11 October 2012

How do you know an extroverted engineer?

xkcd: my normal approach is useless

He looks at your shoes when he talks to you.

Engineer jokes abound for a reason. I think we get trained to think in math and look at everything as a problem to be solved. Plus we seem to have a lot of introverts. A lot of the time most people spent practicing social interaction, we spent hanging out in our heads and never finessed the people skill set. Personally, I think we hang our old posters up in academic buildings so we can look at something besides the other people in the hall as we walk by them.

However, yesterday I walked across the engineering quad and was looking at someone who had a sweatshirt I liked. In that awkward moment when he realized I was staring, instead of the standard everyone-looks-away or the territorial stare-down, he just smiled and everyone's day carried on a little more smoothly. Now, to implement a new interaction algorithm!

03 September 2012


With semester starting, my attempts to make updates on my projects here became amazingly abortive. A few things that I think are of note, with updates to follow:

  • Note taking and textbooks on a tablet
  • Learning vector calculus - kind of like learning another language
  • Calculating my physical displacement - very eureka! 

28 July 2012

Hanging out with the Higgs

When a world megaproject creates one of the most expensive scientific instruments of all time, a lot of attention seems an understandable result. The CERN quest for the Higgs boson has certainly garnered a lot of attention, and not just in the scientific community.

Commentary has come from many sources (my favorite being The Onion explanation of the LHC or the "$50 Billion Science Thing"). Even Dilbert got on board:
Dilbert finds the Higgs Boson

The Higgs-bleed through into culture has shown up in fiction, satire and television

Alpinekat made a rap explaining the fundamentals of the epic scientific endeavor several years ago, which I love sharing. 

Most recently, someone sent me a video of the sonification (translating measured data into sounds) of the Higgs-Boson. For more explanation, including the sheet music, check out the article Higgs the musical.

23 July 2012

What's the most probable sorority/fraternity?

Using the list of the fraternities and sororities on North-American Interfraternity Conference and the National Panhellenic Conference wikipedia pages I did a quick and dirt frequency analysis (more commonly used for cryptography/code breaking).

Apparently for a sorority AΔΣ is the combination of the most popular letters, while ΔΣΦ takes it for the frats. 

10 July 2012

All the world's a microscope stage

There's an insane amount of detail in the world that we never see. Not just the Sherlock Holmes-like attention to environment and behaviour, but the underlying structure of things. Random little things that you never expected.

For a microscopy project, I needed to select something unusual to image with both SEM (scanning electron microscopy, which is launching electrons at a surface to look at details that are so tiny even light waves are too big to show you what's happening) and AFM (atomic force microscopy by dragging a tiny tiny needle across the surface of things like it's reading molecular Braille). 

When I came home, the most unusual thing on hand was my pet hedgehog, Ada.
Ada the hedgehog

I took samples of her quills and some baby quills from the breeder Hamor Hollow to compare how sharp the quills were at different ages.

Look at a quill with the naked eye and you wonder...What's that made out of? What is it really shaped like? How does that grow to be? I thought of pins or needles the way I've seen them for sewing. Surely nothing more complicated than the ring-like growth pattern of trees. I've pet Ada and felt quills, so I could confidently tell you the sides were smooth. 

No. All of my guesses were right off.

SEM of a baby hedgehog quillLooking at a quill from a baby, we can consider the pokey-pokey factor to be significant. For comparison, the metal of a light use razor is about 127 um thick, narrowing down to microns (and depending on the use).

Adult quills look similar, but the tips become increasingly worn on older quills, like a razor blade.

After the AFM scan, it turns out the edges of quills have lots of little flakes. Like human hair that you seen in all those shampoo commercials.
The  first image (brown scale) is the scan of the boundary between two of the flakes on a single quill, showing the changes in depth like a topographic map. On the SEM image of a baby quill (gray scale) you can see how much smoother the flakes are when the quill is brand new.

