Bard's new Citizen Science Program

You can see it in the dust up over climate change or vaccines, most people just don't understand what science is, how it works, and how it's different from anecdotal evidence.  This translates to some very real world policy issues.

Bard College, in my backyard, is creating a new program to try to address this:

But to many of the college’s faculty, and to Leon Botstein, who has been the college’s president since 1975, there was still something missing: a true introduction to science and scientific thinking for the vast majority of Bard’s students. “People who graduate in fields other than science often do not understand science,” he says. “They do not know what the limits of science are and what science can do. It’s catastrophic.”

Next January, Bard’s science and math faculty – along with postdoctoral students and faculty from other institutions -- will try to change all that with the Citizen Science Program, three weeks of science learning modeled on the success of Language and Thinking. Also required of all 500 of the college’s freshmen, and ungraded, Botstein hopes it will become similarly entrenched as a landmark of students’ first year at Bard.

This is going to be 3 weeks of quite intensive education.  From the curriculum:

This program will merge three distinct, yet thematically interwoven week-long rotations, each designed to address the overarching question How can we reduce the global burden
of infectious disease?
In one rotation, you will focus on the concept of laboratory experimentation by exploring the question How can infections be treated? This rotation will be spent in the laboratories of the Gabrielle H. Reem and Herbert J. Kayden Center for Science and Computation. You will get hands-on experience exploring how antibiotic resistance develops in bacterial strains and how DNA is moved from bacterium to bacterium (similar to how antibiotic resistance in Staphylococcus aureus produced the “MRSA” strains so often reported in the news).

The second rotation will focus on the question What factor best explains a person’s probability of exposure to disease? The disease we will study is tuberculosis. You will look at a number of factors, including the science of Mycobacterium (the organism that causes tuberculosis), risk factors that can cause a tuberculosis infection to become worse (including HIV infection), and global locations where the tuberculosis burden is heavy and light. We will discuss what can be done to alter the number of infectious cases within these different environments.

The third rotation will focus on the spread of infection by exploring the question What intervention—such as vaccination, treatment, reduced exposure—is the most effective at reducing transmission of an infection? You will learn to use state-of-the-art computer simulations to help you understand and observe how an infectious disease can spread throughout a community, and how different treatment options can be effective in limiting disease.

I especially appreciate their focus on a real world problem, then using science to explore facets on how to address it.  It's an inspired approach, and I wish them the best of luck in implementing it.

The 4th Amendment and the Cloud

Ars Technical has a great write up of some of the uncertainty of the 4th amendment (protection against unreasonable search and seizure) when it comes to your data in the cloud.  The first test case for the Supreme Court of this is actually going to be around text messages.

But what may be just as telling as the forthcoming Quon decision itself is the way in which the Justices attempted to wrap their heads around the technology during oral arguments. For example, Justice Anthony Kennedy asked what happens if a text is sent at the same time that one is being received. Justice Antonin Scalia and Chief Justice John Roberts expressed surprise that the text was routed through the service provider and did not go directly from person to person.

"Could Quon print these—these spicy conversations out and circulate them among his buddies?" Scalia also asked. And even the lawyer representing Officer Quon was not sure whether deleting a message on a pager would also delete it from the service provider's records. This might paint a worrisome picture of a judiciary that will be making important technology-related decisions with only a limited understanding of that technology.

It's also interesting that protection to voice communication wasn't extended until 1967 when a court case around wiretapping phone booths came to the Supreme Court.

Unreasonable expectations

Yesterday I learned something about the new Ubuntu 10.04 release.  I doesn't like to start correctly when applied to a 8 year computer, that has a 11 year old second video card sitting in it.

This is the desktop under my desk in the office that I largely use as a media streamer, irc proxy, and second display (via synergy) for reading documentation while I write code.  It turns out that some number of years ago I'd stuck my old 3dfx card (a company that when bankrupt in 2002) in there, I think when I was testing Linux DRI on that machine and the i815 driver wasn't working correctly.

Honestly, I'd forgotten the card was in there for years.  But when the machine did funny things to the display during boot (just about the time it's trying to switch to it's fancy splash screen), I spun it around to play with cables and noticed the second vga port.  Removing the 3dfx Voodoo3 pci card and rebooting fixed all the issues.

Lesson learned: don't leave hardware you don't ever plan to use plugged into machines.

How to put the window buttons back to the right on Ubuntu 10.04

Ubuntu 10.04 introduces a change on the default placement of the close, minimize, and maximize buttons on windows.  This is a clone of where the Mac folks have had them over recent years, and a design that I remember originally back in Beos days.

