rianjs.net

I generated a public/private key pair on my local machine, and I'm experimenting with copying files from here to my webhost without needing a password. Took me a few minutes to get working, but now I'm playing with it, and I accidentally hit [tab] while putting together an scp command on my local machine.

It autocompleted with what's available on the remote machine.

That's pretty fricken cool.

On the heels of my four-part series this week, Amazon has launched their new "Green" section:

I have my doubts about the real usefulness of some of the top items, but the CFL bulbs are the real deal. While I would never pay $14 for one of them, we did get about three dozen of them on sale at Building 19 for something like 17 cents each about two years ago. We noticed a significant drop in our electrical bill right away on the order of ~$10-15 a month. Bonus has been that we've had to replace one(?) bulb during this time, so we still have quite a few of them stashed away. One downside is that they don't seem to last very long in New England weather extremes, so we use a regular incandescent for our one of our outside lights.

[Completely unrelated, but what's up with the crappy graphics Amazon uses for their navigation? Honestly, Amazon, just use text. It loads faster, and looks nicer.]

This is part one of a four part series:

1. Part 1: The Little Things
2. Part 2: Wind and waste heat
3. Part 3: Petroleum, plastic, and data centers
4. Part 4: How it will shake out and conclusion

PDF of the whole thing (2,158 words).

There are other ideas floating around, that might be of interest to private individuals, too. Wind turbines floating on neodymium magnets resulting in ultra-low coefficients of friction that generate electricity spring to mind. Even the smallest gust of wind could offset electricity costs for your home. Applied on a larger scale, these turbines could be installed next to stretches of highway where the wind created by vehicles speeding by generates power for the grid. The ideas get progressively more sci-fi and less based in reality, but all have research and/or working prototypes to support them.

Ultimately, I expect a handful of green power generation strategies to become prominent, based largely on a region's geographical needs. A landlocked country has little use for generators that harness tidal forces, and a country without large amounts of sunlight will have little use for a solar grid and might be better off with a network of small, personal, near-frictionless turbines to produce a great deal of power. This will have the secondary effect of changing power companies' dynamics. Private individuals and businesses may end up selling a significant fraction of the electricity that they generate back to the power company, as already happens on a tiny scale.

Rather than being a sunk, overhead cost, electricity could provide a smaller, secondary source of income in some cases. This is good for the overall health and robustness of the power grid itself. Rather than a centralized source vulnerable to operator error, equipment failure, or even an unlikely terrorist attack providing us with all of our power, a grid of consumers becomes a grid of hybrid supplier-producers. This has an effect on the dynamic of the producer-consumer relationship, too. The consumer has more power because they're doing more than just consuming. They become more of a partner in the relationship.

So while no single master stroke of technology is going to save the world from global climate change, or rescue our economy from its dependence on foreign oil, there are a number of initiatives that, in aggregate, are having a real, profound effect on our economy. That effect will only become more pronounced as time goes on, and these technologies that are mostly in the lab make their way slowly into the real world. It's important to note that while being green is trendy and gets a lot of press and has considerable mindshare, particularly among the youth, it's not this trendiness or mindshare that's going to create lasting change. It has certainly sparked social change, which is good, but as always, it is the bottom line that will be the driver for bigger and better things. It will be economic forces that determine whether we continue our destructive tendencies or move towards a more renewable future. My money is on green, because that's the direction the invisible hand is pushing us in. Green is, quite simply, how we do more with less, and create new markets while we make our way in that direction.

This is part one of a four part series:

1. Part 1: The Little Things
2. Part 2: Wind and waste heat
3. Part 3: Petroleum, plastic, and data centers
4. Part 4: How it will shake out and conclusion

PDF of the whole thing (2,158 words).

Plastic, another petroleum product, is a problem in the making as well both in terms of making more and recycling what we've already used. Recently, a 16 year old Canadian high school student conducted a series of experiments designed to isolate organisms that might degrade plastic bags. After collecting soil samples at a local landfill, he spent 3 months culturing them solely on a diet of polyethylene film strips. He narrowed it down to four types of bacteria, and grew each on agar plates, and discovered a new species of bacteria that eats plastic bags more ravenously than Pseudomonas, the only known plastic eater to that point. Burd found that only 0.01% of the microbes' body mass was released as carbon dioxide, allaying fears that his technique, if implemented on a wide scale could increase the amount of greenhouse gases released during recycling. It's estimated that these plastic bags will take between 50 and 1,000 years to break down on their own in a landfill. And microbes have been shown to do the opposite as well: taking toxic styrene and turning it into a biodegradable plastic called PHA. Both processes have economic implications, and each seems to be another tiny nail in Malthus' coffin.

