The Kindness of Internet Acquaintances
The tail end of last Winter term found me getting muscle spasms in my neck and shoulder, and eventually landed me in the emergency room at 3 am. I'm two-thirds of the way through grading the first formal lab report for my class this term, and I'm pretty sure I've identified the proximate cause of those muscle spasms, because they're back. It's the combination of spending long periods typing at my office computer (I collect and grade lab reports electronically), and occasionally hunching forward to bang my head on the edge of the desk when confronted with a particularly egregious sentence.
I've still got six more reports to grade, so I shouldn't spend much time typing on my blog. Happily, people I don't know well at all have done me a huge favor by writing what I would've written on a couple of topics of interest to me.
John Scalzi on Intelligent Design and Michael Behe's testimony in Dover:
It's embarrassing. It's embarrassing for Behe, who claims to be a scientist. It's embarrassing for Behe's employers (who have been forced to acknowledge the embarrassment Behe causes them on a regular basis by posting a disclaimer on their web site), and it's embarrassing for anyone who likes to imagine that science should actually be about science, and not about comforting people twitchy about the fact they share a common ancestor with whatever animal it is they like the least. It's not embarrassing for those people, of course, but the fact it's not makes me embarrassed for them. I think it would be ashamed to go through life so afraid of ideas that I'd be willing to force ignorance on others to make myself feel happy and safe. Seems a little selfish, and a lot sad.
EvolutionBlog on liberal education:
Education isn't primarily about facts or job training. It's about exposing yourself to all of the things human beings have been up to for the last few thousand years. You read the works of the ancient Greek playwrights not because you really care about their nifty plots, but because by reading those works you immediately realize that the concerns of people thousands of years ago are pretty much the same as their concerns now. You read Dickens or Shakepeare or Hemingway (or Agatha Christie or Stephen King (yes, they belong in the canon too!)) because by doing so you appreciate for a moment what the English language can be made to do. You learn science partly for the specific facts you learn (you really ought to know that the Earth orbits the Sun an not vice versa), but mainly so that you can marvel for a moment at the sheer ingenuity, persistence, hard work and cleverness that went into figuring all this stuff out. You learn history not just because you should know when the Civil War was fought or what the Mayflower compact was, but because everything that happens today finds its raison d'etre in the past, and knowing something about the past can not help but make it easier to make good decisions today.
(via InstaMunger)
Read, as they say, the whole thing.
Posted at 7:52 AM | link | follow-ups | 4 comments
Maximizing the Bird-to-Stone Ratio
I'm a huge college basketball addict fan, but I've long felt that the NCAA is one of the most misguided and obnoxious organizations around. They tend toward dropping the hammer on relatively well-behaved schools for minor offenses, but turn a blind eye to really large-scale systemic problems with college athletics (and the vast sums of money they generate as a result). The problem is, most of the people who make a stink about this are so morally compromised that it's hard to really support them. It's sort of like the NFL contract rules-- if Marvin Harrison were to make a stink about the fact that he can be cut without warning, but can't renegotiate his deal after a good year, I'd have no trouble being on his side, but Terrell Owens? No thanks.
Anyway, via Yoni Cohen, here's a case with some potential:
[JMU senior guard John Naparlo], whose stage name is "Johnny Napp," has not participated in any of the Dukes’ first four practices because of NCAA eligibility problems caused by his budding country music singing career.
Chief among those is the use of his likeness on the album cover of his debut CD "Cowboy Up and Party Down" as well as on promotional posters and his Web site, johnnynapp.com.
The pictures call Naparlo’s amateurism into question because they violate an NCAA bylaw that says an athlete may not "permit the use of his or her name or picture to advertise, recommend or promote directly the sale or use of any commercial product or service of any kind."
Naparlo was told by the NCAA in mid-September that he must remove his picture from his CD and promotional material before he can practice. In addition, he can’t record or tour during the basketball season, because athletes are not allowed to work while their sports are in session, according to another NCAA bylaw.
