I manage a small private forum, and I’ve noticed that sometimes it slows to a stop. In fact, something is hammering on my server and running so much ASP script that my whole website becomes slow. Then I figured out why:

Yahoo Search Robot! It was logged into my forum 11 times. I turned off guest read access, and everything instantly became normal and speedy again. But the Yahoo search server was going nuts. 20 guests (as seen above). Then 30 guests! All getting access denied.

I also turned on robot exclusion on my site (robots.txt and the META tags), but this is not stopping the Yahoo Search Robot from trying to crawl my forum in the most obnoxious possible manner.

Many libraries have found they can save space and make information more available by scanning their vast holdings of journals and conference proceedings, converting the result into PDF formated files. That’s good news for people doing research on-line. But here is a problem:

This particular journal article contained a photograph, but when journals are quickly scanned and compressed, the result is the almost complete destruction of the image data. This is actually a pretty good example. It’s not unusual for photos in a paper to be reduced to a black rectangle with some white blobs.

Here are a couple more examples. This is bad news, since as Adobe’s advertising campaign says, libraries can “Pitch the Paper!”. Let us hope that we do not find one day that some important historical images are completely lost in this process.

Of course, PDF is capable of storing images at higher resolution. But it takes time and care to scan a paper well enough to preserve a good record of a photograph. In practice, it just doesn’t happen — thousands of journals have to be processed, using default settings that compress the data well but ruin the images.

Personally, I wish scientists would use rich text format instead of PDF. I’d like to be able to read a science paper, click on a graph and see a spreadsheet of the original data pop up, as it does with MS Word. But the ability to embed original data and images in a paper is rarely used.

Two recent questions strangely converge, what is a good rocket fuel and what is a good battery for a laptop PC. It got me interested in how much energy can be stored in a given mass.

At Bell Labs, I remember a great talk given in 1985 by the head of the Physics department. He talked about the exponential growth in network bandwidth, memory density and computing speed. But he noted, battery lifetime is limited by chemistry and is only asyptotically approaching a modest limit. The latest seems to be the Lithium Ion Polymer battery, which is finding use in mobile devices. The recent stories of exploding laptop PCs are a clue to just how much the energy density of batteries is increasing.


Let’s look at the specific energy density (kiloJoules per kilogram) of some technologies:

Ultracapacitor ............... 50 kJ/kg
Lead Acid Battery ........... 100 kJ/kg
Ni-Cd Battery ............... 200 kJ/kg
Alkaline-Manganese .......... 300 kJ/kg
Lithium ion Polymer ......... 600 kJ/kg
TNT ....................... 4,000 kJ/kg
LOX/LH2 .................. 13,000 kJ/kg
Nuclear Fission .. 90,000,000,000 kJ/kg

We can see that chemical batteries still have a long ways to go before approaching the energy density of chemical fuels and explosives. Liquid oxygen and liquid hydrogen (LOX/LH2) is the about as good as it gets from chemical energy density, which is why they were chosen as the ideal rocket fuel by K.E. Tsiolkovsky over a century ago. And wouldn’t it to use a laptop for one week instead of just six hours, per charge?


Ultra capacitors are interesting. Although they are just now catching up to batteries, in theory they might someday rival the density of any chemical energy source. One advantage they already have over batteries is almost infinite lifetime. The D-cell compatable capaciters seen above are rated for 500,000 duty cycles.

When I show people pictures of Venus, I get two very common and somewhat perverse questions. OK, I’ll give them a shot:

What is left of the Venera landers today?

Venera-9 style landers were built mostly from titanium and ceramic. So structurally, they are probably still sitting there. Of course, internal electronics (plastic insulations, circuit boards, etc) would be burned up. It is possible that liquid in the large chemical batteries might have boiled and ruptured the main spherical pressure hull, but only if the internal pressure could exceed 100 atmospheres. I’m not sure it could do.

What would happen if I stepped out onto Venus?

