After only knowing my classmates through electronic messages for two weeks, I was finally able to meet them all face-to-face last night. They are a really smart group of friendly, easy-going professionals about my age. The two profs involved in our first class also seem friendly, dedicated, and motivated to make sure we all finish the program. Then I went home and stayed up a little too late posting follow-ups to greetings and responding to posts on the class website, so I'm really pooped today.

One thing I'm a little worried about is that 4 of the 7 students in my cohort have been through part of the program before. Since they've actually completed several classes already, they won't have to participate or come to classes regularly until later this year. I was looking forward to the small group experience many people have described to me, but three people seems ridic. There's only so much feedback I can imagine giving to the same two people for months on end before I sound like a broken record--it's not like I'm a career counselor or a psychologist. I guess I am living in interesting times, so we'll see how it goes...

Molly called me from Israel yesterday on a real phone. Not for snuggle talk, but to have me walk someone through playing with their cable modem to see if I could fix their internet connection. Their alternative connection has apparently dried up, as well, but it was really bad in the first place; IMs go undelivered and/or arrive in batches 10 at-a-time. She can't really access many websites or check her email yet.

This fancy "high-speed" connection was supposed to be hooked up before they got there. It wasn't working when they arrived, so they called the cable company. Two techs played around with it all day long, managed to get one ancient G4 mac laptop connected through it via ethernet, and then left. The cable box seems hardwired to the MAC address on the...Mac, so we tried to clone the address and apply it to their wireless gateway. Although the crappy cable box gave the crappy wireless AP an address, the two refused to route any traffic between each other. Later in the conversation, I learned that have to start a PPTP session from the Mac to the cable box to even use the internet, and even then it runs at like 3kb/sec. Holy WTFs, Batman! We tried to restart the cable box at one point, but then the cable connection itself decided not to come back up (the specs so flashing cable light forever == call the cable co and pray). They are screwed. To make things worse, the guy I was talking to handed the phone back to Molly just in time for the battery to die, and she hasn't called me back or IMed since then. SAD FOR ME.

This really sucks. Andrea assured us all that family communication has been possible in the past through a decent net connection, and that they'd have a fancy new high-speed setup this year. So far I've only been able to (poorly) IM Molly twice. If she can't even check her email, there's no way we're going to be able to skype  :(

My cat was pestering me too much tonight, so I decided to distract her with some catnip. We don't have any toys I can stuff it into anymore, so I just poured out the loose-leaf nip that came with her cardboard scratcher box, and rubbed it in. It took her a minute to realize what I was up to, but then she went APE! She started meowing up a storm and scratching the crap out of that box, and generally purring and rolling around on it and carrying on much more loudly than before. She has been stratching at it and sniffing it for an hour now. Oiy, I should have just let her meow at my chair a little longer...

Molly confirmed that she is alive and well, it just took longer than expected to get their internet connection working. I missed her (she IMed early this morning), but hopefully we'll be able to talk a little later in her day.

Molly has left the country in search of ancient ruins in the Israeli desert, and I'm all alone for the next two months. SAD FOR ME, but HAPPY FOR MOLLY, because she likes scratching in the dirt looking for stuff. I MISS YOU SWEET-PEA!

Anyhow, I have once again started school. I'm finishing my undergrad at Concordia University in Saint Paul, and I'm taking Japanese at the U of M again this summer. I've been running a lot on the treadmill (thanks to Zach for helping me work some kinks out of my new shoes), and I want to keep up the physical fitness momentum. While molly is gone I'm going to try to ride my bike as often as possible--hopefully I'll stay out of the car altogether, barring emergencies or downpours when I need to be at school. 私の自転車が大好きを乗ります！

My prediction for NPR in 2018: 'Carbon Traders Leave Fouled Environment, Popped Economic Bubble'. In an interview last month by NPR's Christopher Joyce, a lot of emphasis was placed on how fresh-faced college grads in DC can acquire a fancy brownstone in Georgetown, a six-figure starting salary, a bunch of pseudo-bohemian friends, AND save the world at the same time by becoming carbon traders. Joshua Albin (alleged "whiz kid") describes his profession this way:

"Carbon trading is super-sexy. I can't tell you how many friends I have that are in big law firms or investment banking or in med school that just fall all over themselves because this stuff sounds so cool."

I can't tell Josh how many get-rich-quick schemes seemed super-sexy at first, only to leave giant messes for other people to clean up later. Also in the story, Nathanial Bullard, a guy who "used to teach school in Egypt; now he's an expert in green energy projects" (talk about your non-sequiturs) implies that a pig farmer in Iowa is sitting on a pile of cash with his "methane-oozing" manure lagoon. Interviewer Joyce eats it up, explaining, "and in fact the farmer could make money on the market by capturing that methane," completely glossing over the fact--or (worse for a journalist) being blissfully unaware--that a manure digester is an industrial operation requiring engineering oversight and a multi-million dollar investment to build and operate--decidedly not something a family farmer would attempt without a commercial partnership and a massive subsidy. My mom was involved in a small-scale manure digester project here in MN, and it's no small undertaking. Also, is the pollution emitted just to construct this new infrastructure factored into the trading scheme? This is just one tiny aspect of the "market"--but one of the only practical examples given--and questions of efficacy and enforceability abound.

