More on Echo360

Because I posted on my blog about Echo360, I have been contacted by a marketing person from Echo360 who is interested in what I have to say. Here is what I told her about my experiences so far.

I think Echo360 is a very useful tool, but there are some other features that would help. One thing I think would be useful is to integrate a discussion forum with the lecture capture. The way I would envision this working is that when a student is watching the lecture online, if he has a question he can click a button and type his question in, then the question would be tagged with a timestamp indicating what point in the lecture he asked that. And there would be a way to link to a particular timestamp in the lecture. Then the instructors/TAs can log on and look at the questions, and when you looked at a given question you would have a link to the associated point in the lecture. Or an instructor could answer by pointing the student toward a particular place in the lecture. This would be useful to me because oftentimes when I talk to students, they say the instructor told them something which doesn't sound right to me, but it's hard to clear up the confusion without the lecture in front of me.

Another problem, which is a little more technical, has to do with the sound. The way our setup works is that there are two microphones, one worn by the instructor and one "shotgun" microphone. The one the instructor wears has much better sound quality but only captures what the instructor says, while the shotgun microphone can capture everything in the room. So one of the things I have to do is when the instructor pauses for students to ask questions, switch to the shotgun microphone, and then switch back when the instructor starts talking. This often isn't very reliable and you don't get all the questions. So a useful tool would be to have two or more audio channels going into the system, and it would automatically switch between them depending on who is talking.

I'm also taking a TA training seminar where we are reading the book "Teaching At Its Best: A Research-Based Resource for College Instructors" by Linda B. Nilson. That book talks a lot about how traditional lectures aren't very good at promoting student understanding, and has lots of suggestions for how to improve lectures. For instance, one way of improving lectures is to periodically stop and give students a question to answer and discuss in order to keep them engaged. Something like this could be incorporated into Echo360 - if you had a system like the one I described in the first paragraph, then the instructor could put a question on the whiteboard and ask students to answer it by posting something. It might even be possible to set up a system where the instructor can set the lecture to automatically pause and ask the student a question, and not to continue until the student gives an answer.

More school stuff

- I have just finished creating, assigning, collecting, and grading the first homework assignment of the year as a TA. This homeowrk was about formal logic, including propositional and first-order logic. (Propositional logic just has variables which can be either true or false, while first-order logic allows variables to be arbitrary objects, there are "predicates" that are basically functions which take objects as arguments and return a boolean variable, and there are "quantifiers" which allow you to say things like "for any X, P(X) is true" or "there exists an X such that Q(X) is true". We thought this part would just be review (these are Masters students in computer science, so we thought they would already know about logic) but some of them had trouble with simple things like the difference between validity (statement is true all the time regardless of the values of the variables) and satisfiability (statement is true for some assignment of values to the variables.) In fact one of the problems we had to "cancel" and make it an extra credit problem because it had to do with proofs in first-order logic, a topic we didn't cover (I didn't realize that we weren't going to cover it when I wrote the homework, and the professor didn't notice when he was looking at the homework before posting it)

- I have my Ph.D. qualifying examination this Thursday and Friday. I have looked at most of the previous quals and it seems like I pretty well prepared for it. Another thing is even if I do decide to only get a Masters degree, passing the qual is a good idea because if I pass the qual I am exempt from the distributional requirements for a Masters (there are still two distributional requirements, hardware and systems, that I haven't taken courses in so I would need to do that if I were to go that route).

My Teaching Assistant Job

This semester I will be a TA for CS 477: Formal Software Development Methods with Madhusudan Parthasarathy. This course is about how to mathematically prove that programs (or at least parts of them) are correct. The first 3-4 weeks of the course is about propositional and first-order logic (which I mostly already know) but after that the stuff will be new to me. There are a total of about 30 students in this class, 15 of which are I2CS students - I2CS is the online program. I am the only TA for this course. I will not be giving any lectures or leading discussion sections but there will still be a considerable amount of work to do. I will be doing the following:

1. Helping to write the homework assignments and answers, and also grade the homework assignments. The professor also has money in the budge to hire a grader but we don't think that will be necessary (also, the graders are all undergraduate students, so it will be difficult to find a grader that knows the stuff in this course).