I can understand that Mark wanted a more Mac like experience with 10.04, as Ubuntu seems looking past MS Windows parity and towards Mac OS parity at this point.  However, for those of us that are straight up Linux nuts through and through, and could give a damn about what The Steve thinks about close buttons, it's actually quite easy to put the buttons back on the right side of the window.

First, get to a terminal and run gconf-editor.

Next, navigate to /apps/metacity/general/button_layout.  You just need to change this key to be ":function1,function2,function3".  If you are using the Human or Humanity themes the correct order would be ":minimize,maximize,close".  If you'd like to use the new Ambiance theme ":maximize,minimize,close" will look better, because the beveling is actually just a trick of the images, it's not intrinsic to the button area.

See, it wasn't that hard, so now people can get back to real work instead of flipping out that their buttons are in the wrong places. 🙂

Hudson Valley Earth Day... and a new kind of green

About the time that Susan and I started dating, Susan when to a book signing with a friend for Dare to Repair.  It's a basic manual of home repair written to a target audience of the recently divorced woman who'd always let her husband do all the handy work.  The book was a modern day women's lib, as there is a certain amount of independence, mental and financial, that comes from being able to fix things.  The contents of the book were pretty basic, but given its target audience, it probably hit the right balance.

More importantly it raised the real issue, people shouldn't be afraid of things that are broken.

I = V / R

The repair skills of a generation ago don't really prepare us for today's world.  A modern car has 100 million lines of source code.  Most of your home appliances are complex computers.  The glowing LEDs that surround you should be a constant reminder of just how much of our world is built on the IC: the Integrated Circuit.

A few months ago, due to some encouragement from the Squid Wrench meetings, I decided it was time to dive back into my college electronics and see what I could relearn.  Susan was originally skeptical, as this was a new hobby with a few hundred dollars worth of startup costs.  As soon as I got my new soldering iron I dug in to fixing some LED light strands that had a few wires sheered.  One strand was successful, the other way not.  A decent start, but not great.

The real success came the following weekend when I decided to crack open a chime alarm clock that Susan had bought a few years ago, which stopped working 6 months later.  It had been laying on a shelf for years.  After some poking and prodding, and adding insulation between a couple of wires, it came back to life.  The clock, new, is more than a hundred dollars, so my investments were starting to show returns.  Later that day I managed to fix a 5 year dead Shruti Box, which was another hundred dollars recovered.

None of these repairs required stellar amounts of skill, but they required determination to open up the black box and try to figure out what was going on on the inside.  It is a determination that is being bred out of us as our everyday devices are showing up hermetically sealed from vendors that just want us to buy a new one in 12 months and throw the old one out.

A new kind of green

Last year our local fairgrounds had a Green Fair and Expo in April, which we went to.  It looked a lot like a home show with slightly more discussion of solar and geothermal.  Of the hundred or so vendors there I found about 3 that I cared to talk to.  This year, it was a very different event.  It was billed as an Earth Day celebration, and the green fair aspects of marketing slowly evaporated as we got closer to the event.  There were hardly any vendors, a lot of the space was given up to musicians and lectures.  Space wise it was poorly set up, and you could see that they got a lot fewer booths than they'd expected.

However, there was one thing that was really inspiring.  There was a table set up for the Hudson Valley Materials Exchange: "a non-profit community warehouse for materials rescued from the business wastestream".  You could buy various wood, metal, plastic and textiles that were the intermediate stages of some manufacturing.  Stuff that would otherwise end up in a landfill.  This group has apparently been around for a decade.

It also spawned a giant swappapalooze area in a separate tent where kids were building all kinds of things with wood, cloth, glue, and anything else they could get their hands on.  That's an inspiring thing to see now adays.  No kids hanging out in the corner with Nintento DSs, just going after wood with a saw.  This idea of repairing what we typically would throw out, and buidling more of what we need is definitely a new kind of green.

What if Earth Day became an anchor event in the Maker Community?  Wouldn't that be exciting.

RIP Carl Macek

Carl Macek was the TV producer that brought American kids Robotech in 1985.  Robotech was spawned out of the Anime series Macross, but because Macross was only 36 episodes, and you needed 80 to get a daily syndication run in the US, he came up with this crazy idea to mash it up with 2 additional series that had giant robots, do a little re-editting, and make an over arching 3 generation story arc.  It sounds crazy, but it really did work.

Kids TV in the mid 80s was thinly veiled toy commercials, however because Robotech's product partner was a model company, they were able to shoot for a higher level of sophistication with an 11 - 14 target audience.  It meant that Robotech had a real plot, real emotions, real conflict, in a way that nothing else did at the time.

