Minesweeper

At this point, Minesweeper is life, Minesweeper is habit, Minesweeper is (maybe) an idol.

D:

So close to breaking two minutes though...

Reflections on Doing Biochem Research

so much depends
upon

1.0 microliters

filled with
polymerase

into the thermal
cycler.

It's About Time: How to Do It and How I Did It

Okay, I recently helped out with running It's About Time at an invitational.  I didn't make the tests, but I saw all the scores.  And...

I will now describe how I think you should It's About Time this year.  I did this event during my senior year, and we medaled regularly, so I think I have some amount of authority on this topic.

Written Test

• Compile a formula sheet.  Practice and get good at doing the math and physics calculations quickly and accurately.  You might want to have someone who is good at these things in this event.
• Look at lots of practice tests.  It helps if your school's team somehow has sets of tests from many invitationals, but you can still find some on scioly.org, I hope.
• Making the binder
• Find a lot of information.  If you have time, make the information easy to read and quick to access before you print them out and stick them in the binder.  If not... well, I guess a lot of printed web pages is better than nothing.
• Approaches on how to make information easy to read and quick to access
• Sometimes you just have large reference tables full of numbers.  Um, yeah.
• Take notes on a computer - to be printed and put in the binder.  When you update the notes, don't just write it in pencil/pen on the printed notes.  If you have time, also update the computer version and print out new versions.
• Put tabs in the binder.  I ended up cutting Post-It notes into thin slices, because I had so many tabs.
• Which topics should I study?
• Read the rules and past practice tests.
• Um, check scioly.org.
• Make sure your partner is also on board.  The test has to be completed in a short amount of time, so you will probably have to split up the test, so both of you have to be good.

Clock

I think you should use a pendulum or spring

I think you should use a pendulum or spring.

We got two objectives: 1) optimize how well you do on the timing part of the test, and 2) minimize the amount of effort/time/money you spend on making the clock, as those resources could often be devoted to other events.

Okay, there's also 3), which encompasses a bunch of miscellaneous personal reasons - the desire to challenge oneself, to build something cool, to satisfy one's curiosity, to have fun, etc.  This may push you to build something more complicated then necessary.  Could be a good or bad thing!  You figure out how to balance personal reasons with the team competition.

• Why water/sand clocks are bad
• Flow rate/viscosity may vary depending on temperature and humidity, and how in the world are you going to compensate for those variables during an competition?  (Granted, the variation may be small, but still.)
• Risk: if you spill, you might get penalized.
• In the competition I helped out in, a team got penalized for refilling the upper reservoir of their water clock in the middle of a time interval.  D:
• (Disclaimer: Never actually tried making one of those; this might not be accurate.)
• Advantages of pendulum or spring clocks
• Can be built to be quite robust
• Small variation in period
• Can be done pretty well with relatively little effort, though it is possible to make more complicated ones.

Basic Design

Okay, from now on, "we" means me and/or the two other guys that worked on the event with me.