Personally, I smell the smug stink of the middlemen who contrived the NINA loan (No Interest/No Asset) when the market ran out of legitimate loans to pack into into what turned out to be worthless mortgage-backed securities during the last few years. And, truly, "mortgage security broker" was the last profession where an interested kid could invent a position with a six-figure salary for doing almost no work at first--and what's looking like criminally negligent work later on. As with that profession, my feeling is that the carbon credit trade is just the next group of idiots our tax dollars will be bailing out in 5-10 years, and here's how: since there is zero enforcement in the carbon trading market, some genius will eventually realize he can get the farmer to just promise to capture methane (or sign a cancelable purchase order for the necessary equipment) in order to generate the credit revenue stream. The genius then buys the revenue agreement from the farmer to package into an investment security. Meanwhile, the coal-fired power plant in eastern Europe that is paying for the credits eventually catches on when the NY Times finally gets around to reporting that the market is a sham. They promptly STOP paying, the bubble bursts, and then we realize we just let the same polluters crap out mercury, uranium, and thorium on the population with impunity (but remember: only the carbon is bad!!) for yet-another-generation. Color me skeptical.

I couldn't tell how I felt when I originally listened to the story, so I didn't blog about it at the time. Today, however, Wired published a series of trolls on environmentalism, and the carbon market was one aspect I thought they bashed rather well. While commenting, I realized I had more to say about the potential for bubble-producing shenanigans. Note: I called these "trolls" because there is a distinct lack of primary research (you know, the stuff journalists used to be expected to perform), so they just substituted glib self-assuredness in its place--wait, that's MY job!

I'm flying out later to see how I can help.

14 tons of spilled Oreo cookies snarl Ill. traffic

I just had the misfortune of being stuck following behind your sandwich-eating self through downtown Saint Paul in my car this afternoon. You definitely didn't notice between mowing that whole sandwich in a 3 block stretch and blaring gospel music out the rolled-down windows of your tinted SUV. I was still behind you at the intersection of Wabasha and West 7th when you wadded up your fast-food garbage and threw it out your window at the red light. I assume that the "I Love My Truck" sticker on your rear windshield is at fault--obviously the quality of your Tahoe SS is so high that you cannot bear to soil it with the refuse you collect while polluting your body.

I know people litter, but I haven't seen an adult do this so brazenly since I was a small child. You ma'am, sincerely suck hard. You are a stentorian example of the hypocrisy we see dressed up in religiosity every day. Congratulations on being a cliché, and please return those vanity plates to the DMV so a genuinely decent person can have them.

Our cat Moki loves it when we leave the sleeping bag on the couch:

I have an HP TC1100 tablet PC that's almost four years old. It has a 1.1Ghz Pentium M chip and 1.5gb of ram, and it runs Windows XP. I  do all of my personal software development on it, and the tablet functionality works great for taking notes and doing handwriting recognition exercises for Japanese class. The original hard drive has long since died, as well as the original battery. While in Thailand, Molly and I got great deals on replacement hard drives when our originals started to go bad during the year we were there. I got a 7200rpm 80gb drive, and Molly got a 5400rpm 160gb drive.

My tablet has since suffered several falls, and the most recent finally killed the new drive (and bent my case). Since I needed a new drive anyway, I decided to try something I've always wanted to do: build my own flash drive. Addonics sells a device that allows you to connect 1 or 2 compact flash chips to a laptop IDE interface, and use them like regular hard drives.

You can't just use any kind of CF card though, because most of them are intolerably slow and/or can't deal with multiple simultaneous reads/writes without dropping down to modem-like speeds. The card needs support UDMA at a minimum, but watch out for the cards that claim two speeds because the slower of the two (or worse) is what you're likely to consistently encounter when using the card as a hard drive. The first card I bought was the 16gb version of the RiData Lightning. It sucked. Took 3 hours to install XP, and it couldn't deal with reading and writing at the same time. Single reads or writes occurred at the stated speeds on their website (i.e., relatively fast), but simultaneous reads or writes brought throughput down to below 2mb/sec most times. Performance as an OS/boot drive was awful, but I assumed a $60 card with a limited warranty and lots of hedging about actual read/write throughput would be likely to suck, so I ordered a fast one too.