Very bad idea. The surface temperature (870 F) is about that of an oven on self-cleaning mode. The high density (1/10th of water) would greatly increase thermal transfer, like a liquid would; however, the atmosphere is still a gas. Supercritical carbon dioxide would not support rapid oxidation (flaming combustion), but it does act as a strong non-polar solvent, sometimes used for dry-cleaning. So I believe it would have the effect of being french-fried in mid air, and quickly reduced to a charcoal figure. A lurid motion-picture special effect waiting to happen. 🙂

Space exploration is a magnet for crank science. It’s nearly impossible to talk about something like intersteller propulsion and keep people on the same page as real-world physics and engineering. And it’s even more difficult to talk about far-reaching ideas like colonizing planets without drifting into the realm of science fiction. But here I go anyway.

Consider the famous scenes in 2001, where a NASA official flies to a beautiful space station operated by Pan Am airlines and then on to a Lunar colony. You’re looking at a simulated trillion dollar infrastructure, but why was it built? Who is using it? Who is paying for it? How does it make money? What are people doing on the Moon that is worth all this? These are issues that science fiction simply overlooks.

As in 2001, the analogy is often drawn between the airline industry and a future spaceflight industry. The difference is, on the Earth there are real destinations to fly to. There are countless social and economic reasons to travel from one populated region to another on the Earth. This is not the same as spending billions of dollars to fly to Mars, pick up a rock and return to Earth. For spaceflight to be practical and large-scale, there must be a reason, there must be a destination.

People talk about things like mining helium-3 on the Moon. Both technically and economically that’s nonsense. At present, there is nothing remotely valuable enough to pay for the cost of mining and interplanetary transport. But more importantly, these ideas represents a fundamental misconception about wealth, in the sense defined by Adam Smith. Real estate is valuable because people want to live there and work there. Human activity is the true definition of wealth, and human presence is what makes a destination interesting.

Thus, colonizing space is a bootstrapping problem. it is a problem in economics, not engineering. If Mars had an atmosphere and a population, it would be of incalculable value, and people would pay to travel there and back. But how do reach that point? The technology of cheaper travel and terriforming Mars is fascinating to speculate about. I believe it could be done almost entirely with robotic technology. But that is not what blocks us from proceeding. The real problem is developing a mechanism for funding, when there is a huge return on investment but a turnaround time of centuries. You would have to create a Martian Futures Market that people have genuine confidence in — a serious enterprise that makes steady progress, backed by corporations with proven expertise and probably at least one first-world government.

Maybe you have to engage people’s territorial and competative instincts. Let’s say America declared that it was going to unilaterally colonize Mars and annex it? After the obligatory student protest marches all over the world, I believe other nations might start a competing program! And then it’s hard for anyone to back down. If both programs make enough progress, investors will want them to merge and cooperate eventually. It is just too expensive to duplicate the effort.

Apocolocyntosis – To turn into a pumpkin. Dating back to Roman times, it was considered a humorous play on the word “apotheosis” (to be deified).

Bezoar – A human hairball, believed to be a universal antidote to poison.

Cenogenesis – The temporary formation of gills and other atavistic features in the human fetus, leading to the adage “Microgeny recapitulates phylogeny”, development of the creature plays out the evolution of its species.

Fusel Oil – Contaminants in alchoholic beverages responsible for hangovers, consisting of higher aldehydes and ketones.

Hagiography – The biography of a saint. 19th century occultist Aleister Crowley wrote a book he described as an autohagiography.

Lysenkoism – Unsupported scientific theory promoted for ideological reasons, named after the Soviet biologist empowered by Stalin who attempted to suppress the theory of evolution in the USSR.

Oubliette – A dungeon cell with a trap door in the ceiling.

Revenant – A person who has returned from the dead.

Trilithon – A pair of tall stones supporting a horizontal lintel, such as at Stonehenge.

Wergild – Money paid to a family as compensation and penalty for murdering one of them.