• Have a sturdy base and frame to hang the pendulum/spring from.  Make sure it doesn't shake/wobble.
• Base
• We just used a flat slab of wood for the base.  But it wobbled.  So we also taped a washer to the bottom of the slab at a strategic location.  But it still wobbled sometimes.  Blargh.  In the end it didn't kill us though.
• In some competitions, you might actually have to put the clock on a desk that is slightly tilted.
• Maybe make something like a tripod?  I dunno.
• Frame
• Um, make sure the frame is heavy and sturdy enough that the motion of the pendulum/spring doesn't affect it too much.  Although maybe it's okay for the frame to shake a little.  Not sure.
• Pendulum/spring
• Controlling the period
• Let T denote period.  Then...
• Pendulum: T = 2 pi sqrt(L/g) for small amplitudes, but is greater for larger amplitudes.
• Vary L
• Mass on a spring: T = 2 pi sqrt(m/k).  As far as I know, this does not vary depending on amplitude, as long as Hooke's law holds.
• Vary m, the mass.  Or vary k by using different springs.
• What's a good period to use?
• Naively, one might want to make the period 0.1 s, since that's the precision that the competition specifies.  But that could be hard to build and hard to count.
• I think anywhere between 0.5 and 2.0 s is doable.  We ended up consistently getting within 0.2 s of the target time by estimating fractions of a pendulum swing.  Remind me to talk about estimating fractions of a period later.
• I think we did something like 1.5 s with a pendulum, so ~0.75 s for one swing.  But then we had another pendulum with a period of 0.8 s.
• Setup
• Pendulum
• Mass hanging from a string
• String is somehow clamped/fastened/secured from the top so the whole thing, y'know, swings like a pendulum.
• Have a horizontal bar or a hook; tie the string to that with an appropriate knot.
• Clamp the string with something
• Mass at the bottom of a rigid bar
• We didn't do this because I tried and there was too much friction at the axle and strings were so easy...
• (I know some teams have pendulums that extend below their base.  Therefore, you would have to set the clock on the edge of a desk or countertop.  Um... I guess this approach works, but I suppose there's the slight risk where you are unable or not allowed to put your clock at the edge during a competition.  It's also just a bit more inconvenient.)
• Spring options
• I saw Harriton's clock.  They hung a mass on a spring and hung the spring on the frame.  In their clock, the mass bobbed up and down.
• Make sure the mass is massive enough (or that the spring has little enough damping, or that there is little enough friction) so that even after 300 s, the thing is still oscillating.

Testing the Mass-on-the-Spring

I have no idea.  Never did it.  I would assume that it'd be like pendulum testing, except the period is more constant, so your life is easier?

Testing the Pendulum

Preface/disclaimer:  I don't actually know if this is the right way to do things.  But it's the thing we did and it kinda worked for us.

• Time the pendulum
• This is the part which was very time consuming for us compared to other people.
• Easy solution would be: find the pendulum's average period by timing the thing with a stopwatch, counting swings over five minutes and then dividing the time by the number of swings.  At competitions, simply count the number of swings and multiply by the period.
• The issue with this is that the pendulum increases its period as the amplitude of the swing gets smaller.  For the pendulum that we built, if we followed this approach, we'd get errors on the order of 0.5 s.
• What I did
• Found a lap timer - a timer that can record lots of time points.  So I could start it, and then press "lap" again and again.
• At first I used an online stopwatch, but it was annoying to parse the times.  I later just wrote a lap timer in Java with a more convenient output format.
• Sat down next to the lap timer in a comfortable position.
• Started the pendulum and the lap timer at the same time.
• Every single time the pendulum swung to its leftmost or rightmost point, I pressed lap.
• I stopped after five minutes.
• This is super boring but I thought the sacrifice was worth it.
• You might be able to do this in an automated way by recording data with a sensor - perhaps a smartphone sensor.  A teammate found that when the metal mass of our pendulum swung back and forth, it caused light levels and magnetic fields detected by our strategically placed Android devices to vary periodically.  And if you have time series data from that, you might be able to analyze it using WWZ (weighted wavelet Z-transform), which can be done in VStar...  But this is all speculation and sounds unnecessarily fancy.
• Do multiple trials.
• Analyze the timing data
• I put the data from multiple trials in Excel.
• I subtracted each row from the row above to find time between swings.
• I made graphs to see how the period changed over time.  Unsurprisingly, it decreased.
• Table: Eventually, I produced a table that went up to five minutes.  The vertical and horizontal axes were number of swings.  The numbers in the table were the times that those number of swings correspond to.
• The times were not just multiples of the period, but were averaged from the lap time data, so they accounted for period change over five minutes.
• This also meant we had less arithmetic to do - didn't have to multiply period by number of swings.
• Caveat:  If your pendulum mass is hanging from a string, then the period will probably change over the course of days or weeks due to the string stretching and contracting.  So you might want to retest a few days before a competition.  This is kinda annoying.
• Fractions of a period - You need to account for those, or you will not be as precise as you can be.
• Find the times corresponding to fractions of a period.  When you're measuring a time interval with your pendulum, don't just round the number of swings - instead, maybe be as precise as "35 and a quarter swings."
• Keep another table of how long 1/8, 1/4, 3/8, etc. of a swing is, so you can add those in instead of doing more arithmetic during a competition.
• Handle reaction time in a consistent way by always starting and stopping in some kind of consistent way.
• Practicing
• Play random time intervals and try measuring them with your clock.  See what the errors are.  Try to find ways to improve.