The PNY Optima Pro UDMA has a single rating for both read and write, a lifetime warranty, and wear leveling support. It was no surprise that the PNY card was ridiculously fast under all circumstances. It felt just like my old hard drive (only 1/10 the size), and XP installed in about 20 minutes. Since it was only 8gb, I still needed the space the 16gb RiData card offered, but I didn't want to kill my overall performance. Here's how I accomplished this:

1. Formatted both cards as NTFS
2. Installed XP on the PNY
3. Assigned the RiData as the D drive
4. Mounted both cards on another system (to gain exclusive access)
5. Moved the contents of C:\Program Files to D:\Program Files
6. Use Sysinternal's Junction utility to link C:\Program Files to D:\Program Files

I also created D:\Junctions to store a few other strategic links to folders on C with large amounts of static data, like the MSDN docs in the All Users profile, and the hotfix uninstall blob under C:\Windows\$hf_mig$ (I also accidentally blew up one XP install by creating a junction to C:\Windows\Installer, but I wrote about that before, and yes, the Windows Installer team are still idiots).

The net result for me was a working system that had a damned fast boot/OS drive and about an extra hour of battery life. It was a very rejuvenating experience for a four year old laptop and battery ;)

I did run into a snag a few weeks ago: the system was choppier than normal and seemed to be taking about 4x longer to boot. I don't reboot very often, and I was under the gun to develop some software, so I ignored it. Last week I played some h264-encoded HD video over wireless while I was on the treadmill, and it was dropping frames left and right and the video was out of sync with audio 20mins in. It seemed my speed issues weren't going away on their own, so I installed Windows XP SP3 a few days ago, hoping to remedy the situation. It really seemed to speed the drives up (the chop was noticeably lessened when accessing the slow drive), and the hi-def video frame rate was almost back-to-normal, but there was definitely still something wrong. Then I checked device manager and discovered that the IDE controller had changed to PIO mode at some point.

This is critical for performance of a flash-based system overall; PIO means the CPU is programming all of the memory access operations performed by the IDE driver. DMA/UDMA means that the device is programming the operations itself and leaving the CPU to do more important things (like decoding video and accessing the network). I mentioned this to my friend Dennis, and he sent me a link to an article with a method for forcing Windows to re-detect your drive capabilities. After modifying the registry, Windows rebooted faster than the day I first installed it, so I knew something was going right. UDMA mode was once again enabled, and I was free to roam about the internets.

Well, it turns out that the reason my controller reverted to PIO mode in the first place is because the cheap RiData card has been throwing a lot of errors during IO--typically when I would launch Visual Studio. Even though this drive is only used to launch apps and store ebooks for jogging on the treadmill, it only lasted three months before breaking down. Fortunately for my data, it broke unlike any mechanical hard drive I've ever owned: I can read every single file on the drive and even create a flawless ghost image of it, but if I try to do something like launch an application that might do multiple reads against the drive, the system log fills with disk and atapi errors and the machine eventually bluescreens if I keep it up.

I'm exchanging the broken RiData for a new one, and letting Molly use it in her camera where it can't do much harm. I'm replacing it with a 300x 16gb from Transcend. Lifetime warranty and wear-leveling, here I come ;) I'll probably reconfigure the machine to run everything on the fast 16gb card and get rid of the junction points. I'll leave the 8gb in the machine as D. My small development virtual machines should run great on it.

So I've been running a lot in the last few months, after a year of extremely low physical activity since we returned from Thailand. After Provamation moved from the Energy Park office to our home offices, I didn't have a nice 10mi/day bike ride to look forward to, and I really let my physical activity level dip. Around February, I started walking, then running a few days a week on the treadmill, and I've put a lot of miles on it.

Right now I'm running 3 days on/1 day off, 6 miles per run. I wear a heart rate monitor and try to stay below 155. Smartwool socks make all the difference in the world; I made the terrible decision to run in cotton socks during my first three runs of the year, and I still have the blister scars to prove it. Molly and I bought smartwool socks for our trip to Thailand in 2006 because they breath exceptionally well, and in that heat, and it's important to have something light when you can't wear sandals. We bought the ultra-thin ankle versions, and they've been nearly indestructible so far--at least two batches of cotton socks have worn holes during the same period.

I reached a personal goal on the treadmill odometer last weekend, which I set to determine when I should by some new shoes. My friend Zach took me to Marathon Sports in Minneapolis, and I bought a new pair of shoes and some proper running shorts (cotton + sweat just doesn't mix). Compared to my old shoes, it's like running with two big pillows strapped to my feet. Definitely a good feeling.

So how can run on the treadmill all the time when the great outdoors is just so damned beautiful? Well, I like to read while I run. I have a lot of audio books to listen to when I want to be entertained, but I prefer to study while I run. There are some excellent philosophy and science books which lend themselves to the audio format, but very few technical books (esp. on software or language) which work in that format. The treadmill is great for reading because I can just open the book and zone out; running is just something my body is doing while my brain is thinking about other things, and I don't have to worry about veering into the street and being hit by a car if my mind wanders too far.