AFM scan of a hedgehog quill
SEM of the side of a baby hedgehog quill

A cross section showed hundreds of hollow unit cells. Well sure, if you're an 8-16 ounce animal, you want your protective armor to be light and transportable. Of course. My brilliant plan to scan the cross section with the AFM was totally thwarted. The tip of the probe that does the scanning is so tiny it would get stuck a thousand times a scan over that surface. That surface should be a little more orderly, too. Cutting the sample for preparation was apparently a little sloppy.
cross section of a hedgehog quill (SEM)

And the hollows in the quills grow lengthwise, with a series of stronger internal support along the length of the quill. Not extruded like play-dough, the way I'd imagined.
cross section of a hedgehog quill (SEM)

At the very least, I'm glad I guessed about all this, because I wouldn't have realized how wrong I was any other way.   

02 July 2012

You're going to rationalize anyway, so why pretend?

We so want to be making decisions in rational ways, but the evidence is to be piling up that we're not rational thinkers, just rationalizers. (Dan Ariely's Predictably Irrational is one of my favorite books on the subject.)

A friend of mine recently was struggling with which law school admittance to accept. She had a gut feeling about what she wanted, but felt obligated to provide very quantitative reasons to friends and family when pressed about her decision.

We created this spreadsheet, weighting the importance of the factors in her decision and the relative value of each school in those categories.

It ended up being quite a felxible tool because weighting and the value assigned could be adjusted until it felt "right." Certainly, it's an arbitrary method, but it allows you to rationalize in a very organized way.

Feel free to save a copy of the spreadsheet and adjust so it's useful.

24 June 2012

Quantum Kitchen

I want my kitchen to exist in a quantum state. When I come home and observe it, it would collapse into a state with prepared dinner. The closest I have come to this is cooking a couple times a week and packing leftovers. There are several standbys that I can 'excite,' depending on my energy level.

From lowest to highest energy level:
0. Microwave popcorn. Open a beer. Sometimes I deny this state exists.

1. Microwave frozen wantons and veggies (mixed veggies are my favorite for this)

2. Boil frozen wantons and veggies

My rice cooker enables the next several quantizations of food effort. These all require you to cook rice. (If you don't have a rice cooker, use a 2:1 ratio of water to rice. Cover it, don't let it boil over, and don't stir. When there is a honeycomb pattern in the rice, it's probably done, but do a test bite to be sure. You can get a lot more involved with it, too.)

3. Courtesy of one of my couchsurfing hosts in Japan: add frozen veggies to the rice while you're cooking it. If your rice cooker permits, you can steam them over the rice.

4. Microwave one of these Indian dinners. Pour it over the rice when it's done. I've tried a couple other kinds, but Patel's is my favorite.

Recommendations for others? Please share!

13 June 2012

Where do the smartest people live?

There's been a snazzy map going around lately. It shows the brainiest cities in the USA based on the performance on Luminosity.

I'm digging the graphic, not just because I like maps and charts, but I've also lived in several of the top 25 areas, and it's good to see they're still representing.

Although the scores were normalized across age and gender, there is a strong trend towards college and university towns in these results, but 
"The result is not driven principally by college students, according to Daniel Sternberg, the Lumosity data scientist who developed the metro brain performance measure. "Since our analysis controlled for age, the reason they score well is not simply that they have a lot of young people," said Sternberg. "Instead, our analysis seems to show that users living in university communities tend to perform better than users of the same age in other locations."
However, the data only accounted for populations metros that had more than 500 observations. I would suspect a skew in those college town populations towards people more interested in and willing to try something like Luminosity to test and improve their mental performance.

Particularly interesting, though probably not surprising, is Richard Florida's report that creativity and knowledge work showed a strong association with Luminosity performance:
"The Lumosity data were significantly associated with both the share of adults with a bachelor's degree or greater (.56) and the percent engaged in knowledge and creative work (.45)."
As xkcd reminds us, correlation doesn't imply causation. Regardless, it's interesting to see what it looks like.

31 May 2012

Graph it all!

Prior to the buckle down for my thesis, I was making a nice graph of the number of pages I read a day. Part of this is because of my compulsive desire to track things, part of it is because I find creating visual representations of data a very appealing past time.

