Picturequote

The more one sees of human fate and the more one examines its secret springs of action, the more one is impressed by the strength of unconscious motives and by the limitations of free choice - Carl Jung

I really can't escape the fact that I have an innate need to build things. Design is feckless without the tangible aspect of the feedback loop, so I've started putting together what will eventually be a wood/metal workshop.

I didn't realize until typing this out, but this is the first time I've never had access to a workshop. We torn down my family's while I was in grad school and during that time I had access to IIT's shop.

I went to an auction for a cabinet maker that was going out of business (Bolhuis Woodworking Co.) and bought a bunch of hardwood and old industrial woodworking machines. More pictures and details to come. It'll certainly be interesting.

My friend Kevin (pictured) and Derek helped me load up the roughly 1,500 BF of maple, white oak, mahogany, etc., and some old Delta Rockwell machines into a truck on the coldest day of the winter.

Walnut Light Boxes for Medium Format Film

My friend Dan had an idea for some light boxes for 120 film (medium format), so we got together and figured out how to actually make it. After a trip to the lumber store and a whole day in the wood shop this is the tentative result. I'll post more pictures and details when they're fully completed.

Dan got this featured on a blog called Shooting Film.

The slide film is roughly 2-1/4" square (70mm, 120 film) shot with a Mamiya C330.

Donating Your Idle Computer Time for Scientific Research

I've been hearing about networked super-computing for a while now so I thought I'd check it out this weekend by setting up my computer on the BOINC network. It was originally designed for SETI@home and later expanded to other areas of scientific research. I signed up for SETI@home (search for extra terrestrial intelligence), POEM@home (protein optimization with energy models, medical research), World Community Grid, and a few others.

My computer is fairly well suited for it too since it's overclocked, water cooled, and has a high-end 8-core processor, but I was curious about the energy usage. I have an electrical tester and I made a short extension cord with individual wires so I can measure amperage (second photo).

First check the voltage.

This is the computer at idle / regular light use., 0.4 amps.

This is the computer running BOINC (100%  CPU and GPU), 1.4 amps.

Voltage x Amerage = Watts
1000 watts in one hour = 1 kilowatt hour (kWh) = $0.10 in Chicago

Idle / typical  -       50 watts (125V x 0.4A) =   12 cents/day, $3.60/month,   $44/year
Running BOINC - 175 watts (125V x 1.4A) = 42 cents/day, $12.60/month, $153/year

There's also a lot of heat given off by the computer, 3.4 BTU's per watt/hour. That's about 600 BTU's (that'd be like adding three people to a room). For what it's worth I figured that during this January it took an average of 6,000 BTU's/hour to heat our apartment (all glass walls, all electric heat... ugh). Long story short it makes sense for me to lend my computers processing power during the cold months in Chicago, but as soon it gets warm out it would start to impose a decent burden on me since I'd have to both pay for the electricity for the extra processing power and the additional AC required to cool said heat generated by the computer. In the mean time I might as well let my computer crunch numbers as opposed to letting the electric baseboard heaters do essentially the same thing without finding the cure for cancer or ET.

And now the real point of this post; networked infrastructural sharing as it relates to electrical generation - it has many other applications too.

The key insight here, as it relates the networked computing, is that there is an enormous amount of power generation potential in the form of idle cars - just like idle personal computers. Our collective cars have over 35 times more power generating capability than all the power plants in the US. Each power source is about as efficient as the other¹.

The way electrical distribution works right now is fairly straight forward. A network of large plants create huge amounts of energy, mostly from burning hydrocarbons, and then send it through a network to users.

large scale power generation > network > end user

Large power plants are being built at a much slower rate than in the past. Micro-scale sources like solar now make up the majority of added power generation. The paradigm seems to be shifting to:

small scale generation that's very spread out / end user > interconnected grid > end user

Basically, energy and distribution is increasingly starting to look like how the internet works. To be fair the actual numbers are quite low right now, but the trend is rather stable. Over the last decade there has been more capacity added to our electrical grid in the form of micro power sources (solar panels and the like)(comparable to my single personal computer) than there has been traditional large scale power plants (comparable to a super computer).

