Dennis Ferguson wrote:
On 2008-01-24, Mike wrote:
For the past few years internet users worldwide have been reaping the
benefits of the dot com crash where assets such as fibre costing
billions were sitting there doing nothing and owned by financiers who
didn't know how to make money from them so they sold them on at a
fraction of their original cost. All that surplus capacity, sold off
on the cheap is rapidly running out due to ****wit ideas like youtube.

I don't think so. It may be that not much fibre has been buried
underground since the 1990's, but the amount of bandwidth you can get
through a single existing fibre pair has increased dramatically since
then due the continuing, fantastic increase in the density achieved by
WDM equipment. WDM equipment isn't exactly cheap, but it is usually
way cheaper than burying more glass in the ground. It is the case,
however, that WDM seems to be nowhere near as effective for existing
undersea fibre, but prices there have risen enough that people are
starting to spend to lay more of it.

I think the problem there is that undersea cables need repeaters, and if
the repeaters weren't designed for a given modulation you are, so to
speak sunk.

I worked on an undersea repeater some years back. Dreadful project..one
factoid that may or may not be true, is that they had to relay lots of
cables in the coastal shallows due to them proving extremely tasty to
sharks...

I add this because its amusing, and it just shows that sometimes things
are not as simple as they seem...



I don't think the problem with building big Internet services is
transmission bandwidth, there's lots of that and it is still fairly
cheap. The problem is that big routers haven't gotten cheaper anywhere
near as fast as transmission bandwidth, so buying routers has become
a considerable fraction of the cost of a service (and, of course, big
ATM switches are really, truly expensive even compared to big routers,
so if your DSL service uses them for backhaul there's a good chance
that is why it is short of bandwidth).

In any case, this has little to do with what BT apparently wants to do,
which is to bury fibre where there is none now. Whether there is lots of,
or a little, bandwidth where the existing fibre is now makes little
difference to the cost of new stuff.

Sensible caching could reduce core backbone loads on 'broadcast' type
material.

Its the last 4 miles of fibre we probably want tho. Rather than copper,
That the current bottleneck.


BT in their approach are no different to anyone else - but they are
taking a long term approach and what they don't want is some tit of a
politician pulling the rug from under their investment and handing it
out to the 'competition'

A possibility they seem to be responding to by trying to persuade the
same tit of a politician to guarantee them a return on their rather
risky investment. Risk-free investments are wonderful if you can
find them.


I think that is less than fair - I am no fan of BT, but I can think of
at least one project - the M25 thames crossing, that was just about
clear of debt when the government 'regulated' it down to the point where
it will never actually make a profit.


Then theres the opposite side of the coin, where the channel tunnel has
been essentially broke since it was built, and the market price of a
ferry crossing is still way less than what they would need to charge to
make it profitable.

Dennis Ferguson