clinton east

The Federal Government have unveiled a discussion paper about Australia’s media in the Digital Age .
The paper is concerned principally with two issues;

1. The use of Broadcast Services Band Spectrum
2. Media Ownership laws.

There has been extensive criticism of the paper, particularly with regard to the second point. It prompts me to say a few things which have been on my mind about the first.
I believe the discussion ought to be framed differently. Isenberg famously said that design-by-assumption works as long as assumptions hold. It’s about time Australia questioned whether, in the long run, the assumption that we need to allocate broadcast free-to-air licences at all. Why? Because spectrum is scarce, and we should be very careful in considering what it’s really for.

Many folk will tell you that there’s as much spectrum as you like; from DC to daylight. That may be true, but in practical terms, it’s a white lie, because not all spectrum is created equal. Low spectrum is favored by operators, because it reduces the capital cost of deploying networks. The reason is very simple; the signal penetrates better, and if you’ve got better signal you need less radio sites (for a given power), so the cost drops - rapidly. That’s why we’ve seen big players switch from 2100Mhz WCDMA to 850Mhz WCDMA. Lower capital cost, but you need the spectrum.

Now, let’s put something in perspective- VHF TV runs at between 40Mhz and 220MHz. UHF TV runs between 525Mhz and 860Mhz. If you count as high as 2100Mhz, even large players consider the costs to be too high. Spectrum is not cheap, and as a proportion of the overall spectrum, broadcast television costs quite a lot.

There are also strong signs that the broadcasting model will not stand up to cheap distribution. If we’ve learned anything from the internet, it’s that people have widely varying tastes. Some people call it content microchunking and rebunding, some call it the long tail. People are after different things, all the time. As distribution and transaction cost to have a one-off piece of content drop, we’re discovering that people vote with their feet and get something tailored to them. It really isn’t such a stretch to imagine somebody wanting Home and Away at 7:15pm rather than 7:00pm, is it?

The current regulatory environment makes perfect sense if you cast your mind back fifty years. The nationwide telephone network was copper. And video could go over copper; that’s as long as you had, oh, say 300 spare copper pairs . Back then, radio spectrum really was the only means of delivering television to the masses. And because it was scarce, we needed licences, and we needed media extensive media regulation.

Is that assumption still true? Well, it shouldn’t be, at least not for the radio part. We now have much more powerful means of distribution. A good quality MPEG stream comfortability fits within the bitrate of an ADSL2 connection. With an assumption the aggregation and core network can handle such speeds, why don’t we just deliver television over broadband? It turns out if you’ve been awake at all in the last five years, that’s what folks have already got planned .

The thrust of Minister Coonan’s discussion paper is about creating more broadcast television and accommodating for what is perceived as changing market forces. But, no matter how efficient you make broadcast, you inevitably run into the same problem; the scarce resource of radio spectrum, limited market entrants, and problems sustaining a competitive market for the public’s attention. Change the assumption about distribution using scarce radio and broadcasting an array of channels, and a raft of problems vanish. We should not predicate Australia’s future on broadcast television, when better technologies exists which encourage competition, and do much the same job.

So, what do we get if we deliver over some better distribution network, such as copper? Well, immediately, each receiving node gets the benefit of a routed data network, and a bona-fide free market for content. If it turns out that consumers want to stream MPEG-4 video of Seven News from their site- fantastic. That’s what IP multicast is for, so we needn’t have drastic inefficiency either.

But hold up Clinton. What about all that wireless spectrum we just freed up distributing data over broadband? What will we use that for?
Well, there’s two, interrelated answers.

