Archive for June, 2010

Wednesday Wii – Train your Brain 1

June 30, 2010

In artificial neural networks, when you are helping your net learn how to respond to inputs, the tried and true approach is to divide your dataset into three parts – train, test, and validation. The first, and biggest, part, is the training set. Think of this like the driver training track at a well endowed high school. The neural net runs through this dataset many times, perhaps thousands of times, in order to learn how to respond to the inputs. You train it, and train it, and every so often you run it through the training set again, in test mode, to see how well it has learned. Then you go back to training. Much less frequently, you show it the second batch of data. This is the much smaller test set. The object is to see how it works with data it hasn’t been trained on. Think of this as taking your high school driver out on the local roads. You know your system has learned all it is going to when the results on the test set start to diverge from those on the training set. At that point you give it its drivers license test by showing it the third, and usually smallest dataset, the validation set. You only get to use this dataset once, and it tells you how ready your neural net is to face the wide, wild, world.

What has this to do with the Wii Fit, you ask? Well, the Wii Fit is training your brain as well as your body. In the balance tests, the key is to get your nervous system to respond to the inputs from your balance organs so that you can stand on one leg without falling on your butt. Think of all those exercises, from the Tree Pose to the Single Leg Extension as the training set. As you do them, your internal neural net system learns to keep your balance under a wide range of conditions.  Periodically, you get to take the Body Test, where you are exposed to, yes, a test set of actions that are not the ones you trained on. And therin lies the rub. If you go into the My Wii Fit Plus section, you have the opportunity to build yourself an exercise routine. In addition, if you click on the little Wii helper you get a chance to practice the various tests it gives you during a  Body Test session. This is generally a bad idea, because what you are doing is training on the test set, and so I refuse to do it, with two exceptions, that I’ll tell you about next time.

The Ergotron Lx Keyboard Arm

June 28, 2010

I run multiple computers, a Ubuntu Linux box for most of my work and research, and a separate Win XP PC for those apps that have to have Windows. Yes, I know about VBox and Crossover and Wine, but sometimes it’s just more convenient. Now, the approved way to move between computers is a Keyboard-Video-Mouse (KVM) switch. I’ve not had much luck with KVM switches, they rarely last more than a year or so, and the latest one has a tendency to lock up linux so bad that I have to do a power cycle to get my mouse and keyboard back. The solution I’ve hit on is to use the KVM switch as a simple monitor switcher, and plug a separate keyboard and mouse into the PC. This, of course, means I have to keep sticking them out of the way and pulling them back out whenever I want to use the machine. Enter the keyboard arm.

After a whole quarter of pulling and putting, with intimations that I’ll be teaching the same class using the same software, I broke down and ordered a swivel arm for the keyboard. The one I chose was the Ergotron Lx Wallmount – surprisingly, there aren’t a lot of others out there (which may explain part of the price). It came today, and I thought I’d build a review as I went along, maybe posting a copy of it to Amazon, whence I bought it.

Packaging was simple, but interesting. The densely packed OEM box was inside a much bigger Amazon box, along with a large crumple of packing paper. The Amazon box was lighter than it looked. The Ergotron box, heavier. Maybe because the advertised weight of the system is 16.7 lbs, while the shipping weight is reported as 16 lbs. Go figure.

Inside the box that was in the box, the arm components (shelf, humerus and ulna (or biceps and triceps), and the wall mount) were well packed and well protected. One also gets two bags of smaller parts, and a six page, nine language instruction sheet, mostly pictures. The pictures are clear, which is good, because the text isn’t particularly useful. It isn’t unintelligible, just extremely terse, and limited to things that are hard to draw, like the fact that they recommend an installation that can hold up to 4x the weight of their system. They are serious — the screws they give you are two 10x80mm concrete screws that look like they’d anchor the tiedown hook for the Hindenberg.

Of course, the mountings have to be strong, because the arm itself is solid. Beat your way out of a crowd of zombies-solid. And heavy. I doubt that 0.7lb of that 16lb is packing material.

Installation was simple. Line up the wall mount holes. Drill. Screw. Assemble. Since it’s a “wall-mount” setup, they want you to put it into two inches of concrete wall, or into the thick part of a 2×4 stud. I was mounting mine on a computer desk, so I drilled two 1/4″ holes through the particle board and bolted it on using a 1 1/2″ long bolt, with a washer, lock washer, and hex nut on the inside. If this causes any durability problems, I’ll report back.

Assembly was equally simple: drop the humerus onto the wall mount shoulder, drop the ulna onto the humerus, screw on the palm (keyboard tray). That last part was the trickiest.  The keyboard tray is heavy, the screws are small, and the arm wants to move away when you approach it with a screwdriver. Getting the first screw in is really a two person job.

They give you stickons to keep the keyboard from wandering around and to help route the cables up the arm. Right now I’m letting them hang off the back, but I can see why you’d want to restrain them.