2. Communicating with the I2CS students via e-mail. Also I will have "office hours" which will be conducted via Skype.

3. Operating the camera that records the lectures to be posted on the Internet for the I2CS students. The camera is mounted on the ceiling and it is operated remotely from a "Media Services" room in the basement. Most of the process is automated; there is an "Echo 360" system that is programmed to automatically record during the period of time that the class is scheduled and then post it on a secure server for the I2CS students to log in to. I just have to make sure the camera is working and pointed at the professor. The system is kind of cool - the projector and the screen that the projector projects on to can also be controlled remotely, just in case the professor can't figure out how to operate them.

4. Updating the course web site with the lecture notes and stuff.

More fun stuff going on

The second session of the 4H robotics club happened today, and we focused on programming this time. There's a special programming language that is used to program the robots, that is based on putting together visual "blocks" that represent commands (like "move", "turn", "play sound") and so on, as well as programming language constructs like loops and if/then statements. The kids seemed to have a slightly harder time doing this part than the building part, and they got about halfway through the tutorial heet that they gave them.

Then in the last 10 minutes of class I did my peanut butter and jelly sandwich activity. At the advice of the other instructors, I used just the jelly in case anyone had any peanut allergies. It was fun and the students laughed a lot whenever something went wrong, like when they instructed "take the bread out of the bag" and I dumped the whole loaf of bread on the table.

Also, in keeping with the "learning how to build stuff" theme, a new facility has opened on the Champaign-Urbana campus: the Champiagn-Urbana Community Fab Lab (CUCFL). The CUCFL is one of a network of dozens of similar "fab labs" around the world, and has facilities you can use to design an object on the computer and then create it. There is a laser etching machine you can use to cut a shape out of wood or acrylic, and there will (soon) be a "3D printer" you can use to design a 3D object in a computer aided design tool and then actually "print out" a 3D model of it. An article in the Daily Illini about this is here. According to some articles about the phenomenon, "fab labs" have proven to be a useful tool in empowering citizens of rural areas and developing countries to exercise their innovation.

Teaching Others

So, last week I had the first session of that 4-H robotics club activity I mentioned before. It went pretty much as I expected. Another student brought in a PowerPoint presentation about robots, and then they built a robot out of Legos from a kit and instructions provided. There were 15 kids, divided into four groups (because we had four kits available). They were able to finish building the robot itself, and in later sessions we will show them how to add sensors to the robot and program it. On the "guide to instructors" on the 4-H website it suggested a fun activity to teach them what programming is all about: bring in some peanut butter, jelly, bread, and tableware, have them write down a list of instructions for making a peanut butter and jelly sandwich, and then they read the instructions to me and I execute them. The point is to show how being clear and exact is necessary - for example if they write "put the jelly on the bread" that could mean put the whole container of jelly on top of the bread. Then they can see how to "debug" the instructions so they work right. The next session will be on November 14 so after that I will be able to tell you how it went.

But coming soon I won't just get the chance to teach elementary school age kids - I'll also get the chance to teach college undergraduates. It turns out that there isn't enough money in the grant for our project to fund me for next semester, so I will have to be a T.A. for next semester. I just learned this today so I don't know what class I will be a T.A. for, or even if I will actually be teaching students (I know that some TAs just grade papers). I will keep you posted with more information once I find out more.

Some more news

1. The Lego robotics thing won't be starting until the 4th week in October, and will only be once a month on the 4th Sunday of each month. So there will be no more news about that until then.

2. I will be presenting my work at the CS Grad Expo on October 4. I will try to get some sort of picture up then.

3. We just submitted our annual report to the National Science Foundation. If you want to see it I can try to put it up here soon.

4. I have gone to several Magic card tournaments in the past few weeks. I have started to get better at the tournaments and I actually got in first place in one of tham (out of about 12 people or so). The only problem is that now whenever I play multiplayer (that is, games with more than 2 people) everyone gangs up on me because they assume that I will beat them if they don't. It is impossible to convince them otherwise even if there is a situation where, say, I have no creatures with any useful powers on the board but the person across from me is just one turn away from using his power that will make all his creatures indestructible for the rest of the game.

5. At Belegarth, I have identified several things that I need to get better at, and I have devised a plan for doing so, as follows. These plans have not been implemented yet but I will post again as I see how they come out.

