The Android OS, despite being less than a year and a half old, has fragmented to an appalling degree. Different manufacturers include varied screen sizes, sensors and actuators. So, application developers cannot predict what hardware is available. And consumers cannot easily predict if an application they download will function correctly.

The operating system itself has even become fragmented. There are five different versions of Android available across different hardware. Developers now have to consider these different versions, not unlike working with different browsers. Developing for the latest version (as Google does), effectively tells anyone with an older version (79.6% of the Android market at the moment) that they’re out of luck.

I will note that the speaker suggested using version 1.5 or 1.6 rather than 2.1 in developing. On the one hand, that does solve a short term problem of diverse systems. On the other, it makes an unsafe assumption of developers: that we’d want to use an older specification. For most programming languages, it is better practice and makes a person more employable to use the latest standard whenever possible.

And, there are some genuine reasons for this fragmentation. With more varied hardware, it becomes increasingly difficult to include the appropriate drivers (needing to write one for every new feature for every individual piece of hardware). Some devices do not contain the memory or processing to be upgraded, which effectively keeps them trapped with their original version.

It’s this fragmentation that will eventually lead to different flavors of the Android OS, more or less defeating the purpose of a unified operating system. Manufacturers may want to consider having an easy Lego-like means of upgrading hardware, but that still does not account for building the drivers (and assumes that manufacturers would even want open hardware).

Ah well. It was a nice idea.

I spent Monday sick and grading. (I did not receive the assignments to grade until after the department grade-by due date and had to rush.)

Grading is a very disheartening. The purpose of the assignment was to demonstrate using a queue. Prof said they could use more advanced data structures. The students did not comprehend they were to still follow the queue requirements (enqueue, dequeue, limited access, etc.).

Out of 70 students only 1 included a queue, reused from recitation. About 20 had included Java’s queue (many of which preceded to use a secondary data structure to do the actual work). The rest used another data structure altogether to reach the same effect: ArrayLists to LinkedLists to HashMaps to trees.

For people that are not familiar with a queue, this is among one of the simplest data structures to implement. It implements a First In, First Out (FIFO) means of accessing data. One can only access the first (or “top”) Object in the queue; all other contents are inaccessible.

… I don’t understand the drive of these students to use more complex data structures (and code design) than is necessary. A queue uses notably less memory and has a faster access time these other data structures.

I read comments that a hash table is more efficient than a queue. Queues have constant time adds (enqueues) and removals (dequeues). A hash table has O(1 + k/n), where k is the number of keys and n is the number of elements; this is between linear and logarithmic time scales (meaning the amount of time to perform an add or removal will increase with an increase in the size of data structure).

This scares me. There will be (if there aren’t already) computer scientists and programmers making inefficient code because they genuinely believe their way is better.

… I talked to the prof in charge (who has been out of town this week) about my issues. My actual boss, the Director of Mentoring and Retention, had spoken to him previously about this. (I was actually third to complain about it to her.) And she was not getting through to him; she was even considering going to the department chair.

After I told the prof what had transpired, the light bulb finally went on for him. Students will not prove they understand concepts unless they are forced to. Getting the right answer is more important than learning, I guess.

For my software engineering course, I am creating a very simple Roman to Arabic numeral converter. (If anyone knows what the unicode character for Roman numeral 5000, a V with a line over it, I would appreciate it for accuracy.) It isn’t the converter program that is important. The purpose is to demonstrate black box testing methodologies: testing without knowledge of the internals of a program.

I developed a test plan and wrote the test cases in JUnit, as instructed, before proceeding on to writing the actual converter. I then ran my tests. A couple failed and a hand full had an error. I rewrote the code; the class is small enough to justify it, rather than worry about where my sometimes failing  bug was.

Yet, I still had the same problems. So, I copied the short bits of code into a main() function to use the debugger. The debugger revealed no faults: my system produced the correct output. JUnit had produced a false negative. I didn’t know it could even do that. It did not occur across similar tests with identical assert statements, so I’m not sure what it could be.

In addition, I will be grading the first programming course, tutoring a foreign visually impaired student, and doing side research for the Director of Mentoring & Retention. I’m not even sure my workstudy is for this semester, but I’m guessing the Director wants to spend every dime (thereby making sure I get the full employment pay).

In most semesters, I have done side projects which dabble into my interest. Yet, I’m not sure I will have the time this time. So, I have decided that I will continue watching how this semester goes and then decide whether I have the time.

If it does work out, I think I would like to do some cell phone programming (probably Android, but maybe the iPhone‘s Objective C). I don’t know what I would want to program though. I would like to involve hardware, like the GPS or accelerometer. I think the best tools are tangible or ubiquitous.

I’ve stopped reading my CS text books because, well, they’re just not helpful. But then, trying to learn anything more unusual like how to use the tools available is still very difficult. So, I’m just going to do a quick overview of what I’ve learned from using different tool sets.