Log scale (on y-axis) of pages read

Google docs 'trends' plot of pages read

I experimented with a number of different types of graphs while I was doing this, trying to find what's most appealing to me. I think the log plot is my favorite. It's easiest to see when I burned out on my "professional" reading at work and tried to smoke through a novel in one night (see early March). The trends plot reveals other things, in particular weekend trips or visits from company where I did essentially no reading and have drastic dips.

I'm undecided as to what I should make my next graph project, aside from the nifty research ones that aren't ready to be unveiled.

20 May 2012

Visible light spectrum to RGB

Some of my recent work has required me to become better versed in color spectrum properties and some of the multitudes of ways we've created to try and classify it. Dr. Dan Bruton has a really excellent and understandable page at http://www.midnightkite.com/color.html, including some FORTRAN code to convert wavelength values to RGB.

It's been very handy for plotting some of the data I've been collecting lately, especially with scatter3, when I still have wavelength as a variable to account for in the output.

I tweaked the code for MATLAB and am posting this version here with his permission. Hope it's useful for someone...

%%Wavelength to RBG code was modified from Dan Bruton's FORTRAN code at
%%Only RGB was kept, Gamma and intensity SSS were for unavailable data
%%Further details and explanation at:

%set wavelength to the name of your vector of values to convert
RGB = zeros(length(wavelength), 3); 
for i = 1:length(wavelength)
    %set wavelength to the name of your vector of values to convert
    w = wavelength(i);
    if w >= 380 && w < 440
        RGB(i,:) = [(w-440)/(440-380) 0.0 1.0];
    elseif w >= 440 && w < 490
        RGB(i,:) = [0.0 (w-440)/(490-440) 1.0];
    elseif w >= 490 && w < 510
        RGB(i,:) = [0.0 1.0 -(w-510)/(510-490)];
    elseif w >= 510 && w < 580
        RGB(i,:) = [(w-510)/(580-510) 1.0 0.0];
    elseif w >= 580 && w < 645
        RGB(i,:) = [1.0 -(w-645)/(645-580) 0.0];
    elseif w >= 645 && w <= 780
        RGB(i,:) = [1.0 0.0 0.0];
        RGB(i,:) = [0 0 0];

18 May 2012

It works with my laptop...

I sit nearest the printer in my office. As with all community resources, some members of the group are far more diligent in upkeep than others. Some members just keep sending copies to the printer when it's out of ink or paper, as if the new electrons will carry a desperate plea and the printer will have mercy on them. More often, someone else will just replace whatever is missing, and the behavior is unintentionally reinforced.

This morning however, my adviser needed to print something and a "fatal error" which I'd never had to fix before was displayed. After the standard fiddles, he went for the failsafe - turn it off and back on. Apparently, it didn't turn all the way off - the job he sent was still saved and began to print shortly thereafter. As he walked out with his pages, he said, "I am an electrical engineering genius."

17 May 2012

The $100 Startup Review

Chris Guillebeau's book The $100 Startup has been out for just over a week. I finally finished it a few days ago and got my post up on Amazon, but I was really excited to read it and highly recommend it.

"In the interests of full disclosure, I received a copy of the $100 Startup in advance, but was so slow reading it because of all the notes I was taking that I’m just now getting to writing my review.

I found this to be a very understandable and approachable guide to starting a microbusiness. There’s no background knowledge assumed or exclusive demographics. The whole book builds on your intentions to provide something worthwhile and your desire to accomplish something independently.

There is very actionable guidance for all stages of starting your own business, from generating your ideas and setting up, to relationship building and marketing up to scaling (or not!) your business operations. I particularly liked the case studies across a variety of industries. There are also worksheets (and additional resources provided online) to work through questions, business plans, product launches and on and on...

Regardless of the stage of your small business, this book has insights waiting to be yours."