There are other initiatives being worked on, primarily in academia, that will have huge implications for our business and environmental future. Generators that sit in the ocean or river and harness the power of tidal forces. There are some problems associated with this method of electrical production, such as how to store this energy meaningfully, but these problems have analogs with other types of green energy production, like wind power. With enough interest, investment, and work, they'll be solved.

Other ideas surrounding the harnessing of the oceans include thermal energy converters. It's thought that the amount of solar energy captured by the ocean is equivalent to 250 billion barrels of oil per day. That means that each day, the world's oceans capture the energy equivalent of 33 years worth of the US's total oil consumption. Obviously capturing the entirety of that energy is impossible and undesirable, but any company that comes up with a way of efficiently harnessing just a tiny fraction of it stands to make billions. Quite likely they will find themselves in an oligopolistic or even monopolistic position, too, as the barriers to entry will be huge, and the absolutely large minimum efficient scale of production will prevent new firms from entering.

"Green" thinking is also driving microscale R&D. Electronics companies are looking at solid state storage as a means of cutting down on power consumption in the datacenter. As more and more of our computing and storage moves to the "cloud", more datacenters are required. Datacenters are expensive to cool, and it's quite difficult to achieve an inexpensive, efficient, useful power density as well. That means that a large scale reduction in the amount of electricity consumed by individual server components will mean that useful power densities can be lower, or more servers can be crammed into a smaller space.

The largest consumers of electricity in the server are the pieces that move, specifically the hard drive. This movement has the secondary effect of creating waste heat which must be compensated for with adequate cooling lest the entire datacenter overheat. So even a small decrease in power consumption in that one tiny segment of that one specialized market will have domino effect across many secondary industries. It's similar in scale to the standby power dilemma mentioned above. And when you see companies like Microsoft, Amazon, and Google moving close to hydroelectric dams to build their datacenters, or moving to Siberia to save on cooling costs, you know these concerns aren't pie in the sky. There are real economic forces at work that are more powerful than the constraining forces associated with having to build fiber infrastructure out to these remote areas.

This is part two of a four part series:

1. Part 1: The Little Things
2. Part 2: Wind and waste heat
3. Part 3: Petroleum, plastic, and data centers
4. Part 4: How it will shake out and conclusion

PDF of the whole thing (2,158 words).

Being green makes good business sense, much of the time. While you obviously wouldn't want a hospital run directly on solar power, it does make sense to build solar arrays in the right places, and wind farms in perpetually windy areas, and then hook these up to the existing power grid. In that context, running the hospital on solar power doesn't seem like such a bad idea anymore. In medicine, we manage chronic pain by coupling a long-acting opioid with a short-acting, rapid-onset opioid. The long-acting agent is used to control baseline pain, and you never use short-acting opioids to manage baseline pain because of the greater duel risks of overdose and dependence. These agents are used to breakthrough needs only. In power generation, the metaphor is analogous: renewable resources provide your baseline power, and your coal- and oil-based electricity kicks in only when necessary. Thankfully, exothermic reactions lend themselves to relatively rapid cycling and are therefore suited to "as-needed" use.

Texas billionaire oilman T. Boone Pickens is seeing why it's valuable to invest in renewable sources of energy. Not only is it good for national security, but it makes good business sense to invest in renewables. With any non-infinite resource, the market is subject to the forces of supply and demand. When supply drops, the price goes up. If demand increases because India and China need their share of the world's petroleum supplies, prices for the US consumer go up, as well as the ancillary costs associated with anything that needs to be transported. As the amount of available petroleum decreases — as it's steadily and inevitably doing — these forces increasingly affect the way you operate your business. For a company like National Grid, eliminating the twin problems of scarcity and competitive bidding are good for the bottom line.

Civil and structural engineers and architects are hopping on the green bandwagon as well. The first of them jumped on because it was hip and different, and enabled them to leverage a different kind of brand image to achieve financial success. Lately, though, buildings that are built to be more energy efficient make economic sense. In Sweden, Jernhusen AB is harnessing the body heat of thousands of commuters that pass through Stockholm's main railway station. The firm believes that the system being designed can provide about 15% of the energy needed to heat the 13-story building being built next to Central Station. This system isn't even particularly radical. It's going to cost about $47,000, and will only require a few pumps and some pipes, since the ventilation system is already in place. I think it's a safe bet that a 15% annual energy savings for a 13-story building will more than cover even the short-term costs associated with it, particularly in a city only ~1,000 miles from the Arctic Circle like Stockholm.