LilyPad

So, the LilyPad series of embedded systems modules is designed for wearable computing. So, the best place to start learning about the LilyPad stuff is from the creator, Leah Buechley. Things to keep in mind:

• I never needed to create a pad for my units to prevent slippage. But, you’re millage may very.
• Pins (rather petals) of the LilyPad Arduino are dictated by an int (integer) value.
  • 13 is reserved for the LED that exists on the LilyPad Arduino.
  • RX/TX are for serial connections (see XBee below)
  • Analog (denoted by “a” on the petal) pins retain the same numbering scheme as the digital ones(so there’s an analog and digital pin 0, which are both labeled “0.”)
• Buttons should be set up connected to two digital pins. One should be written to be HIGH while the other is written to LOW. Then, just do something whenever the HIGH pin reads as LOW. (This is contrary to the standard debounce that is done with buttons.)
• Do not EVER cross conductive thread lines or allow them to touch one another. This will destroy one or more modules.

I have used all of the LilyPad modules available. Except for the slide switch it’s pretty much “find the closest example and repeat” (if anyone knows how to use the slide switch, I’d like to know).

XBee

Based on the ZigBee standard, XBees are designed for WPAN. Indoors, they are restricted to within a room, but can be separated by miles outdoors.

This forum post is probably the easiest I’ve seen in getting started. Things to also consider:

• RX connections go to TX and vice-versa. Weird, counterintuitive, but that’s how it goes.
• With a firmware update, it’s possible to have the XBee use actuators. This isn’t direct control, more like if one XBee has a pin go HIGH, the XBee it talks to will also have that pin go HIGH.

Personally, I found it a lot easier just to use the XBees for communication and an Arduino for the actual sensor data and actuator control.

Trackmate

Trackmate is a computer vision based tangible object system. From a user standpoint, it’s pretty easy to control: cut out some inch-square icons, tape them on to objects, and use the interface.

From a programming standpoint, not so much.

• I recommend the Minim Processing library for any audio work. Seriously, don’t even bother trying to do this from scratch.
• The code itself, when I last looked at it, was very hardcoded, so don’t expect to change the behavior of the system unless wanting to write it from scratch.
• Review Processing. While it is derived from Java, it leaves some more advanced things out.
• Review matrix algebra and vectors. While you won’t be doing math directly, it will help to make sense of some of the values.
• I did the initial setup using the PlayStation Eye camera as recommended. While I could use the camera’s drivers and operate it from outside Trackmate, it didn’t work with Trackmate (looked like snow). Don’t know why, but it just wouldn’t work for me.

Today, I missed class because there was more grading for me to do. This extra work was because the prof I am TAing for accepted homework assignments via email from a handful of students and didn’t bother to tell me. 2 of these students emailed after the due date.

None of these students has a learning disability, let alone one that affects their ability to turn work in. And, even if they did, university policy states that disability status cannot force a professor to have a flexible due date for students with disabilities.

I’m pissed for several reasons:

1. The students know damn well they need to use the checkin program and how to use it. Department machines can be accessed remotely.
2. Turning in work late isn’t okay. Unless there’s a medical emergency, then tough shit.
3. These students were out of town the weekend that the assignment was due. They had 1 and a half weeks to do and turn in an assignment that should take at most 2-3 hours at their level.
4. I wasn’t informed until today about these students, 10 days later. I’m expected to finish grading within 5 days of the due date. (I ended up missing a class to do this new grading.)
5. The professor didn’t consider the graduate TAs saying “no” to the students requests. This is rather disrespectful.

Special treatment is very different from accommodation. I have never turned in an assignment late or partially completed; no one has ever told me that these are acceptable things to do. No one has said that misinterpreting directions, or choosing to view them as “not including me,” as being an acceptable thing to do.

I have begun the process to file a complaint about this. And it’s not just about me. It’s about the graduate TAs that are expected to tell me what the professor is doing. And it’s about the 65 other students that did follow directions.

Damn, I wish my job wasn’t so stressful.

This semester has been hell though with grading. I am grading, by hand, the quality of the code submitted (the equivalent of a grammar check on an essay) for the second semester programming course. This latest assignment is taking 20-30 minutes to grade each student and with 66 submissions I can feel my brain melt.

Because I’m having to use more mental and time resources than the Director expected when she signed me over to another prof, she has asked me what work I want to do my final semester. And the idea that I have in mind is very unusual.

We have 200 and 400-level independent study credits. For several years, excluding last year, there has been a group independent study in the spring with students from both the 200 and 400 level. This is led by the Director and usually another faculty member with approximately 10 students. The purpose being to do undergraduate research as a group or groups and practice presenting findings in the spring poster session.

I have asked to be the teacher for this course next semester.

I know that at some point of my academic career I will be a lab/recitation teaching assistant. Between my social anxiety and attention deficit, I can honestly say that teaching will be difficult for me. So, I want practice first.  The group study seems like a good place to get my feet wet.

