|Homepage / Publications & Opinion / Silicon.com
Peter Cochrane's Uncommon Sense: DIY v DIFM Networks
And the winner is...
The 1950s saw an explosive growth in the US TV network with some communities beyond the reach of their nearest transmitter and unable to receive a reasonable TV picture. It was not economically viable to install sufficient transmitters to service every small community so alternative solutions had to come from within these far-flung places.
A few knowledgeable individuals set about building towers with high gain antennas and amplifiers to capture sufficient signal. They then distributed home-to-home using coaxial cable and electronic amplifiers. This was the birth of the Community Antenna TV (CATV), which ultimately became CAble TV in the 1970s and 1980s.
The last decade has seen rapid growth of the mobile phone network, hi-fi, video, games, surround sound and PCs, all based on DIY purchase, installation and maintenance. What is critical is the speed of deployment. In less than 12 years on the planet digital mobiles have overtaken the fixed line phone population, installed over 100 years.
If the TV broadcasters created the programme material and also developed, manufactured, sold, installed and maintained our TVs, then I think we can assume that none of us would own one! In contrast, the telephone companies installed all the wires to homes over a 100-year period and mostly precluded homeowners from doing their own network wiring.
So the IT world has been a contest between DIY and the Do IT For Me (DIFM) approach. I think DIY is about to take over and win.
B2B activity enjoys high-speed networking across corporate LANs and has expanded world trade while simultaneously realising vast operational savings. In contrast B2C has realised an insignificant fraction of its full potential. The principal reason? A lack of bandwidth to the home.
The key question is: When will we all get significant bandwidth to the home and what will the technology be? Will DSL technology save the day? I think not! This is the last gasp of the ancient copper network and increasingly looks like a band-aid solution. It is wholly underpowered even before it is deployed. A 256Kbps or 2Mbps download capability may have been a reasonable prospect a decade ago but the advance of PC technology has seen a huge gap widen between the two domains.
Copper is a static technology that has changed little for the last 50 years. The cables were originally designed for speech signals and now transport data a thousand times faster. In contrast computers, optical fibre and data networks have realised performance improvements by a factor of more than 1,000,000 in 20 years. Moreover there is no foreseeable slowdown in the rate of advance expected for at least the next 20 years. The copper and glass paradigms are entirely out of kilter and copper severely limits any progress.
What can be done? The good news is that much of the local loop cabling is installed in buried ducts that obviate the need for huge amounts of civil engineering. So installing optical fibre is a straightforward proposition. It has been economically viable since 1986 when most could not imagine a need for connections beyond those provided by copper.
Today we see commerce, creativity and productivity crippled by a lack of adequate road and rail transport compounded by no bandwidth to the home and office. Economies are being impacted by the lack of bandwidth and an inability to connect.
Could satellites provide an alternative? No. They will never be able to provide a significant amount of capacity compared with fixed cables. In their history they have made little impact other than in broadcast and the accessing of countries and locations where it is difficult to install cables. Right now the easiest - and interim - last mile technology to install is actually line of site radio, which can deliver bandwidths of more than 10Mbps.
For the network companies the most pressing problem is the lack of an economic business model. The fundamental problem stems from the point economics employed - it is extremely difficult to justify the installation of anything new when you have an established base of copper cables. This is especially true if the copper was paid for a long time ago.
But if we factor in the cost of ownership and new-install costs, then fibre and radio are very competitive. Copper demands a huge amount of ongoing attention with thousands of repair crews dedicated to repairing damage caused by water and corrosion. Fibre is impervious to water and can be programmed remotely because of its huge bandwidth capability. In short, physical re-routings are no longer required.
The scale of the problem is massive in terms of the number of people required and the physical time to install and there is probably no chance that any nation will see a complete change over to fibre in the local loop in anything less than a 10 years period - at a cost of over $10bn.
What is the general public to do in order to get adequate internet connectivity and to continue the growth of the business-to-consumer boom? The self-install of wireless networks is an obvious solution and one that communities are capable of completing with minimal technical skill.
So history may repeat itself with WLANs echoing the earlier CATV. Today Manhattan has over 12,000 self installed WLAN/Wi-Fi systems that could potentially offer contiguous service across complete city blocks. Much of London is the same.
All that is required is for regulators to legislate for the resale of existing broadband capacity. Install Wi-Fi on every broadband link and immediately many more can get access. Coverage will spread like a virus infection and like an infection it will cost nothing to disperse!
This column was compiled after walking around Singapore with my laptop logging hundreds of Wi-Fi installations and observing a near total absence of overhead plant and not many manhole covers! Despatched to silicon.com from my hotel bedroom LAN connection.