Of course, we cannot eliminate uncertainty altogether, so we rely on “experts” to help us measure it - is a risk really a RISK or is it just a risk? When experts agree, we start to relax because we compound the weight of their opinion - that many experts cannot be wrong!

Furthermore, we assign greater weight to those that agree with us than we do to those who don’t… It is human nature - the herd mentality that you see so much of in today’s social networks - the result is that you get a WikiMyth or even an urban legend.

The Thought Process
The problem with our method is that we don’t know whether experts are primary or secondary sources. Are they truly original thinkers who have put all the pieces together themselves, considered diverse views with an open mind and arrived at a point of view? Are they still open to alternative views, even after they have built a position and presented it to the world as their truth? Or are they people who have picked up a theme and then massaged all the evidence to fit what they are comfortable with and conveniently ignore the stuff that doesn’t fit?

Both types can be horribly wrong in their conclusions, but when primary sources are debunked, they have nowhere to hide. For every Nostradamus there are probably a hundred David Ickes. History reflects very differently on those two “prophets”.

When it all goes wrong for secondary sources, they can often be found comforting themselves with the blanket that they were (after all) just following the crowd. Where do you see yourself?

The more time you give yourself for the uncertainty to fade, the better your conclusions. The flip side of this in business is of course that the more time you give yourself, the more time you give your competitors. If you want the first mover advantage, you need to evaluate the risks and take a position. You have to be prepared to be wrong.

So what on earth does this have to do with Telco…?
I am of course referring to the bombshell that was delivered about a week ago by Dr John Papandriopoulos, where he claims a solution that will deliver 250Mbps over copper wires. If this is true, it puts a whole new perspective on the need to lay fibre to the home. Why would you, if you can get the same speeds on existing copper?

Just as we were beginning to believe that FTTH was the only way to give homes 100Mbps access - the consensus of the herd - some bright spark comes along and delivers a whopping great “what-if?” Hold your horses…?

Uncertainty is the biggest barrier to investment, and this new dynamic adds a whole new wave of it to the investment case. Is this real? Does it work? What does it really mean…?

What is this all about?
Let me try and pick this apart a little. Is it real? In theory, I think we can be pretty sure that it is. Firstly, VDSL2 is theoretically capable of 250Mbps already.

Source - Infineon

Secondly, Dr John is/was a student at a well respected university, who awarded him the Chancellor’s Prize for his work and whose own venture fund is funding the turning of theory into reality.

It is worth noting that the thesis containing the theory was published in 2006, over a year ago, so they have time to scratch the surface. The fact that the uni itself is backing it to the hilt - there are no other venture partners - is a sure sign that this is real.

What happens next?
The question of who owns the Intellectual Property is very interesting; it seems that Melbourne Ventures are the contacts for now and that might suggest that the University owns the rights to the invention, particularly given the contribution of tutors who are paid employees of the organisation.

The University has an IP policy which deals with this issue up front, which John will have known about but I hope he got a good deal on royalties as the inventor… It could be the copper equivalent of finding the cure for cancer and even if it did get him a job at ASSIA where no doubt he will be highly valued, he may never come close to another breakthrough like this. Let’s hope this doesn’t leave a bitter taste.

Does it work?
Here we can be a little less certain. Yes, the formulas certainly work and yes, it almost certainly works in the lab - that is what unis do best. Will it work in the wild? Maybe? Probably? Almost certainly…? I don’t think we will know for sure until the patent is awarded, the theory is embedded in equipment and users trials in significant volumes start.

This all needs time and depending on who you are, this delay will play very differently. If you are an incumbent that owns the copper, then it might make sense to wait and see a little. If you are a competitive carrier considering a fibre build, then perhaps it gives you a window of opportunity while the incumbent you are competing with is wracked by uncertainty. Uncertainty is an opportunity as well of course…

One situation it doesn’t change is new builds, where there is a choice that must be made between laying new fibre or new copper. There, the only decision is to go with new fibre a) because its performance is known and b) because it doesn’t cost you more (and some argue costs you a lot less to maintain).

But what does it all really mean?
Dr John is moving to the US now and leaving his beloved Australia behind. Quite what his role at ASSIA Inc will be, is unknown, but the company’s chairman is Stanford Professor John Cioffi who said last year in an article for The Online Reporter “The main obstacle for the advancement of DSL technology is the interference (’crosstalk’) generated from different DSL lines that share the same telephone cable binder,” This was published just over a year ago - long before last week’s developments.

Looking underneath the headlines, the claim is as follows:

“Two main factors limit DSL speed: long line lengths and crosstalk interference … Our technology … aims to manage this crosstalk interference, consequently allowing telecommunication providers to maximize the data-rates of their networks. We can do this dynamically, and adaptively, to try and get the “best compromise” of interference between neighboring lines to maximize performance.”

