23 April 2008

It's About Time!

Finally someone comes out and tells the truth about cell phones and aircraft instrumentation...and it only took the airlines opening up Wi-Fi on flights to make someone actually do their research and explain it to the rest of the world. Cell signals do not interfere with aircraft communications or other instrumentation. The radio waves used by cell phones are limited to specific frequency bands shown here, and they do not overlap with those used by aircraft communications as seen in the U.S. Frequency Allocation Chart. They've traditionally been prohibited on board airplanes because they play havoc with the cell phone networks.

A cellular network works because the towers have overlapping areas of coverage -- think of a Venn-Diagram -- and when the phone is in the intersection of two towers' coverage areas, it chooses to use the one that gives it a better signal. The problem is that the radio waves used by cell phones get blocked by buildings, mountains, trees, and other unfortunate side-effects of civilization and geography so the towers are fairly close and powerful to provide good coverage. A plane flies over all the detritus on the ground and your cell phone can receive a decent signal from towers miles away. This means that the phone is constantly switching towers to find the best signal and that puts unwanted strain on the network. A cell phone will crash a plane the same way it will crash a car -- by distracting the driver, not by interfering with instrumentation. And airline pilots get slightly more training than automobile drivers.

Wired magazine ran an article in the current issue that mentions this, although it doesn't do much to explain it. The primary focus was to let people know that several major airlines are providing Wi-Fi service to passengers via-cellular networks. The way these work is by having a "cell tower" in the plane connect to the base stations and all the network devices on the plane connect to its tower instead of the cell sites, thus avoiding the network problems described above.

I may have some details wrong in the above description as I don't know a great deal about radio communications, but the gist is correct and this has been a pet peeve of mine for a long time, so I'm glad the myth is finally getting resolved. It's as bad as the 10% of your brain myth.

~KMarsh

Never Gonna Give It Up

They just RickRolled karaoke night at the dining hall....

Old Engineers are Amazing

I just spent the last 40 minutes up to my elbows with a fellow student in the nether regions of a pinball machine. It's an old Adams Family pinball table that the student hangout and pizza joint recently acquired and it wasn't working; the ball would vanish down a little chute and should have been knocked to the side and then shot out. When the solenoid triggered to knock it to the side, though, the ball didn't move. We pulled the ball out to look at the chute and I found a light bulb from elsewhere in the machine had been knocked off and gotten stuck in the chute, preventing the ball from going through the hole in the side. We pulled it out and the machine worked.

Mostly. The screen didn't, which is kind of a pain. It appeared that everything else worked, so after some rooting around we discovered a key hanging inside the case of the machine that, once we tested it, opened the panel that the screen is on. We took it out and looked around to find a very smoky looking fuse -- we suspect it was blown when Ross zapped himself under the table, but we couldn't quite tell and didn't have a spare on us. We'll go back soon and check to see if that was it.

But the cleverness that you see in old mechanical things! It's really amazing what those old-timey engineers could come up with. Take the bumpers, for instance. They are a plastic or metal tube that the ball collides into. Around the base is a pressure switch that the ball presses when it hits the bumper. That switch triggers a solenoid (almost everything in the machine is a solenoid) that drops a metal ring from the top of the bumper. The ring is concentric with tube part of the bumper but slightly larger around so that it hits the ball somewhere on the bumper-side of the ball's center, forcing the ball away from the bumper. There was also a solenoid pointed at the side of the case. The best guess that Ross and I could come up with was that it was for making big thumps as sound effects.

~KMarsh

22 April 2008

I Love Mudders. And Planes.

I should preface this post with the following confession: I am a compulsive e-mail checker. When I'm at my computer I'll usually have a window open to GMail and I have this nasty habit of replying to e-mail as soon as I see it rather than waiting until I have, say, finished my homework.

