davidad

Kilowatt hours

Wed, 28 Jul 2010 12:51:28 -0400

A lot of people don’t seem to understand the units for energy and power. Here’s a quick tutorial.

The metric system unit of energy is the joule. A joule is about a quarter of a calorie, which is defined as the energy required to heat one gram of water by one degree centigrade (1.8 degrees Fahrenheit). A joule is also about how much energy you expend by lifting an apple from the floor onto a table. It’s defined as the exertion of a force of one newton over a distance of one meter. Now, that energy can be delivered quickly or slowly, and it’s the same amount of energy, but if it’s quicker, it’s more power. The SI unit of power is the watt, which is defined as a delivery of one joule every second.

Power * time = energy Energy / time = power

Now, for some reason, utitilies have chosen to meter energy in kilowatt hours, rather than megajoules (1 kWh = 3.6 MJ). This seems to cause no end of confusion. This post was prompted by someone, upon being told that a system delivered 100kW, who clarified “you mean 100 kW per hour?” This was the third person I had heard reference kilowatts per hour, and two of them were supposed to be energy experts. Kilowatts per hour does not make sense as a unit. Kilowatts are already units of energy per unit time (kilojoules per second). You can multiply kilowatts by hours to get kilowatt hours, a unit of energy, but that’s almost certainly not what you meant if you said kilowatts per hour. Now, you could define a unit of power based on kilowatt hours per second, wherein the hours/seconds divides out to a dimensionless factor of 3600, but this is rare.

What are these units for? Well, the capacity of a power source, or the demand of a power load, is measured in power - watts or megawatts, joules per second or megajoules per second. But cost is measured based on energy - dollars per megajoule or dollars per kilowatt hour. This is because it costs more to run a system for longer, but it doesn’t affect the system’s capacity.

Any questions? If you’re still confused (or more confused than you were…) please ask.

Template of nanotechnology talks

Fri, 16 Jul 2010 19:56:10 -0400

Ralph Merkle’s “secret of giving nanotech talks in any field”:

[Field] is critically dependent on [products]. [Products] are made from atoms. Nanotechnology will let us make [products] that are stronger, smarter, faster, lighter, cheaper, and just better. This will have a huge impact on [field]; for example, we could even make products that are [astonishing parameter] and cost [remarkably little]!

I managed to squeeze in a bit of recording at #SingularityU,...

Sat, 26 Jun 2010 03:27:33 -0400

I managed to squeeze in a bit of recording at #SingularityU, under the auspices of an audio aid to abating audience anarchy as each lecture session begins. Enjoy…

Live from Singularity University

Thu, 24 Jun 2010 04:47:07 -0400

Live from Singularity University:

I’ve been using Twitter to take notes of the lectures given at SU for the past couple of days. A lot of interesting ideas are being presented, and I’m doing my best to convey them in 140-character format (actually more like 110-character format including the hashtag and the speaker’s name). If you’re interested you should also do a search on Twitter for #SingularityU to get some other perspectives as well.

Machine Learning & AI

Sun, 20 Jun 2010 11:33:19 -0400

A fellow SU student asked me about the difference between machine learning and artificial intelligence last night, and I thought I would summarize my answer for the record, in case other people are wondering, or might wish to correct me.

Machine learning is about solving small problems to within reasonable tolerances. Artificial intelligence is about trying to solve big problems (or all problems at once).
Machine learning tends to be handle numbers, while artificial intelligence tends to handle symbols or words.
Machine learning algorithms tend to give you little insight into how they work, while artificial intelligence systems can often be queried about their “train of thought.”
Machine learning techniques, especially in recent years, can often be justified by mathematical proofs. Artificial intelligence schemes generally cannot.

A cute video illustrating the classic...

Fri, 04 Jun 2010 19:24:48 -0400

A cute video illustrating the classic teleporter-that-destroys-the-original problem.

A solution to videoconference gaze?