PROBLEM: Sometimes I can't remember who is on my team. Asking the target is rarely useful because it simply alerts them that I am about to shoot them. Not asking is problematic because it sometimes results in me shooting people on my own team.

SOLUTION: Take photos of as many players as possible. Write a computer program that will do the following: (1) display a random selection of these images, each labeled as "red" or "blue", then (2) display images one after another and ask me to identify which is on which "team." This way I can practice team identification.

PROBLEM: I am supposed to only "half draw" the bow back when shooting from under 15 feet. However I sometimes have a hard time determining whether it is 15 feet.

SOLUTION: Tape a piece of tape near the side of my glasses, with two tick marks on it. The distance between the tick marks will be calibrated such that the apparent distance between the tick marks is equal to the apparent height of an average-height target at 15 feet. (The principle is similar to the principle described here, except that I don't need to know the exact range, just know whether it is more or less than 15 feet.)

PROBLEM: Different people have given me conflicting answers as to what exactly "half draw" means. Basically, the "draw length" is the distance between the nock (the place where the arrow is attached to the bowstring) and the front of the handle of the bow when you draw it. The maximum allowable draw length for full draw is 28 inches. The question is that even if you just put the arrow on the bow in the "neutral position" and don't draw it at all, the "draw length" is not zero; it is about 7 inches or so (because the bow is curved.) So does "half draw" mean 14 inches (halfway between 0 and 28) or 17.5 inches (halfway between 7 and 28). I have gotten both answers from different players, and sometimes they tell me one thing but when I have them demonstrate and measure it, it's clearly something else.

SOLUTION: Bring a tape measure to practice. Have as many archers as possible demosntrate where they think "half draw" is, and measure it. Take the average of all these measurements, then put a "half-draw mark" on the arrow at that location. Get the half-draw mark checked by a herald (that's like a referee). Additionally, in case I am playing and there is a different herald who disagrees with the first herald on where half-draw is, bring replacement tape so I can re-mark the arrows if necessary.

A new semester

So a new semester just started. A few cool things have been going on:

1. I signed up for two classes: "Adaptive + Multigrid Methods" with Luke Olson and "Finite Element Analysis" with Dan Tortorelli.

The class on Iterative+Multigrid Methods is about techniques for solving large systems of linear equations. Large systems of linear equations arise a lot from approximating a partial differential equation by a separate linear equation at each point on the mesh. "Iterative" means that you solve the equation by coming up with an approximate "guess" and then repeatedly improving the "guess" until you get close enough to the solution. "Multigrid" means that rather than just using one mesh, you have several different meshes of different sizes, and you use the solution to the coarser mesh (which can be computed faster) in order to get a better guess for the solution of the finer mesh.

The Finite Element Analysis class is in the mechanical engineering department, so most of the students are mechanical engineering students. It is interesting to learn more about how the mathematical techniques I am learning about are actually used to model physical systems, although one problem (from my perspective) is that since most of the students are not computer science students a lot of time is spent going over basic programming concepts that I've already seen over and over. For example today the professor spent most of the class just explaining how to write a program that reads input from a data file and puts it into a matrix.

2. As for my research, we have gotten to the point where we can produce reasonably good looking visualizations of the meshing process. Once I finish that part (probably in the next few weeks) I am planning on making a web page where I can put them up so you can look at them.

3. In Belegarth, last week there was the Numenor "Opener" to mark the start of the semester, where lots of people come, including some from other groups, and they do lots of different battles. One of the battles was a "Unit Battle," where the different "units" (units are groups of people that fight together and often have distinctive uniforms) all fight. For that battle I temporarily joined a unit called House Valdemar. During that battle the leader of that unit (who is also the owner of one of the game stores I play Magic at) was so impressed with my archery skills he asked me to join the unit. The way it works is that now I am a "petitioner", and after a couple months the members vote on whether to keep me in as a full fledged member. So I guess it's kind of like a fraternity (not that I would know).

4. I have volunteered to be an instuctor for a 4-H club activity that teaches kids how to build robots using Lego Mindstorms toys. The way this happened was that one of my classes is in the engineering building, so when I was getting out of class I saw a flyer up on the wall advertising this, and it sounded really cool. It's going to be an hour once a week for 6 weeks, and it probably going to start in a couple weeks or so (they haven't set up the schedule yet). I did check to make sure it will end before December so it won't interfere with our vacation plans.