With that in mind, consider the following. I sent an e-mail to the student mailing list (students dash L, we call it) asking for students interested in a glider trip with the aeronautics club next year at 22:30 on a Monday night. I had the good luck to send it right as the mailing list moderator was cleaning out his in-box so it got sent right away. Within 3 minutes I had 7 responses. 3 and a half hours later, in the middle of the night, I had 34 responses. One day later and I had nearly 50 responses. I did not expect this, but it is exciting all the same. The 3 minute one is really the important one.

As for the trip, well I'm taking over joint leadership of the Barnstormers (HMC's aeronautics club) next year and one of the suggested events was to schedule a trip for the club to go on a glider trip. We need to submit our budget proposal by Friday, so I needed to gauge interest: a 30 minute flight is about $100, we might get a 10% discount, people'd probably be willing to spend about $20 out of pocket and the club pays the rest if we can get our hands on the dough. Thus, it's just a matter of how many people are interested. With 50 expressing interest right now I figure we'll probably find ourselves with about half that actually ending up wanting/able to go depending on when it's scheduled.

Another thing to note is that Harvey Mudd College used to have an aeronautics program in which students would learn about aviation and eventually get their private pilot's license. Iris Critchell was the driving force behind the program, and she's still here at Mudd. Look for more about the Bates Program and Iris in upcoming posts.

~KMarsh

19 April 2008

Registration and Current Coursework

I've settled on a language now: I'm going to take Russian next semester. I wanted to learn a new alphabet/writing system but I'd been told that Chinese and Japanese are huge time sinks and I don't want to deal with that on top of clinic et cetera. I sent an e-mail to Professor Little who teaches the engineering Project Management course to ask him how much it will overlap with the Software Development course I took last semester and depending on the answer I'll take either that or Computer Networks.

Last night the entire school decided to party. I'm not entirely sure why, but among other reasons it seems that everyone realized that we're near the end and the mountain of coursework is, in fact, doable. For some value of doable, at least. I've finished one of the movement analyses I have to do and have started researching my big paper, so that's good, but VLSI needs to be documented, so we have a huge task writing that up now as well. My robotics project has been scaled back to the problem of finding textbooks instead of generic objects because two of our team members are in VLSI and couldn't work on the robot much recently. I'd better get back to researching.

~KMarsh

14 April 2008

Registration is in the Air

The end of the semester is upon us and as the flowers struggle up through the melting frost (yeah right) the time for pre-registration is come. We just got our pre-reg packets which include our yellow card, tons of fliers, other random spammy papers, and the all important course catalog for the 5C's. Here's the idea: Similar to room draw, everyone at Mudd, by class, gets a time slot to register for classes next semester (officially registration isn't until next semester starts, but we pick our classes and everything now during pre-registration). We fill out the provided card with the classes we wish to take (as well as some alternates in case classes fill before we get to register) and have it signed by both our humanities adviser and our technical adviser. When the time comes we stand in a huge and delayed line, skipping classes and playing general havoc with the school to try to get into the classes we need to graduate.

Tentatively, my schedule for next semester looks like this:
  • CS Clinic (cool industry research/work project class)
  • CS Colloquium (get talked at for an hour a week by people doing CS-y things)
  • Computer Vision (taught by Zach Dodds, who is Awesome, about once a decade)
  • Occult and Magical Philosophy (taking it on Pitzer with friends -- fills Humanities reqs.)
  • Some Language Class
    • German -- 4 days a week and I've wanted to take it for a while.
    • Japanese -- possibly quite useful, but 5 days a week and people tell me it's hard/huge time requirement.
    • Chinese -- 5 days a week, but taught on Mudd by a new prof. Fills an on-campus hum req. but new prof is a gamble. I'm not remarkably interested in it.
    • Russian -- 5 days a week. Slavic language may be useful for travel in Eastern Europe.
  • Some Other Class
    • Project Management -- Engineering class, may have overlap with Software Design.
    • Networks -- I don't know much about computer networks and I really should learn.
    • Enterprise and Entrepreneurs -- Econ class on Mudd, fills on-campus hum req. Easy prof.
    • Materials Science & Energy Conversion/Storage -- something cross listed under Chem, Engineering, and Physics. Sounds interesting, but held at 8:00-9:15pm
Pretty cool choices, this semester, all in all. Oh, and for those of you who wondered, VLSI has improved immensely -- we got the microcode done and are just puttering around with optimization now. I still have to start that research paper, but I'm really enjoying myself at the moment.