Zoarium – A super organism formed from the combination of individual creatures.

Metal dyes are based on transition metals, whose inner electrons happen to resonate at frequencies in the visible-spectrum of light. From left to right are seen:

Nickel Amonium Sulphate

Copper Sulphate (“Blue Vitriol”)

Potassium Permanganate

Cobalt Glass

I wrote a program in C++ to calculate interplanetary trajectories. It looks up the planetary positions in the JPL ephemeris database, and calculates the Keplerian orbital elements of a probe trajectory by Lambert’s theorem. So, given the departure time and arrival time, I can find the unique orbit that fits those two points at those two times. Here is the trajectory of the Soviet probe Venera-1.

I also compared distance values with Soviet ranging data. The first three ranges were measured by the CWFM radar data, measured from the Pluton system in the Crimea. The rest were calculated by their ballistics center.

Russian Value My Calculation
Feb 12, 03:45 GMT 26000 km 23269 km (1st telemetry session)
Feb 12, 11:25 GMT 165000 km 142001 km (2nd session)
Feb 17, 11:35 GMT 1889104 km 1889500 km (3rd session)

Mar 4 00:00 GMT 6.9 million 6953121 km
Mar 25 00:00 GMT 15 million 15243053 km
Apr 13 00:00 GMT 28 million 27796154 km

I believe my early numbers would be better if I was accounting for the Earth’s gravity. The probe would be launched faster than I calculate, but then slow down from the pull of gravity until it assymptotically approached the Lambert orbit.

When I switched from UNIX to Windows ten years ago, I became a big fan of Visual C++. I was happy to see today that Microsoft is making an Express version of the tool available for free, forever (it was originally intended to be free for a one-year trial period).

http://msdn.microsoft.com/vstudio/express/

They also are giving away a free version of SQL Server, which is good news for small websites. I’ve used the full version of SQL Server, and it is also remarkably easy to set up. And along with DB2 and Oracle, it is a serious product, not a slow buggy ersatz database system (you know who I’m talking about).

I was pleased to see that Daffodil has finally run the TPC-C benchmark on some of the free database systems ona 1.7 GHz PC. This is something the makers of these products have resisted for years, although until recently they have not really had the feature capabilities:

PostgreSQL – 62 transactions per minute
MySQL 5.0 – 483 tpm

On the TPC website, let’s see what the big boys do. On PC workstations:

MS SQL Server, 2.8 GHz Dell – 38,622 tpm
DB2 on HP Proliant – 18,661 tpm

And of course the professional systems are tuned for multiprocessor systems, where they go absolutely nuts:

IBM DB2 – 3,210,540 tpm
Oracle 10g – 1,601,784 tpm
MS SQL Server – 1,231,433 tpm


Over the years, I’ve been accumulating a small database of color spectrums, as part of my C++ graphics library. I’ve been tinkering with a solar-system renderer/simulator lately, and I decided to work on calibrating planetary images. NASA is notoriously bad about punching up the color of planets to make them look more “interesting”.

So what is the average color of the planets? Easily answered with a little digging, because the spectral reflectivity (albedo) has been measured carefully for all of them. Take that, multiply it by the spectrum of the Sun, convert to CIE XYZ, and then convert to 24-bit sRGB, a gamma-corrected color format that computer monitors and HDTV’s are calibrated to display. I scaled XYZ to make Y equal to the total geometric albedo of the planet (but I am not including the effect of distance from the Sun).

I have never seen a correct image of Venus, which is usually an ultraviolet imaged colorized orange or blue or even purple. It fact, to the naked eye, Venus is cream colored and featureless. The Earth is somewhat violet because of the blending of blue with the reddish colored land. And Mars is more brownish-orange than red. Overall, the real palette of the planets is more subtle than usually portrayed.

I added a second set of colors for major moons and the three major asteroid types. Note how dark carbonaceous objects like Phobos and Hygiea are — a very primitive form of matter.