12 May 2012

Wearing pajama bottoms to a thesis defense

The last part of March and all of April was dedicated to a high intensity freakout over finishing my MS thesis and giving my defense. Graphene-silicon interfaces, hybrid devices...totally fascinating and life consuming stuff. Interestingly, this was all done remotely. Research and measurements were finished before I moved halfway across the country to my new university, but my masters adviser and I spent some significant time since then determining what all those measurements meant.

Regular phone calls were critical to keep everyone on track. Once a week, all the questions and concerns, updated graphs and edits were given a thorough run down. Logistics were sorted out - who would be on the committee? Do I need to schedule a room? Can everyone work with soft copies of the thesis or should I mail drafts back east? I think this was a key piece to keeping on schedule and still defending by the semester deadline.

Thesis drafts and edits were kept in Dropbox. My adviser pointed out that it provided a nice, passive-aggressive way of reminding him to check my updates since a little notification would show whenever I saved my edits or a new version.

The thesis itself was conducted via Skype. There were minor hitches (video freeze, anyone?) but the call clarity was solid and there was a live presentation of the slides at both ends. Some practice runs of the presentation involved as much debate of technical issues as they did the material of my presentation. I'm not sure if this took away from anything, but it made me very aware of the experience for all parties.

Most importantly, I passed. With all the typing and talking behind me, I think this might make a smashing article for one of the IEEE journals. Especially since all this free time has just opened up in my schedule.

24 March 2012

Task manager app search

I've been trying to find a good task manager that will sync my phone and my google tasks account. Best case scenario: multiple lists (so I can track projects independently), tags (I like to sort by type of task), deadlines, reminders, recurring tasks and no ads (hate how they slow things down and decrease security). I'm happy to pay a couple bucks if there's a solid free version I can test out. I've been reviewing the options in google play and these are my findings, and personal experience to date...

Edit: View the full page more easily here: https://docs.google.com/spreadsheet/pub?key=0AoQs6J3aWrFPdFUzRWYwT1BZUVNPa25VbXUycXUydUE&single=true&gid=0&output=html

22 March 2012

I know you from *somewhere*...

An encounter with social grace is always a good moment for me. Today, I ran into a department head when I was going out to lunch. I'd previously met him when I'd been visiting as an industry partner rather than my current persona of a slightly disoriented first semester grad student. I said a “hello” and slowed down past his table, and he stopped me, “I know I recognize you, but I can't remember where from.” After a quick refresh, he was very interested in how I was doing here. As one of those people who is great with faces and shamefully poor at remembering names (the faces are so effortless and the names such a struggle...what makes this happen?), I've often used a similar line to re-engage with people and wondered how it felt to be on the other side. I certainly preferred it to the minimal acknowledgement I was expecting, and it was nice to know he recognized me.

16 February 2012

At least whiteboards are quieter...

It's a little bit shocking to me when lectures are delivered exclusively on a chalkboard. Just in general. In engineering and technology disciplines, I find it alarming. Think about the resources and media that are available. We have videos, animations and pictures all over the web. There are online university classes, where you could have “guest” lecturers from some of the world's most prestigious institutions. And why so many lectures anyway? I think labs and simulations are even better. I certainly prefer them, being a visual and kinesthetic learner.

I don't discount the value of writing things down. Working through problems by hand can be much more tangible. Writing down a derivation helps me remember and makes it easier to catch errors during lectures. Plus, I haven't learned to type equations fast enough to keep up in class. Now I'm considering tablet options instead.

I'm surprised that we still get assignments handed out by worksheets, turned in handwritten, no website with additional references or course information, and a general 'hardcopy-ness' to every aspect of a class. This certainly isn't true of all my classes, but I have a fascinated frustration when I see the resistance to learn and adapt to new tools.

I'm giving myself a project this semester of trying out a bunch of new scientific and math programs on my computer. I primarily use a Linux system, so I'm giving preference to open source (and free - I'm a grad student after all) programs. I thought using my homework assignments would give me specific problems to try out with these tools, rather than a vague task that I'll never start. I started to notice that resistance in myself before I left my old job, and now I need to train myself out of it.