~KMarsh

09 April 2008

Its the Small Things that Really Count

Since I've come to Mudd I have noticed several small things that aren't mentioned during admissions at any college (that I found) that make the Mudd experience so much better. These are official campus programs, so they're more than just the (generally wonderful) congenial atmosphere between students, staff, and faculty, but they don't get any recognition from the admissions office.

First is DOS, the Dean Of Students office. Dean Chris (also known as Dean Fun at times) hires his "DOS Muchachos" to aid in planning study breaks and events for the student body. Things they get up to include $2 tickets to the midnight release showing of movie X ($5 if you go IMax, I think), water fights in the quad on Friday afternoon, "s'mores on wheels" rolling through the dorms in the evenings during winter, and the "Indiana Jones Interactive Experience" projected on the enormous blow-up screen (think bounce castle, but with a screen) in the quad complete with giant rubber ball covered in brown paper chasing a student in an Indiana Jones hat through the audience.

Next up we have CAP, the Committee for Activities Planning. This committee subsidizes off-campus activities such as movies, concerts, and trips to Disneyland. If you want to do something fun but think it costs just a bit too much, go to CAP and they'll probably offer to subsidize it if you can scrounge up about 10 interested students.

For faculty-student interaction, the Leonard Fund is hard to beat. This fund will pay for up to $8 a person for up to 8 students eating out with a faculty member. The most notable use of this was spring semester 2007 when a group of my friends would go to Jack in the Box every Wednesday night at around 3am with Professor Bob Schaffer. He could almost be a CS prof with those hours.

Finally, one of the nicest amenities and perhaps the least touted is the summer storage. Each dorm has a storage room or two that residents can use to store their stuff over the summer when they're off campus. Mudd also orders in a bunch of "storage pods" which are plywood boxes that are left on campus in the weeks leading up to graduation that are then carted off to some secure storage facility. Most other colleges and universities that I've heard about do not do anything like this, leaving their students to either rent public storage or cart everything back home/to their summer housing (whatever that may be).

~KMarsh

03 April 2008

I'm Okay...Really.

Those of you who read Trevin's blog will have heard already about VLSI. For those of you who don't, VLSI is an acronym for Very Large Scale Integration and is the name for the processor designing, laying out, and fabricating integrated circuits -- computer chips. HMC's VLSI class is taught by a man who is a legend in the industry and as such is incredibly hard. I'm a computer science major, not an engineering major, so I decided to take the class because I enjoyed the simple digital design class and wanted to learn down to the electron level how computers work. That thirst has been slaked.

When the class started, it was incredibly interesting to me and even though it was a lot of work (we had a 16 hour problem set once...) I was really enjoying it. Now in the second phase of the class we are working as a 14-student team on a single, big project: we're building a 6502 microprocessor -- the chip that was the heart of machines like the Apple II and the Nintendo Entertainment System. This sounds cool, and it is, but my partner and I were brought onto the microarchitecture team and for the first few weeks were slogging through this code that we didn't write and had no idea of how it was structured. The last few weeks have been spent finally understanding the code and trying to fix it. All the while the other teams are moving forward and if they're not waiting on us now, they certainly will be if we don't get finished soon.

I think you'll understand, then, when I say that I'm just done with that class. I can't drop it (not because I'm not allowed to, but because I've got this far and really would like to finish it, and also because other people are depending on me to do my work and if I leave then suddenly someone else has to step in and take over for me), but I'm sure my grade is suffering terribly because of the slow progress we're making and, worse yet, this class is eating into the time I should be devoting to my other classes -- I have little time to be working on robotics, which is still fun, and I have a research paper for a humanities class that I have yet to start. Hopefully we'll get this microcode finished this week so we can start paying attention to other things.

~KMarsh