Since every mechanical system wastes energy in the form of heat, recycling waste heat is also becoming more popular. Estimates of the amount of energy lost in the form of heat — expressed in terms of electricity — from smokestacks in the US alone is at 50,000 megawatts, more than half of what this country generates from its aging nuclear fleet. Initiatives to turn this waste heat directly into electricity are already underway, and can be built on small scales that make it worthwhile for these industrial companies to invest in.

This is part one of a four part series:

1. Part 1: The Little Things
2. Part 2: Wind and waste heat
3. Part 3: Petroleum, plastic, and data centers
4. Part 4: How it will shake out and conclusion

PDF of the whole thing (2,158 words).

Being "green" is a badge worn with honor by companies and individuals alike. Just like consuming organic foods, it's as much a yuppie status symbol as it is a lifestyle choice. At least it used to be. Venture capitalists are on the lookout for interesting green companies these days because there's lots of money to be made by reducing energy consumption, cutting back greenhouse gas emissions, and generally doing more with less. From a very high, forward-thinking level, it seems bizarre that we haven't been doing this right along: if economics is the study of human behavior in a world where there are limited resources but infinite demand, it makes sense that we would want to do more with less.

If it requires 50% less electricity to run an efficient datacenter, and 50% less gasoline to get from point A to point B, then that means we can house twice as many servers and end up with the same electric bill, and travel twice as far without burning any more gasoline. So every halving of the resources required to do something results in a net doubling of whatever you can do with that resource, all things being equal.

This model is easily applied to things like fuel efficiency and other commodity consumables, where small increases in efficiency result in immediate, apparent cost savings to an otherwise ignorant consumer. However there are other, less intuitive places to look where small savings aggregated across millions of people results in real macroeconomic benefit.

For example, there is a push right now to get rid of or improve "standby" modes for most electronic devices. This is what is widely considered the "off" position for most things, but in reality is actually the low-power mode wherein a device is not performing its primary function. The clock on the VCR and the microwave. The "breathing" the LED in your Macintosh computer does while it's asleep. Your laser printer while it's on standby waiting for a print job. Same for your fax machine.

A single device can suck up as much as 30W of electricity every 24 hours. Multiplied across all of the consumer electronics in your home, multiplied by the number of households in the United States, and you quickly realize that this is a boatload of wasted electricity. In fact, in the lifetime of a single electronic device, this savings is estimated to be $10. This is one of the reasons that President GW Bush directed the entire federal government to buy low-standby-power devices back in July of 2001 (PDF). Uncle Sam buys a lot of electronics. That means tens and possibly hundreds of millions of taxpayer dollars saved by one superficially insignificant initiative.

If I were a venture capitalist today, I wouldn't be looking at Internet startups. While the Internet is a sexy market and commands a lot of mindshare, I don't think that it's the future. We're coming to the end of the Information Age. No, the problems uncovered by the Information Age aren't solved. Search isn't solved. Scientific computing as a whole is still quite nascent.

But we are gradually working our way out of this Age and into the next: the Renewables Age. Just like we're still using what was developed during the Industrial Revolution, so too will we continue to use and develop the goods and services developed during this Information Age. So while computing and information management isn't going anywhere, it will be superseded by bigger economic concerns. Namely, renewable energy.

I firmly believe that a few well-placed, relatively modest investments today can very probably yield absurd returns on investment sometime down the road. It would, however, be a very long-run type of play, and many VCs aren't prepared to make an investment that won't pay off within ten years.

With that in mind, I wrote the following as part of a larger essay for an economic history class about a week ago. What is interesting is that when I started the paper, I had no idea where I would end up. Like most Americans, I hadn't thought about the "greening" of the economy at all. I didn't have a conclusion in mind when I began. The final result was this, and I think I make a pretty good case for renewables and investment therein. At the very least, I have convinced myself, and I don't really know how you can argue against it unless you're talking timespans of less than ten years.

I've broken it up into a couple of shorter pieces because it's simply too long to post as-is. No one would read it. Because it's unedited, the beginning of each piece might feel a little jarring beginning on day 2.

1. Part 1: The Little Things
2. Part 2: Wind and waste heat
3. Part 3: Petroleum, plastic, and data centers
4. Part 4: How it will shake out and conclusion

PDF of the whole thing (2,158 words).

Let me know what you think…

I've been wanting to learn MySQL and PHP for a little while now, because I think it'll be useful in the coming months, so I decided that I should probably install Linux, rather than installing the two on Windows. (There's something about that that just feels wrong.) I have something I'm itching to build, and it's time to start picking up some new skills again. I've had a spare 500GB drive lying around for a couple of months, so I popped it in on July 4, downloaded Ubuntu 8.04 Hardy Heron, and installed the OS almost as a spur-of-the-moment thing.