1. Fixed distribution networks aren’t scarce, but wireless spectrum is. It makes sense to use small, cellular networks regardless of what you’re doing with the spectrum. Remember the access network problems we’re seeing get slogged out in the news? What better for competition than freeing up some low frequency spectrum for competing access networks? And having nice, low frequency spectrum will mean having low enough capital costs such that competitors might be encouraged to join the fray. An it’s not like using broadcast spectrum for small networks is exactly a new idea.
2. But really, the whole point of wireless is mobility. That’s what you get from wireless that you don’t get from a wire. And so that’s where we should focus our attention. It’s where plenty of economic value is too

The Federal Government have unveiled a discussion paper about Australia’s media in the Digital Age .
The paper is concerned principally with two issues;

1. The use of Broadcast Services Band Spectrum
2. Media Ownership laws.

There has been extensive criticism of the paper, particularly with regard to the second point. It prompts me to say a few things which have been on my mind about the first.
I believe the discussion ought to be framed differently. Isenberg famously said that design-by-assumption works as long as assumptions hold. It’s about time Australia questioned whether, in the long run, the assumption that we need to allocate broadcast free-to-air licences at all. Why? Because spectrum is scarce, and we should be very careful in considering what it’s really for.

Many folk will tell you that there’s as much spectrum as you like; from DC to daylight. That may be true, but in practical terms, it’s a white lie, because not all spectrum is created equal. Low spectrum is favored by operators, because it reduces the capital cost of deploying networks. The reason is very simple; the signal penetrates better, and if you’ve got better signal you need less radio sites (for a given power), so the cost drops - rapidly. That’s why we’ve seen big players switch from 2100Mhz WCDMA to 850Mhz WCDMA. Lower capital cost, but you need the spectrum.

Now, let’s put something in perspective- VHF TV runs at between 40Mhz and 220MHz. UHF TV runs between 525Mhz and 860Mhz. If you count as high as 2100Mhz, even large players consider the costs to be too high. Spectrum is not cheap, and as a proportion of the overall spectrum, broadcast television costs quite a lot.

There are also strong signs that the broadcasting model will not stand up to cheap distribution. If we’ve learned anything from the internet, it’s that people have widely varying tastes. Some people call it content microchunking and rebunding, some call it the long tail. People are after different things, all the time. As distribution and transaction cost to have a one-off piece of content drop, we’re discovering that people vote with their feet and get something tailored to them. It really isn’t such a stretch to imagine somebody wanting Home and Away at 7:15pm rather than 7:00pm, is it?

The current regulatory environment makes perfect sense if you cast your mind back fifty years. The nationwide telephone network was copper. And video could go over copper; that’s as long as you had, oh, say 300 spare copper pairs . Back then, radio spectrum really was the only means of delivering television to the masses. And because it was scarce, we needed licences, and we needed media extensive media regulation.

Is that assumption still true? Well, it shouldn’t be, at least not for the radio part. We now have much more powerful means of distribution. A good quality MPEG stream comfortability fits within the bitrate of an ADSL2 connection. With an assumption the aggregation and core network can handle such speeds, why don’t we just deliver television over broadband? It turns out if you’ve been awake at all in the last five years, that’s what folks have already got planned .

The thrust of Minister Coonan’s discussion paper is about creating more broadcast television and accommodating for what is perceived as changing market forces. But, no matter how efficient you make broadcast, you inevitably run into the same problem; the scarce resource of radio spectrum, limited market entrants, and problems sustaining a competitive market for the public’s attention. Change the assumption about distribution using scarce radio and broadcasting an array of channels, and a raft of problems vanish. We should not predicate Australia’s future on broadcast television, when better technologies exists which encourage competition, and do much the same job.

So, what do we get if we deliver over some better distribution network, such as copper? Well, immediately, each receiving node gets the benefit of a routed data network, and a bona-fide free market for content. If it turns out that consumers want to stream MPEG-4 video of Seven News from their site- fantastic. That’s what IP multicast is for, so we needn’t have drastic inefficiency either.

But hold up Clinton. What about all that wireless spectrum we just freed up distributing data over broadband? What will we use that for?
Well, there’s two, interrelated answers.

1. Fixed distribution networks aren’t scarce, but wireless spectrum is. It makes sense to use small, cellular networks regardless of what you’re doing with the spectrum. Remember the access network problems we’re seeing get slogged out in the news? What better for competition than freeing up some low frequency spectrum for competing access networks? And having nice, low frequency spectrum will mean having low enough capital costs such that competitors might be encouraged to join the fray. An it’s not like using broadcast spectrum for small networks is exactly a new idea.
2. But really, the whole point of wireless is mobility. That’s what you get from wireless that you don’t get from a wire. And so that’s where we should focus our attention. It’s where plenty of economic value is too