Data Table

As mentioned before, if you have a pendulum, you should probably print out two tables: one that says how much time each number of swings corresponds to; one that tells you how long fractions of a swing is.  Here is an example:

Check with the event supervisor at impound.  The supervisor would either want you to impound your tables along with your clock, or he would allow you to keep the tables in your binder but then use them during the clock part.

Even if you don't want to time your pendulum's every swing for five minutes and just found the average period, I still recommend making and printing out a multiplication table of your average period.  That way, you don't have to waste time on or screw up the arithmetic during a competition.

Fancier Designs

As mentioned before, you might not want to do fancier things unless you really want to and have the time and already have a backup clock in case the fancier one fails.  Some possibilities:

Pendulum with an Escapement

If you are doing It's About Time, you better know what an escapement is!  Anyways, the energy from the gradually lowered mass means that your pendulum has approximately constant amplitude and period throughout the entire five minutes.  Yay!  One of us did make an escapement pendulum; the gears and escapement were 3D printed.  It generally worked pretty well.  Cons: Sometimes the escapement would just stop inexplicably.  Also, we had to wind it back up at the end of each run.

Other

• Gravity escapement - I saw a team with one.
• Deadbeat escapement - Supposedly better than any run-of-the-mill anchor escapement.  But harder to build. Maybe 3D print one?  I designed one and 3D printed it, but it did not work.  It just got jammed. :(
• Complicated K'Nex clocks - I've seen a couple of those.  They might work well, but they look quite time-consuming.  Hmm.  Maybe a viable option.
• Their periodic noises may distract and disrupt the counting of other teams, but they're not loud enough to disqualify you.  Yay?

Redundancy

At states, our device had not one but two pendulum.  One of them was a regular pendulum - a mass hanging by a spring from a horizontal bar.  The other one was the 3D printed escapement pendulum. So each one of us operated/measured one of the pendulums, and we obtained two independent measurements on each interval, so hopefully that meant we were more accurate?  On one of the trials, the escapement pendulum ran out of string and stopped prematurely, but fortunately the other pendulum still yielded a time measurement.  Hooray for redundancy?

Conclusion

Good luck on this event.  In closing, I have a few random bits of advice for you.  Don't...

• butcher a Newton's cradle and use the balls as pendulums.  Unless you're very very desperate.
• worry about what other people think about you.  If your heart tells you to move you body side to side or nod your head up and down to match the motion of your pendulum/mass-on-a-spring, go ahead and do it!
• forget to have fun.  Except when you're behind on time and/or thrown into the event and/or completely screwed.  Then, just try your hardest to mitigate your impending failure. :P  (lol jk but really tho)

About Me

Hi! My name is Jeffrey Huang, and I am currently

• (1/12/2015) a senior at Conestoga High School, somewhere in the depths of the suburbs surrounding Philadelphia.
• (1/24/2016) a freshman at Cornell University.

That should be enough to identify me uniquely.

I like math and I like science.  And I have a lot to say about Science Olympiad, if and when I ever get around to saying it.

As you probably guessed from my last name, I am Chinese-American.  I have lived in China for three years, so I can speak and read Chinese.  Can't write any more; can still type.  The title of this blog is one possible interpretation of my Chinese name.

Home

Written 4/22/2014

I am a wanderer.
Do not ask me, “Where did you come from?”
My homeland is far, far away, and words are weak.
How could you believe what I have done?
Understand what I have heard?
See what I have lived?
How could I translate from a language in which
every word has its own emotion,
every saying its own story,
And every idiom its own soul?

Do not ask me, “How was it back there?”
It is different here,
but everywhere I’ve been has been the same.
Everywhere I’ve been,
The sky had clouds and stars,
The hills had grass and birds and trees,
Roads had cars,
Schools had bells,
And humans were humans.
I have not seen war.
I have only seen small cruelties.
I have always lived near a river, near the sea.
And humans are humans.