The tray itself is sized for a standard keyboard. My big old ergonomic aircraft-carrier style keyboard hangs over a bit on the sides, and an inch or so in the front.  The action is very smooth, both horizontally and vertically, and the tray stays where you put it. I have mounted it at the near edge of the computer table, and it articulates enough to put the keyboard in front of me, squared up to the monitor. There is an extension on either end that will hold the mousepad. It’s smallish (7×7) and it only sticks out on one side at a time. No using one side for the mouse and the other for your coffee cup.

So far, I only have three complaints. First, the mouse tray slides too easily, so if you push on it when moving the arm it wants to dump the mouse. This is evidently a feature, not a bug, because they give you a wall-mounted holster where you can put the mouse before you stow the keyboard for landing. Second, there’s a surprising amount of play in the joints, so that in certain configurations, deploying the mouse pad causes the tray to tilt in that direction — over an inch. The mouse won’t slide off, but marbles won’t stay on. There is also some shake in the arm when you type, which is to be expected, given that you don’t have three point bracing. Finally, the primary vertical wrist control (flapping your hand) is tightened via a recessed hex nut. They give you a wrench to reach it with, Unfortunately, the angle is off just enough that the end of the hex wrench can’t fit into the socket unless you move it to an extreme up position. The wrist is positioned well enough and is tight enough to not be an issue, but if it was an issue, fixing it might be a problem.

I’m giving it four stars.

I have published a five-month update.

Wednesday Wii, Happy Anniversary

June 23, 2010

Happy Solstice, and the start of Aere-Litha. So, this is Week 104, and on Friday it will be exactly two years since we started using the Wii Fit and WF+. In those two years I lost 25 pounds, and gained back five, lost a belt size, and immeasurably improved my balance and flexibility. In the Sun Salutation, when it said “touch your toes”, I was initially straining to get my palms on my kneecaps. Now I can, with a little bounce and at the end of my workout session, touch my ankles. Small victories.

My problem right now is that I otherwise seem to have plateaued out. My weight doesn’t want to change, my balance hasn’t continued its improvements,  my scores are within a few points of what they were six months ago. More aggressive measures are needed.

This summer, I publicly (well, as public as a one hit a day blog allows) declare I am going to do more Wii, more off-Wii walking, and more off-Wii balance training.


June 17, 2010

Anthony Cordesman recently published a commentary on the current state of affairs in Afghanistan. Comments on the commentary were then made by Abu Muqawama, Max Boot, and Michael Cohen. Let me add my own two cents.

First, he questions the strategic reasons for the war:  “The US has no enduring reason to maintain a strategic presence in Afghanistan or Central Asia. It has far more important strategic priorities in virtually every other part of the world….” Then, he gives us one: “The key reasons for the war remain Al Qa’ida and the threat of a sanctuary and base for international terrorism, and the fact the conflict now involves Pakistan’s future stability.” OK, two, actually, three. All are likely to be wrong.

Let’s look at AQ and international terrorism. These can be seen as two different concepts. AQ is a coordinator, an inspirer, a financier — one might even say venture capitalist. It doesn’t need Afghanistan as a sanctuary. It has places all over the world it can go to. What it needs is a secure communications system, and access to large amounts of funding that it can funnel to the people it inspires. Most of that requires the services of a modern economy (although the Moslem informal banking system is apparently a help). It doesn’t particularly need training camps, if what it is doing is indoctrinating a few people and showing them how to take over airplanes. The best agency to defend against AQ right now is probably the Treasury Department, followed by the FBI, and FBI liason officers at the US embassies.

The groups clumped under the title international terrorism appear to be mostly involved in training foot soldiers and NCOs and exporting their skills to insurgencies in other Moslem countries. They require camps where people can run around, shooting AK-47s, but they are not a major threat to the US, except where we are on their soil. Most of the output of these camps appears to be going to AF and PK, not to NY.

The stability of PK wasn’t an issue until we started pressuring them to push forces into the tribal areas of the NW frontier. If the PK government were to stop, to go back to ruling this region through benign neglect, the single most potent reason for a destabilizing insurgency would disappear.

We solve the international terrorist problem, and possibly much of the AQ problem as well, by removing the source of their recruiting appeal. We leave AF to the Afghans. We reduce pressure on the PK government to stir up the tribes on the NW Frontier. We continue our pullout from Iraq. We change our Middle East policy to be overtly more even-handed between Israel and the Palestinians, instead of being knee-jerk supporters of whatever policy the hopelessly fragmented Israeli government comes up with this week.

The second thing that Cordesman questions is whether or  not the war in AF is “winnable.” He thinks it is, but not with the original goals. He thinks we can achieve a “good enough” solution to make it seem like we have won. This is a far cry from what we wanted. It’s still probably unachievable. One commentator was quoted as saying that you had to spend a year on the ground in AF to be qualified to make any kind of statement about the country. I haven’t done that. Neither have our policy makers nor most of those who advise them. Even the reduced goals that Cordesman has espoused seem to have little to do with the overt strategic reasons he gave above. Here’s a not-very-insightful list of what I think the spectrum of post-US AF possibilities looks like.

