).  And yes there was a lot of exuberance in the late 90’s –Yahoo’s purchase of Broadcast.com and AOL’s of Spinner probably had something to do with it. 

Om closes with:

    The Internet enables consumers to be highly selective. In other words, digital music is the first big market that allows mass customization. With an iPod/iTunes or Creative Zen/Napster combination, I can create my own radio stations, sans ads. It doesn't even cost much. Companies like Napster will rent you as much music as you want for $15 a month.
    Why listen to someone's generic playlist when you can create your own? (Looking for help finding new music? Free online music recommendation engines like UpTo11.com can find tunes you might like.)

    Do you really need online radio? I didn't think so.

First, I consider Napster & Rhapsody to all be in the IP-Radio family.  But, to answer the question:

a.    Streaming radio is free.
b.    Most people are passive.  They are not going to go to recommendation engines to find new music.  Instead they discover new music from friends and other trusted sources (blogs, publications and yes, even radio!).  Radio is a formidable music discovery tool (just ask the labels) and IP-Radio opens up great possibilities.  Yes, you say, but they can have their personal radio station be algorithmically programmed.  To which I answer:
c.    Don’t underestimate the human element.  I love listening to algorithmically-created, playlist-based music, but I also love it when I hear a trusted source talk to me and tell me why they love this track, or the history behind this artist.  That is part of the programming success of XM & Sirius.  People like to hear other people; they just don’t like it when it devolves into incessant chatter and interminable commercial breaks
d.    Playlists will rise in importance.  Yes, you’ll create you own, but you’ll also be tuning into those of others.  You’ll share yours with your friends (and introduce them to new music), and vice-versa.  With Apple’s iMix playlists, you can buy what celebrities are listening to.  Playlists are a subtle form of tagging that you could ‘subscribe to’ (haha, I’ve mentioned both the long tail & tagging meme in one post). 

So I think IP-Radio, with its different flavors (algorithmic playlist streaming, podcasting, simulcasting, plain vanilla streaming) is here to stay, with streaming being a key component, but not necessarily the only one.