Do not ask me, “How do you like it here?”
You are curious, but my mouth is tied,
for I cannot judge, cannot condemn
and every place was special when I lived there.
I loved the green mountains tinted gray behind the fog.
I relished the noise of energetic traffic
jostling and flowing in the streets.
I liked the bluish tint of driveways before sunset.
I marveled at the view in the backyard of
houses and trees across the glen.

Do not ask me, “How did you get here?”
The journey was not easy, yet it was not hard.
But as I go on living, a sorrow lingers below.
I am of this place now, but I have learned
That one can pretend that one will never leave
Yet the time of departure will come, regardless.
I can only walk so far before pausing,
Looking back, recollecting.
I will always remember where I’ve been,
but I’m afraid that I will forget.

Do not ask me, “Where are you going?”
For I think I know, but I know I’ll think differently.
I do not know what I need to know but don’t.
The world is wide; I have yet to find my way.
I am a wanderer.

Theory of Numbers

I wrote this essay as a class assignment.  We are working on college essays, and I wrote this in response to a University of Chicago prompt asking for a theory that explains numbers.

I propose that all numbers besides the counting numbers are completely subjective and do not exist except in human thought and imagination.  Most numbers can only describe the universe approximately, and are thus not real.  Natural numbers are useful, and rational and complex numbers might still find real world applications, but I can easily construct a whole number that dwarfs the size of the visible universe, like 10^10^10^10^10^10.  There also exist surreal numbers and infinite ordinals, which are also numbers despite exhibiting some strange properties.  Mathematics is a human endeavor, and though we have used it as a tool to describe the universe and advance technology and engineering, mathematics is only a tool of our own making, nothing more.

Numbers, when applied to the physical world, are simply imperfect labels of things.  They are almost always approximations.  One can never measure the mass of an object, like an apple, exactly, since there is always some error.  Our world is not ideal: if we carefully made a square of side length 1 m and tried to measure the diagonal, it would not be exactly sqrt(2).  Our theories for describing nature may not be perfect, either.  The laws of Newtonian mechanics provide a very good approximation for how projectiles fly and galaxies twirl, but it is not exact, and it would be a bit arrogant for humans as a whole to presume that theories of quantum mechanics and relativity are the last, best description of the universe. (On the other hand, natural numbers, used for counting, seem to be fundamentally a property of the universe: if I say three electrons, I mean exactly three electrons.)

As it stands, numbers are limited in predicting things.  Perhaps it is simple to describe the trajectory of small spherical frictionless cannonballs in a vacuum, but fully-fledged human beings are another thing.  We cannot predict the actions of humans, except in a few isolated cases.  The universe is effectively non-deterministic – there is much we do not know and cannot know.

Yet we use math everywhere.    Even complex numbers, which might initially seem useless, find applications in electrical engineering and other branches of science.  Constants like pi or e are ubiquitous.  Still, this is only one system of math that happens to work for us.  If we were to encounter an advanced alien civilization and compare our knowledge of physics and math with them, we would probably find some similarities and many differences.

Not all numbers are useful for describing the universe.  With exponent towers, we can easily make a number so big or so small as to be meaningless, like the aforementioned 10^10^10^10^10^10.  Mathematicians have created much larger numbers, like Graham’s number, but that deals with the abstract problem of coloring the edges of n-dimensional hypercubes.

Mathematics, for the most part, is self-consistent – but only because we made it so.  We have number systems, like the set of rational numbers, and we can do arithmetic with them.  But did we simply make all of this up?  Mathematicians have “discovered” the ordinals, some of which are larger than any countable number – effectively infinite.  Yet there are infinitely many infinite ordinals, and mathematicians have defined ways to add, subtract, multiply, and divide them.  Both the rationals and the ordinals are self-consistent – under the established rules we cannot find a contradiction.

Ultimately, I say that numbers (besides the natural numbers) are only made up by humans.  We can create new systems of numbers, as we like; we can play with our existing rules and make “discoveries” and theorems based on those rules.  With the numbers we have, we can only approximate the behavior of the universe, and we cannot predict everything.  Numbers are only a game – a game of our own making.