> A particularly apt application of the term, as early computer weather
> modeling work was one root of the development of
> Chaos Theory. See James Gleick, Chaos.
> <http://en.wikipedia.org/wiki/Chaos:_Making_a_New_Science>
As a tactical planner in the telecom industry, it is my job to make
wild projections about the future. Not just for next year's budget,
but for five plus years down the road. I'm REALLY GOOD at building
formulas and so forth, but that always has to be tempered with my gut
feeling and local knowledge. The numbers are a good starting point,
but don't assume intangibles. For example:
1. Historically speaking, bandwidth demands and capacities have
doubled about every 20 months since the first commercial
implementation of the telegraph. What makes me think that this trend
will stop any time soon?
2. That growth has NOT been a straight line, but has seen substantial
growth plateaus.
3. That growth has NOT been even across all geographic locations.
4. The customers and uses keep changing.
So, in my models, I do use historical growth patterns to predict
future growth, but also have to pay attention to local economic
issues. Does the community in question have a large employer which is
continuing to expand or is it a mature business ripe for a takeover or
undercutting by a foreign company? What kind of economic diversity
does the community have? Is there a data center located there? (data
centers are sometimes located in some pretty out-of-the-way places
because that's where the fiber routes are). I do look at local
conditions and even "as run" data to determine what places have
over/under performed. Through all that analysis I can conclude the
following:
1. Bandwidth and consumption will continue to double about every 20
months over the long term over my entire region. However, any one part
of that region WILL severely over or under perform the averages.
2. The current rapid expansion is NOT a predictor of things to come.
We've been in a relative 10 year plateau and we're now in the middle
of a growth spurt. Once caught up, there will be a little bit of a
breather as it becomes incremental additions instead of the current
wholesale technology swaps.
So, to apply this to global climate, I look at some of the same types
of factors:
1. What has been the overall historic LONG-TERM trends.
2. Determine what is either a localized issue or what is a contributor
to a stair-step of change.
It's easy to look at the "hockey stick" of climate change models and
predict that this will continue on to Armageddon. But the problem that
I have with that is that for the hockey stick model to continue would
require the ignoring of the long-term trends which show that rapid
change is just a burp in the curves and not an entire redefining of
the curves. I do believe that Earth science shows us that historically
speaking that an action (increase in CO2) will result in a reaction
(increase in things that consume CO2) that will temper the increase.
We just haven't seen the growth in CO2 consumption yet.
--
Ken Norton
ken@xxxxxxxxxxx
http://www.zone-10.com
--
_________________________________________________________________
Options: http://lists.thomasclausen.net/mailman/listinfo/olympus
Archives: http://lists.thomasclausen.net/mailman/private/olympus/
Themed Olympus Photo Exhibition: http://www.tope.nl/
|