1. A return to the Taliban

2. A stable government, with insurgent participation,  probably not pro-US, but not necessarily anti-. By stable government I mean that they have, you know, a flag and a head of state, and some way of guaranteeing the safety of embassies. I don’t necessarily mean that the government’s writ runs beyond rifle shot of a paved road.

3. A pro-US government, capable of defeating the insurgents over the long haul

4. A western-style democracy

The last is not achievable in our lifetimes. The third one will require a decade of US military presence and casualties, with all the domestic angst and international terrorist recruiting fodder and PK instability that implies. The first two are the most likely results of the US ending its presence in AF in a timely fashion.  I have no idea which is more likely. My position is that any of these, once achieved, solves most of the problems and answers most of our current reasons for being in AF. The question is, which can be arrived at with the least expenditure of lives and treasure. I’d say that getting us out faster is better.

Reading List Page

June 16, 2010

In response to user demand (i.e. one search in a two year period), I have decided to move my reading list from its obscure position as a long-dead post to a new glory on its own page. Look to your right, under Pages.

Wednesday Wii

June 16, 2010

Back to my Wii Fit excercises, after a three month absence. This quarter was stressful because of my mother-in-law’s illness and because I had two somewhat unexpected statistics classes to teach. So, while I was able to do a weigh-in and body test daily, I had no time for the actual exercise regime (or for cleaning out my RSS feed, which now stands at 400, down from 800 earlier today). Now, the quarter is over, and I can get back to work.

My first session was only 30min long. I found that those exercises not involving balance  (warrior pose, half moon) went about as well as they had before — I even made the top ten on one exercise. On each of the exercises requiring balance, either static (tree pose), or dynamic (sideways leg lift), I was down about twenty points. Shoulders and back feel the stress of renewed exercise, but nothing painful. Weight holding steady at 215, as it has (+ 2lbs) for the last three months. Next week is week 104, the end of two years of Wiiing (or is it Fiting?), so we can call the upcoming year a New Hope (unless somebody’s used that already).


June 13, 2010

So, based on this recipe for glazed carrots, I decided to try the veggie side of sous. Then, while fiddling with the crockpot settings, I had second thoughts. The key to VSV is that the pectin in vegetables needs to be heated to above 180F before it will break down. That’s pretty close to a normal simmer of <212F (OK, 30 degrees below, but it’s not broiler-meat difference). It might be that (a) the temp doesn’t matter so much, since it’s not so far off, and (b) the exclusion of the water does matter. That is, this is a real sous-vide issue, and not a low-temp issue.

As I understand it, when you are cooking vegetables there’s two main techniques: submerged in a liquid — water or oil — or submerged in a gas — steam or air. If you boil/simmer them, the water takes much of the flavor. If you fry or braise, you run into the same issues you have with meat — the heat source is much hotter than the desired temperature. If you steam them the juice runs out. If you roast them, the juice evaporates. Cooking in a bag in not-quite-boiling-water would seem to have all of the advantages. It preserves the juices, and it doesn’t overtemp the outside.

Using the crock-pot without the insert, I found it was relatively easy to hold the water temperature between 188 and 192F. Warmer than called for, but easily done. Due to scheduling problems, the carrots spent more like 90min in the water, and even with the air sucked out the bag didn’t want to submerge. Various heavy objects placed on top didn’t help, unless I was rehearsing for a Funniest YouTube skit. Finally, I gave up, and just turned the bag periodically.

These are “baby carrots”, i.e. real carrots what have been cut into smaller carrot-shapes. First taste prior to browning showed them not-soft-but-not-crunchy, and hot all the way through. Need salt (I didn’t use any).

The pork chops are almost ready. Time to go do the final processing on the carrots.

————- munching sounds ————-

The result? Not bad. More liquid than expected — it took 4min to cook off, instead of 2. Not enough sugar, so that it wasn’t really glazed. Worth trying again, but not the source of true carroty goodness. I’d give it three stars, out of a possible four. It really needed salt.

They hate us for our freedoms 9

June 6, 2010

This has been a hectic quarter, and I haven’t had time to do much Deep Thinking on Important Topics. I hope to wrap up this THUFOF series come summer (i.e. mid June).

One thing that has come to mind, that I thought I’d throw out into the bit-stream is the incompetence of the Times Square bomber. He didn’t know what fertilizer/oil combination to use, he didn’t know to open the propane vents, he didn’t know how to make a decent fuze. He didn’t know enough about stealing cars to know there are multiple VN ID locations. He was an incompetent loser.

No he wasn’t. He was an ordinary guy. Desperate, but ordinary.

He didn’t know these things because he was, at one time, like us, a normal, law-abiding citizen. He was in desperate straits. He finally snapped. He’s like an office worker who goes on a rampage with a letter opener, because that’s all he’s got. Now, admittedly, this wasn’t the action of a moment. He visited Pakistan. He moved his family. He planned. But even with a running start, he was such a normal, gormless, guy that nothing worked.

Would you have been able to do any better? Would your neighbor?