It's interesting not because I have any great love of country music, but because somebody really clever and Machiavellian ought to be able to parlay this into a smackdown between the NCAA and the other great obnoxiously meddlesome organization of our time, the RIAA. I'm not sure what the best course to follow is-- spread a rumor that NCAA President Myles Brand downloaded "Cowboy Up and Party Down" without paying for it?-- but somebody smart ought to be able to make this happen...
(Exam day today. No Deep Thoughts for you.)
Posted at 7:55 AM | link | follow-ups | 1 comment
Heighten the Contradictions
I've got a big stack of lab reports to grade, which is going to preclude writing long, detailed blog posts about, say, how to properly collimate a diode laser. I also need to make up an exam for Thursday (ed.: Um, that's tomorrow, isn't it? Hush, you...), and soothe the nerves of a bunch of students who are all freaked out about the exam, the labs, and, for all I know, how to properly collimate a diode laser. Whee!
I do want to post a quick note here to say that Ralph Nader is doing an appearance here next week. They're trying to keep it small-ish, so the announcement to the general campus didn't go out until yesterday, and it's not being advertised to the general public (hence the lack of a date from me). When it did, it created a small amount of buzz, and triggered about a ten-minute rant from me about how I find Nader reprehensible (at least in his present incarnation).
I'll probably end up going, just because they never have enough people at on-campus events. Whether I'll be able to avoid saying something that might negatively affect my tenure chances is another issue (a fair number of faculty are active Nader supporters).
Anyway, if you have suggestions of reasonable-sounding questions to ask that will highlight the reprehensible side of his campaign, or would just like to comment about Nader generally, you know where the comments are. Civility will be ruthlessly enforced, however-- I don't want a flamewar here, and I'll disemvowel or delete any posts trying to start one.
Posted at 7:57 AM | link | follow-ups | 11 comments
Tuesday Random Tracks
Another Lab Day, another collection of pop songs. These are from the four-and-five-star playlist. (Shuffle play on the whole library throws up lots of songs that even I don't recognize. This is safer.)
- "I Feel So Good," Richard Thompson. Having seen the man live, he's not all that credible threatening to take someone apart tonight.
- "Romeo & Juliet," Dire Straits. Mark Knopfler does a love song the way it's supposed to be.
- "The Devil Went Down to Georgia," Charlie Daniels Band. did I mention that I was raised by hicks? Also, the Devil totally won.
- "Carry That Weight," The Beatles. Abbey Road is the best album the Beatles ever made. Discuss.
- "Street Fighting Man," The Rolling Stones. Mick Jagger has a little more street-fighting redibility than Richard Thompson. Not much, but a little.
- "Growin' Up," Bruce Springsteen. Springsteen trying to be Dylan doing the Angry Young Man thing. Not his best, but not bad.
- "Short Skirt/Long Jacket," Cake. "With fingernails that shine like justice, and a voice that is dark like tinted glass."
- "One Headlight," The Wallflowers. A good grad school nostalgia tune.
- "Drown," Son Volt. You do know there was another guy in Uncle Tupelo with Jeff Tweedy, right?
- "Birthday," The Beatles. Not what you'd call one of the lyrical highlights of the Lennon/McCartney catalogue.
- "I Found A True Love," Wilson Pickett. She was driving a Mustang down the dark end of the street in the midnight hour.
- "The Girl With The Wandering Eye," The Rugburns. A weird little pseudo-Celtic ditty, that isn't actually all that juvenile.
- "On The Road Again," Bob Dylan. Honey, how come you don't move?
- "The Sick Bed Of Cuchuliann," The Pogues. A song with more words than a man with teeth that bad should be able to manage.
- "If I Had a Rocket Launcher," Bruce Cockburn. How could I not buy a tune with this title?
- "Grey Ghost," Mike Doughty. It's sort of hard to tell the songs on Haughty Melodic apart, but they're all pretty good.
- "All That I Know," The Get Up Kids. I believe they started as emo, but I think they blew that with this record.
- "Fruits of My Labor," Lucinda Williams. A new acquisition, via Kate, but it made a good first impression.