I was amazed at how quickly and painlessly the OS installed. Back in 2001-2002, the last time I made a serious go of running Linux (Debian), it took almost 18 hours to get the OS installed and configured to the point where I could use X. It took a lot of help from the guys in #linux, too. I never would have been able to do it without their help.

But this time it was as easy as putting the CD in, booting from the optical drive, entering my desired username, password, and localization settings, and that was it. About 45 minutes later, I booted into Gnome, and had working sound and networking, right out of the box. I ran the GUI Update Manager, which found a proprietary display driver for my GeForce 8800 GT — in addition to other normal system updates — and a single restart later, I had graphics acceleration and an up-to-date OS installation.

Total time to working install: ~60 minutes, including software updates. It was faster and easier than installing XP, Vista, or Mac OS X Leopard, and by a significant margin. In terms of ease of installation, I would rank them in this order:

1. Ubuntu 8.04
2. Windows Vista
3. Mac OS X
4. Windows XP

I also installed the Redhat Liberation Fonts following the instructions here, and pretty soon, I had non-crappy looking fonts. (You'll have to change your system fonts in both Firefox and Ubuntu itself to use them before you'll notice a real difference.) While they're still not as nice as Windows or OS X, they're quite a bit better than what you get OOTB with Ubuntu. And I'm not even sure that they're *bad* per se. I think they're just different than what I'm used to.

That ended my adventures in Linux for a few days until tonight. I needed to get a significant amount of schoolwork done which unfortunately requires Internet Explorer. I'm pretty well caught up there, so I've been booted back into Linux for the past two hours or so. In that time, I've gotten my laser printer working using the generic PCL6 drivers with Foomatic/pxlmono. Print quality is great, and I can even configure the printer to print as "draft" quality without messing around. I couldn't even do that with OS X.

My only complaint so far as that when the machine is woken from sleep or hibernate, there is no sound. Apparently this is a common problem with a couple of possible solutions, but I have not felt any inclination to try any of them yet. Maybe in another couple of days or so.

Hopefully I will be booted into Linux for another couple of days before having to go back to Windows for any reason. Alas, I am unable to ditch Windows entirely as I do some pretty intricate page layouts and PDF conversions from within Word, but hopefully I can minimize the amount of time I spend there.

To figure out for the future, should time and attention span allow:

• iTunes replacement that can import old iTunes metadata
• Figure out these page layouts and PDF conversion in OpenOffice, if possible
• Get my Blu-ray player working, which should be possible with mplayer and some other tools

My main focus right now, though, is building my project, which has some very serious monetization possibilities. Now if only I can get someone else to answer his phone.

I chuckle every now and again when I see the MSM reporting on blogs. The usual suspects almost always turn up: TechCrunch, the HuffPo, GigaOM, BuzzMachine — as well as a smattering of the hot blogs du jour. This time it was Steve Rubel's MicroPersuasion and Passive Aggressive Notes.

I must confess some incredulity, because I have never seen Ars Technica mentioned in a story that focuses specifically on blogs. This despite being relegated to merely a "blog" (albeit acknowledged as an influential one) most of the time by the mainstream media when they reference a story that Ars breaks.

Now, the HuffPo is a huge website. Probably a little bigger than Ars with 5.7M unique readers per month. TechCrunch is markedly smaller, and GigaOM is smaller still (1.37M pageviews/month or so).

It makes me wonder why these particular blogs are chosen. Is it because the stories about blogs are by their nature more noise than substance? Indeed these stories are often widely hyped when they hit and will make their way around the 'sphere several times before disappearing like yesterday's newspaper. (The blogosphere echochamber at its finest.) Ars seems to be anti-hype most of the time. It's been known to take a somewhat dim and sometimes even contrarian view to what's hot in the blogosphere this week — "The Cloud!", death by blogging! — if indeed what the blogosphere is focusing on this week is even worth talking about at all. (Usually it's not.)

So here are some sites that Business Week may want to think about including, because these sites are the real movers and shakers in the Internet publishing world. This list is by no means comprehensive, and I make no comments about their content or quality of the sites, only their size. This list isn't sorted in any meaningful way:

• High-Def digest: ~7.23M
• Destructoid: ~4.25M
• Gamersyde: ~5.17M
• SB Nation: ~8.76M
• Celebrity Baby Blog: ~8.64M
• Confessions of a Pioneer Woman: ~6.05M
• Fark: ~37.09M
• BoingBoing: ~6.93M
• SeriousWheels: ~9.05M

For comparison, TechCrunch sits at ~7.5M pageviews per month, and Ars Technica sits at ~30M.