1. No preset agenda.While I enjoy structure when learning, it will take some coaxing to keep me from going on a tangent.
2. Minimal grading compared to normal TA work.
3. Encouraging creativity and critical thinking. I’m passionate about learning; creative and critical thinking are fantastic skills for others to develop.
4. Writing reports and researching literature are two of my strongest skills which I can teach to others.
5. Small group size. Less people means less anxiety.

The Director approves of this suggestion. Now, we just need to know whose approval we need for me to do something that’s never been done before by a graduate student, let alone an undergrad.

Yesterday, I went to a talk on science education in the twenty-first century. The main points from the talk were:

• Traditional lecture-style teaching results in 10-20% retention of knowledge.
• Working memory can last for at most 7 distinct topics.
• Examples, analogies, and images decrease the working memory load.
• Explain things as cohesive units rather than as disjoint parts.
• Even if the choice is insignificant, the ability to have a choice increases motivation.
• Most students leave classrooms with less “expert-like” thinking than when they entered.

… So, if it isn’t really clear, I research a lot on education in STEM areas. And, the longer I am in college, the more I am convinced that many professors have no idea where the bar should be set on how students should preform. A particular favorite is a professor who explained his curving scheme thusly:

Of course there will be a curve! I list everyone by their uncurved score and then [moving his hands in a window/gap creating fashion up and down] go ‘hmm there.’ And divide the grades that way [with partitions].

In all honesty, this seems to be the primary methodology for curving computer science courses, including ones where exams are already curved. And it seems detrimental to students learning. This arbitrary curving indicates that one did better than some percent of students, while giving no evidence of what material was actually learned.

I am currently grading the second programming course in the curriculum. And I feel like banging my head against a wall while doing so. Students make mistakes on material covered in the prerequisite course. Repeatedly. After I dock points and write a comment on what error occurred.

The professor for this course disproves of  me grading students harshly. This includes docking points for having unnecessary looping and requiring that all of their variables be more meaningful than ‘x.’

The professor I am working under is also the professor for one of my courses. My project is notably more difficult than others (many have settled on game A.I. while I’m doing computer vision and machine learning). On more than one occasion, he has indicated that he feels it is too complex for me. I’m not sure if this is cultural or because he feels I can’t succeed or that it shows an example of higher difficulty that others should be able to do.

From elementary through high school, in gifted education and accelerated learning courses. College is the first setting where I am always in classes that are designed for the middle. College is also, interestingly, the first time that I have tested notably worse than my peers and been told that I can’t do something.

I find myself questioning why I am compared to everyone rather than judged on my own qualities. I have had professors grade me more harshly for being clever, and I don’t agree to this. But, if I’m choosing to do work at a different level, why should I be discouraged? And, to be fair, I think students working at a lower than mean level have a right to work at that level.

Professors need to work on showing students why subjects are important and how to learn them. Professors are typically passionate about their subjects, they should share this passion.

From both professors, I have gotten the impression that the project is intended to demonstrate that I learned something. So, rather than worry about the scale, I’ve decided to combine the two projects into a super-project.

Not to say it won’t still be quality work. I just as soon do both together than to risk doing one project poorly.

So, what will I be doing?

1. Embedding a camera, RGB LED, LilyPad Arduino, a mode switch, and an xBee onto a canvas.
2. Connecting an xBee to my netbook, Spinnel.
3. Creating a touch and look modes.
4. Having Spinnel react to the actions.

The touch mode will use conductive thread embroidered into the canvas. When the user touches it, the node will activate and do a specific task.

In look mode, the camera will take pictures.  These photos will be sent to Spinnel for processing. Using OpenCV library, a series of movements will be interpreted as a gesture and the system will activate an event based on this action.

So, why can I combine these projects? Well, a lot of it comes down to the size of the project. In A.I. many people are doing something that could be done in a couple weeks, like programming a game-playing algorithm. Because I’ve decided on a more advanced topic, I get a lot more flexibility in how and with what tools I approach it with.

Now, it’s just a matter of getting the time to do it.

First off, the emphasis on having courses on a curve. I understand the significance of having the course weighted for even distribution. But, when the curve is set up with a C+/B- as the median, I’m not convinced that students understand what they are suppose to. In this system, you can tell how well a student did relatively, but not what they know; there isn’t a checklist of knowledge in this case.

I, as I have mentioned, am doing grading this semester. It’s depressing how few students follow directions.

… Yet, I can’t entirely blame the students. They are being taught to program in the lower levels. The importance of theory (algorithms, logic, and critiacal thinking) aren’t emphasized. It’s very difficult to blame a student for not understanding, if (s)he doesn’t know why things are important and why we do things a certain in way.

Even the tutors and some of the grad students don’t get the signifigance. I just want to hit my head against a wall sometimes.