It manages crosstalk. It does not eliminate crosstalk, although it is dynamic meaning that it copes on the fly with variations. What does this mean in reality? I am uncertain.

Is this what we have been waiting for? It maximises the data rates of their networks… at least until an even brighter spark comes along and increases the maximum theoretical speed of copper beyond VDSL2’s 250Mbps. Impossible? How can you be so certain…? What about IPSL from Rim Semiconductor’s 384Mbps over copper (field trials coming shortly) or further developments from Cioffi’s group that could deliver 1-2Gbps on copper. Is it real? Does it work? What does it mean…?

Is copper a bottomless pit?
The steps being claimed are still very large and that leads me to believe that actually, it may be while before we reach the point of technology delivering diminishing returns. Of course your fibre investment is future proof to some degree, but as sure as eggs are eggs, rolling it out will get cheaper the longer you wait.

“According to Japan’s incumbent, NTT East industrial FTTH deployment costs have come down from USD 5,400 per subscriber (including construction and equipment) in 2002 to just under USD 900 in 2006. It is forecasted that costs could fall to USD 650 in 2009.” according to Finnie, G. (2007), FTTH Worldwide Market & Technology Forecast, 2006-2011, sourced via Katja Mueller’s report on FTTH in the UK.

Getting more out of copper gives the bridge that you need to wait. Of course, you cannot wait indefinitely though - unless of course you are a monopoly that is immune to political pressure. If I had to take a position now, it would be to roll fibre to the cabinet but not to the home. Why?

It seems that in spite of all the advances, there is one constant. Speed over copper degrades with distance. That is true of VDSL2 and even IPSL, and of course we know it is true of ADSL and ADSL2+. I have written to both Melbourne Ventures and Rim Semiconductor asking them what their inventions deliver over longer lines which I will update you with when I hear more, but at the moment we are where we were - shorter lines = faster speeds. And, we have a lot of long lines in the UK.

Going the whole hog
There is another viewpoint though. I have heard it claimed (for example by Angus Flett, Director of Product Management at BT Wholesale) that “If you do VDSL2 … then you have to do fibre to the cabinet, and if you do that then the economics mean you might as well do fibre to the home.”

Intuitively, I’m not sure I buy that and I have seen data comparing the two that doesn’t back that statement up either, with an order of magnitude of around 1.75 to 2 x differentiating the cheaper option of stopping at the cabinet and using xDSL from there.

But there are other factors like maintenance cost. I heard an eminent technologist say recently that if BT ditched its copper and went for an all fibre network, it could also ditch around 100,000 staff which would deliver the ROI in a flash - even with BT’s very generous redundancy packages.

There have been some very well researched studies by thought leaders who also argue that FTTH is a commercial slam dunk. Perhaps all that BT is waiting for is for Ofcom to give them a greater return guarantee before they take the plunge?

Honestly though, have we really thought this through?
There seems to be a heightened sense of certainty that FTTH is the answer, to the point that some people have actually forgotten what the question is. If the question is “how do we spend £25bn”, then yes it probably is the right answer.

One of the most admirable statements I heard at Telco 2.0 recently was Steve Robertson, CEO Openreach stating that they will only do FTTH if and when they can do so ubiquitously and still make money. No digital divide, not £10bn for 90% coverage, not $10bn for 75% coverage. 100%? Probably not in reality, but 99.6%?

But even so, what is the question? Are we trying to deliver more gigabytes? Or gigabits per second? We use around 1% of the installed local loop’s gigabyte capacity and yet we demand upgrades. For sure gigabit per second capacity is subject to congestion, but there are different ways to address that. One is of course to build faster pipes, the other is to spread the usage better on existing pipes. Which would be cheaper…?

So, what if…?
What if Dr John Papandriopoulos’ invention, or Rim Semiconductor’s IPSL, or ASSIA’s future wave of Cioffi inspired technologies really can deliver what they promise? Does the fibre case still stack up? Especially if we can buy time by learning to use the 1% of capacity we use today more efficiently by looking again at the relative importance of speed versus storage.

The more I look at this issue, the more I see false certainty in the conclusion that the FTTH lobby has reached. Is is not the only answer to the demands of 21st century computing, even when that phrase is taken in its widest possible context to also include the complete replacement of 20th century broadcast networks which is another case that for me is not proven.

But, as I said earlier, business cannot wait for certainty. All it can do is take the best advice on board and judge the relative value of doing something against doing nothing. Doing nothing is a valid strategic move, remember, especially for monopolies.

One final what if for you to consider… Copper is not dead, that we can be certain of. What if the regulator made the following offer to BT: you can build fibre and charge what you like for it, but in return you must divest the copper network including IRUs on the physical premises and backhaul.

If you were BT, what would you do? If you were Virgin Media would you buy it? Sky, Carphone Warehouse…? What does that tell you about your real position on FTTH?