- "My Best Friends Girl," The Cars. I'm a child of the 80's. Sue me.
- "Bagheera," Blues Traveler. Best Kipling reference ever by a jam band.
Posted at 7:32 PM | link | follow-ups | 10 comments
Fun in the Lab
Fiber optic cables are basically long threads made of two types of glass. The outer part of the fiber, called the "cladding" acts to confine light to the inner part, the "core." You can sort of picture light as bouncing back and forth acros the core as it travels down the fiber.
In a single-mode optical fiber, the type used in most laser experiments, the core diameter is somewhere around 10 microns, or about a tenth the diameter of a human hair. That's really, really small, and getting light through a fiber demands that the light be sent into the fiber along an extremely narrow range of angles (in terms of the "bouncing back and forth" picture, you want to minimize the number of bounces off the walls, so you need to send the beam straight into the middle of the core). Coupling light into optical fibers is one of the reasons God made graduate students. Sadly, I don't have any graduate students to push the task off onto, so I have to do it myself.
The basic process goes like this: first, you take the laser beam you'd like to send through a fiber, and bounce it off two mirrors (well, there are almost always more than two mirrors, but you need at least two that you can adjust freely without screwing anything else up). After those mirrors, you send the beam into a fiber coupler, a little widget consisting of a lens and a fiber cable connector-- light hitting the lens is focussed down onto the end of the fiber, and hopefully, some of it goes through. Each of the mirrors has two knobs, one for horizontal tilt, the otehr for vertical tilt, and the fiber coupler has horizontal and vertical tilt adjustments as well, giving you six knobs that you can adjust to try to maximize the amount of light through the fiber.
The first trick is getting started-- you need to get a tiny bit of light through, to give you a signal to optimize. I usually handle this by taking the far end of the fiber and connecting it to a photodetector in such a way that the only light hitting the detector is what comes through the fiber. Then I plug the photodetector into an oscilloscope, crank the sensistivity up as high as I can, and twiddle the tilt adjustments on the fiber coupler until I see a signal.
Once you've got something, the optimization process is straightforward: you start by tweaking each of the six knobs in turn, and maximizing the amount of light you can get through using that knob. The horizontal and vertical adjustments are theoretically independent, but in practice, each knob you turn affects the others, so you need to cycle through all six knobs multiple times until you reach a plateau. This almost always gets things to a point where I can take the fiber off the scope, and put it into a power meter (an analog power meter, with a needle gauge-- you can't adjust things worth a damn looking at a digital readout).
Once you hit the maximum transmitted light with the six knobs individually, you start on combinations. The first step is to just work the mirror mounts-- pick one screw, turn it a little bit in one direction, which lowers the transmitted light, and then adjust the corresponding screw on the other mount to maximize the intensity. If it goes up above where you started, make a second turn in the same direction. If it ends up lowering the intensity, go back in the other direction. There are usually several cycles through this process, for both horizontal and vertical tilts.
After that, if you don't have enough intensity coming through the fiber, you get to the really annoying part-- adjusting one of the knobs, and then cycling through the other five to maximize the transmitted intensity. You can end up making pretty substantial improvements this way, but it's miserable, tedious work. (Hence, graduate students...)
If that doesn't do the job, then you need to start playing with the size of the input beam, which means putting lenses in front of the fiber coupler. Which, in turn, almost always means going all the way back to the beginning, because it's nearly impossible to drop in a lens without moving the beam away from the ten-micron core.
If all goes well, the whole process, start to finish, takes less than an hour. On a bad day, it takes, well, all damn day. On a really bad day, you end up tearing the entire set-up apart and starting over. At the end of the day, if you're really good, you can get something like half of the laser power through the fiber. If you're really good. I tend to regard thirty percent as a complete victory, and I'm usually happy with twenty.
And if you should happen to do anything that changes the alignment of the laser upstream of your two mirrors, you get to do it all again...
And that's how I spent my Sunday morning. Still, it beats grading lab reports.
Posted at 12:08 PM | link | follow-ups | 11 comments