Wed, 02 Jun 2010 17:49:54 -0400

One of the common problems with videoconferencing today is that it’s impossible (with current technology) to make the camera and the display totally coaxial, so you get a somewhat disconcerting lack of eye contact. What if you put one camera on either side of the display and used an algorithm like this to synthesize the view from a “virtual camera” aligned with the display?

Stephen Fry is brilliant.

Tue, 01 Jun 2010 23:36:56 -0400

Stephen Fry is brilliant.

Acapella, with reverb/EQ/mixing. No words this time. Unless you...

Mon, 31 May 2010 14:44:06 -0400

Acapella, with reverb/EQ/mixing. No words this time. Unless you count “bow”, “oo”, and “hey”, that is.

"James Clerk Maxwell might be spinning in his grave. But is he rotating around his long axis, or..."

Mon, 31 May 2010 09:52:28 -0400

“James Clerk Maxwell might be spinning in his grave. But is he rotating around his long axis, or flipping endwise?!!!”

- William Beaty, in a thought-provoking article about magnetic circuits

HyperCard: arguably the greatest piece of software ever...

Sun, 30 May 2010 15:26:00 -0400

HyperCard: arguably the greatest piece of software ever produced, for any platform.

"Many people with jobs have a fantasy about all the amazing things they would do if they didn’t need..."

Sun, 30 May 2010 12:13:00 -0400

“Many people with jobs have a fantasy about all the amazing things they would do if they didn’t need to work. In reality, if they had the drive and commitment to do actually do those things, they wouldn’t let a job get in the way.”

- Paul Buchheit (via wizardry)

Benford's Law

Sun, 30 May 2010 00:30:00 -0400

Benford's Law:

is the observation that most “real-world” data is not uniformly distributed - in fact, the probability that the first digit is a “1” is almost one-third, rather than the expected one-ninth. The reason for this is that the *logarithm* of most real-world data is uniformly distributed. I find this mildly fascinating.

"On the airplane of science, nontrivial explanations are not the beverage cart or the in-flight movie..."

Thu, 27 May 2010 09:42:10 -0400

“On the airplane of science, nontrivial explanations are not the beverage cart or the in-flight movie — they’re the wings.”

- Scott Aaronson

Jurassic Park

Sat, 22 May 2010 18:54:56 -0400

Jurassic Park:

It’s only a matter of time now.

Summarize Me

Thu, 20 May 2010 14:00:00 -0400

I may be designing a separate davidad.net from blog.davidad.net shortly, as a landing page for those who don’t already know me. However, I’m not exactly sure what I’d say. Thus, I turn to you, people who (presumably) are convinced that I’m interesting. How would you describe me to someone who clicked or Googled my name and doesn’t know anything else about me?

Last night I recorded some unoriginal guitar stuff. It’s...

Thu, 20 May 2010 10:00:00 -0400

Last night I recorded some unoriginal guitar stuff. It’s not very good, but I want to publish more recordings before I go to CA for the summer and probably lose recording capabilities, so have at.

Today’s attempt at song! I still fail at writing words, so...

Wed, 19 May 2010 00:28:21 -0400

Today’s attempt at song! I still fail at writing words, so these words are totally improvised, and may or may not make sense. But I hope you will agree that the piano chords, at least, are superb. :-)

Loop Perforation

Thu, 13 May 2010 22:47:37 -0400

Loop Perforation:

Martin Rinard’s group at MIT has achieved significant peformance gains in a huge variety of applications by a very simple optimization: simply skip every other iteration of certain loops. This is a brilliant idea, but it’s also a symptom of huge problems in computer science. We should always have the option be more flexible about how we process information in artificial systems. Computer science is culturally opposed to this; certain philosophers as well as much of the lay public presume this to be a technical limitation (that is to say, a fundamental limitation of technology as opposed to humans), but it’s not.

"Says the frequentist, “You’re a Bayesian? I’m going to kick you in the posterior!"

Thu, 13 May 2010 02:49:25 -0400

“Says the frequentist, “You’re a Bayesian? I’m going to kick you in the posterior!””

- Dustin Smith