OT:Energy content of human labor was: Real Linux Men (was: [SGVLUG] Guns, 4WD...and Linux)

[David Lawyer]

Since a couple of people have posted off-topic, now it's my turn.
But if you look at the last paragraph, I speculate how this applies to
Linux.

To estimate the energy require to make corn ethanol, just put all the
facilities and support needed for making ethanol in a small city
surrounded by corn farms.  And make a colored fence enclosing this
community (and farms) with only one entrance gate to the outside world
with no people flow thru the fence or gate.  The only export from this
community is ethanol and the imports consist of all goods needed to
support the ethanol production and the people who live in the
community all of which are directly or indirectly associated with the
production of ethanol.  

Now for every person in the community who works on the farms or in
ethanol production, a few other non-production people will be needed
to provide services to people.  There will also need to be spouses and
children supported to sustain the population.  Also, retirees from
this community will need to be supported as well.  Thus you will need
government, teachers, schools, roads, food, clothing, hardware, energy
depreciation of infrastructure, etc. (excuse the possible double
counting).

So for the ethanol production to yield a positive energy return on
energy invested, the energy represented by the flow of energy into the
enclosed community must be less than the flow of energy out in the
form of ethanol.  But let's look at the just the input energy required
to support the production workers.  To support a person in the US with
a typical lifestyle requires about 100 times the food calories one
eats.  To support all the other people in the community that directly
and indirectly support the production workers, this 100-times becomes
about 400-times.  Then since a person only spends say 1/3 of their
energy working (including the energy spent sleeping as non-work
energy) we now have about 1200-times.  Thus the fuel energy of a
worker producing ethanol or corn is over 1000 times his caloric food
energy expended.  Note that this is only counting the human energy and
hasn't even considered the energy content of fertilizers, farm
machinery, ethanol plant energy (direct and indirect).

This 1000-times model will also be true for other industries as well.
This make the energy content of fertilizers, machinery, etc.
considerably higher.  One conclusion of this is that the people who
have estimated an energy gain on biofuels are very much in error.  And
even the few studies that show an energy loss, while they do consider
human energy, have grossly underestimated its magnitude.

For Linux programming:  Imagine a fenced town like the ethanol
community described above where the only output is Linux programs.
The fuel energy embodied content of the programs will likely be a lot
more than just 1000 times the caloric food energy used while
programming since the rate of burning calories while programming is
low (about 105 kcal/hr vs. 890 kcal/hr for "walking up a 36% grade at
1.5 mi/hr with a 43 lb. backpack -- ref: Table 6 (p.220) Physiology of
Exercise, 1959).  Well, someone did mention backpacking.

			David Lawyer
PS: I would refer you to my draft article about this but it still
needs a lot of work partly because I wrote most of it before I came
across the concept of an enclosed community (putting a boundary around
the problem).  In fact, this email may be more clear and concise than
what's in my draft.  You can find it by typing "human energy
accounting" to google.