I have fond memories of waking up early on Saturdays and Sundays to watch it.  In college I managed to get video tapes of the show from my friend Julie, who was 3 years older, so had the sophistication to actually record these things when she was a kid.  Later, when Harmony Gold finally rereleased Robotech on DVD I managed to do what most people can't, buy a little piece of my childhood back.  I still love the series to this day.

Carl, thank you for what you created.

Instead of a moment of silence, I'll leave you with the opening credits from Robotech as the best tribute that we could give to Carl.

What it means to be a force of nature

Flights are still grounded in Europe due to the Eyjafjallajokull volcano, echoing back to the 3 days of air space closure in the United States after September 11, 2001.  But unlike that event, relaunching the planes isn't a matter of adding locks to airplane doors, or hiring thousands of unskilled workers to make you take your shoes off before getting on planes.  Instead, we just have to wait for the ash to clear, and hope that the volcano doesn't erupt again.

The last time this volcano was active was in 1821, and it continued to have on and off again eruptions for 2 years.  Just think of that for a minute: what would a world be like where Europe was a no fly zone for a year?

Rambling thoughts on C# on Linux

Livnat Peer just posted an interesting look at converting a large source base from C# to Java.  This was done because when Red Hat aquired the company that wrote KVM, they also got a huge .NET management application that they wanted to run on Linux.  It's a pretty interesting look at the various approaches you could take, and how they were eventually successful.

C# on Linux is an interesting beast.  I like C# better syntactically than Java.  Properties are just too damn useful.  Having to have lots of getFoo(), setFoo() in Java when we've got this perfectly good key on our keyboard '=' that everyone has known about since they were 7 bugs me architecturally.  It is a short coming that Java will probably never get past.

Mono, the open source C# runtime, was the only open source Just In Time Compiler (JIT) you could get your hands on a few years ago.  That made it a huge boon to language implementers, and was the defacto runtime that people would play with and hack on to build scripting engines inside over other applications.  It's the reason you'll see Mono specifically show up all over the place in the gaming industry.  Since that time Java went open source, under GPL, and LLVM, which is under a very permissive license, really grew up.  This gave developers interested in language design some options for VMs they could run on top of.

But, there is always another hand.  Microsoft casts a long shadow over C# on Linux.  The Mono project remains many years behind Microsoft on features, and many more years behind that on stability and performance.  While I was working on OpenSim, I was continuously frustrated by how much worse the environment performed on Linux than on Windows.  Any project that is written in C# will be relatively poor performing on Linux.  The word relative is in reference to the same code on Microsoft .NET, it's still 20 times faster than if it was in Python.  Microsoft's sword rattling over Linux infringing their IP ensured that the Mono community remained somewhat small and close nit, with no large organizations investing in it other than Novell.

Mono makes for some decent desktop Applications.  I use three of them on a regular basis: F-Spot, Tomboy, and Do.  I can't function on a computer without Do any more.  But I still have a personal grudge with Mono over a simple fact: I can't watch Netflix Instant on Linux.  There was this theory that because of the way the media framework worked that it was going to work "real soon".  That was 3 years ago... and I'm still waiting.

C# has the basic issue that Java had for a long time, it's a vendor language.  And that's just a tough thing to really believe in, unless you have a sufficient reality distortion field.  Java has finally transcended that.  It took building a community process for future features and open sourcing the JVM.  Google's entirely parallel Java implementation for Android was additional proof that it's no longer in the hands of a single vendor.  And while Java remains far from perfect, if you are on Linux, and want performance, it's a pretty decent approach.

Fast food statistical lies

Rafe over at rc3.org provides a very important bit of fact checking to the fast food infographic that is circulating.  I thought 3800 seemed really high, but I didn't bother with digging deeper.  He did.  I'm coopting his post in it's entirety (it's short) for those who aren't reading his blog.

I’ve seen a number of links to the Everything You Need to Know About Fast Food infographic, which is chock full of interesting statistics, not all of which I’m entirely sure are accurate. It lists the average caloric intake for Americans as 3,760 calories, but I am pretty sure that is impossible.

This statistic comes from the UN Food and Agriculture Organization, but unfortunately the link is dead. Other sources report that the average caloric consumption for men in the US is around 2,618 calories for men and 1,877 calories for women — significantly less. A little back of the envelope math shows that the number cited in the infographic is impossible. The average height of men in the US is 5′ 9.5″ and the average weight is 191 pounds. The resting metabolic rate for 25 year old men of average height and weight is about 1900 calories. If you eat only 100 more calories a day than you expend, you’ll gain 10 pounds per year.

I’ve seen this number being repeated a lot, and I wanted to point out that there’s absolutely no way it’s correct. I haven’t seen the original source data, so maybe it’s a valid statistic that’s being misused, but in the context people are using it, it’s wrong.