To add to this, if cars were electric they could plug into the grid and act as a sort of buffer for renewables and traditional power plants. It seems that by distributing the generational capabilities it would become more stable in both reliability and potentially price since the market paradigm would shift from monopolies to pure competition (this is a bit facile but I don't want to get off topic).

This is a trend that is apparent in many other aspects of our lives and technologies, but I chose this example because it made me think about my incentives to loan my computer to scientific research. Currently if you have solar panels on your home and you generate more power than you consume it goes back into the grid². The systems are designed to try to balance out your use and how much your array generates because, at least in Illinois, the utility company won't give you any money if you generate more power than you use. It's really dumb. It creates no incentive for people to add additional capacity. Currently networked supercomputing is viable because it's for a good cause, but why not use it for business purposes too and actually pay users for their resources? It'd be like Zip Cars except Zip car doesn't even need to buy cars, so long as the numbers make sense it seems like an efficient thing to do. In this case especially since nothing tangible is actually being handled by the borrower.

The question is no longer, will solar take off? Is wind viable? Yes, it can be and it's increasingly the reality we live in - which I'm excited about. The real question is how will the incentives be aligned? Will we invest in a smart grid to share our personal small scale power generation capabilities? Will electric cars that act as an energy bank become the norm? Will people get paid to generate power, to let their car drain a bit at 1:00 PM on July 1st when everyone in the US has an air conditioner running? What are the liability laws like if you generate lots of power in a remote area then cut it off leaving others without power? These are serious legal and infrastructural questions that will shape our coming world.

As the rate at which technology advances increases it seems that many of our most pressing problems are political in nature. People have to come together and do some big things. Technology advances exponentially whereas human understanding and will power do not, and it's becoming increasingly apparent. I work for men who are a few generations ahead of me. I learn a lot from them and vice versa, but they're also stuck in a world that's very pre-1990's America. Expensive office space. Everything of value is tangible. Exploitation of workers is the profit margin. Microsoft Office. Email addresses provided by your ISP. These people control our world and I'm not convinced they're aware that the internet and all that comes with it has radically changed our world.

1 - The average car is about 25% efficient while a power plant is generally in the low 30's then some distribution loss.
2 - It's called a net meter. Your meter literally spins in reverse when you're generating more power than you're consuming.

Government Spending: Low and Falling

Just a quick note about an article by Paul Krugman that appeared on his blog today.

Krugman posits that the increase in government spending is mainly due to two factors.

1 - Since the economy has contracted, government spending is going to make up a larger percentage of GDP. Pretty straight forward.

2 - More people are unemployed and are thus collecting from various social safety nets: unemployment, medicaid, food stamps, etc.

The blue line is government (at all levels) spending as a percentage of GDP, so in the last 25 years it was fairly consistently was between 31%-34% of GDP. I'll also note it fell under Clinton and rose during W, although not by too much. Today it's 36%, actually not that high, relatively speaking. The red line is this same measure but as it relates to potential GDP. Basically, what our economy can and should produce. This number is about 34% which is much closer to the trend for the last 25 years.

Source: Paul Krugman's Blog: The Conscience of a Liberal

So there has been some increase in government (again, this is all levels of government, not just federal) spending, but a large portion of it is clearly due to our economy contracting (red line). If one considers #2 from above then it becomes clear that our government, sans safety net programs, has actually contracted in the last several years. This is extremely troubling. 

Politics and our general inability to come together as a nation and implement our hard won knowledge (see Great Depression) has prolonged this recession and harmed many lives in the process. This isn't "oh we just don't have quite as much money..." This is more young people committing suicide, high school students skipping college, older workers not being able to retire when they planned to (even when working class people's life expectancy has stagnated. Source: 1, 2), and increasing rates of depression that are no doubt related to the inability to find a decent paying job. Bad things happening are a fact of life, but watching them happen when they're largely preventable is tragic