So, I could read a physical book with the aid of one of those plastic book holding thingies, but I don't use that method too often as I don't enjoy the posture my upper-body takes to see the book in the holder, and turning pages when you're sweaty isn't fun at all. My solution is to read ebooks (native PDFs or books I've scanned with a book scanner) using a monitor mounted at an angle complementary to my normal running posture and treadmill location.

Overall, it's a great space for running + reading. The open window lets in a nice, cool breeze this time of year, and between our courtyard garden and the park across the street, there are enough birds, critters, and children to tune out most of noises of the city. Hopefully school will give me new reasons to do running + reading instead of making it harder.

This show is by turns one of the most incredible, depressing, and informative episodes of This American Life that I've ever heard. I listened with rapt attention to how the mortgage-backed securities industry went from reliable, safe, fixed-interest mortgages into a death spiral where at one point $500k+ loans to were handed with ZERO verification to people with no income and no assets (they even had an acronym for them: NINA) just to satisfy the hunger of the investors. Then one day in October of 2006, everyone noticed that these NINA securities they'd packaged up as AAA-rated investments actually contained huge numbers of mortgages where the owners defaulted on their first payment. Then things got really bad.

A special program about the housing crisis. We explain it all to you. What does the housing crisis have to do with the collapse of the investment bank Bear Stearns? Why did banks make half-million dollar loans to people without jobs or income? And why is everyone talking so much about the 1930s? It all comes back to the Giant Pool of Money.

This American Life - The Giant Pool of Money #355

I have two external 500gb drives on my main machine that know how to go into power saving mode when they aren't needed. For some ridiculous reason, Acrobat always spins them up when I scroll two or three pages into a document. This usually happens while reading PDFs through my browser, which causes the browser to freeze up while the Acrobat plugin blocks IO, and it can take up to five seconds if both drives are spun down. I can't find any reason for this to happen (no env vars point to either drive, no adobe software or PDFs on either one), but I have found a few other people on the net complaining about it and trying to figure out how to make it stop. Another reason why Foxit Reader rules and Acrobat drools.

Avoiding spin-up on external SATA drive?

My friend Dennis and I are studying together for the .NET 2.0 MCTS tests. We started with the 70-536 book, Microsoft .NET Framework 2.0 Application Development Foundation. It's a great resource if only because it represents a fairly comprehensive list of the base class library capabilities in a format readily consumable in a learning context. Like all MS Press books aimed at certification, each chapter begins with a list of test objectives covered, lessons that break down the objectives into learnable units, labs, review questions, case scenarios that apply what you've learned to a set of requirements, and suggested practices that test your ability to improvise.

I've been writing .NET applications for about five years, so I had a lot of experience at the start of the process. After reading the book and doing a once-over of the review questions from all of the chapters, I took the exam. I passed, but it was hard one. According to my transcript, I've taken 21 different Microsoft certification exams since 1996, and this was the first one I thought I might not pass on the first try. It wasn't for lack of preparation--the base class library is simply enormous, and they give you 3 hours to answer 40 questions from about 20 distinct and disparate areas of the framework. A few of the questions amounted to having memorized some fairly obscure methods and properties to pick the correct code sample for a given requirement from a group of extremely similar code fragments, but overall it was just a fair (if exhausting) evaluation of your knowledge of the framework. To be sure, during the feedback period after the test I wrote extensive comments on the questions I thought relied on lucky memorization of specific APIs. I feel that the test should get away from questions which Intellisense (or F1 help) can immediately resolve, and focus whenever possible on determining the examinee's understanding of how properly to apply the BCL's capabilities, and whether or not the examinee understands the when and the why of it all.

Despite my professional experience, I don't think I would have passed the test without reading the book. The BCL is just too broad for most developers to cover entirely--let alone memorize every API as they go--even over a period of years. Unless you have the luck to work on such an enormous array of projects and in such varied roles as to have practical experience with all of the major areas, you should consider the book invaluable. For example, my first professional exposure to the drawing, encryption, and code access security APIs was from three totally unrelated projects at different organizations--and I still learned a lot by reading those chapters in the book (as one would expect).

Anyhow, we're reading the WinForms book next (for 70-526), and the web book after that (for 70-528). I don't know anything about the "Professional" certs (the MCPD exams), or whether they incorporate new learning or just the MS "way" of planning and testing like the old WOSA books (which were tedious in the extreme), but I'll look into it eventually. I did notice that the WinForms pro cert book has several used copies available for ~$3 on amazon...that can't really be a good sign since I'll definitely be keeping my 70-536 book around as a reference.

I've registered for a degree completion program for adult learners at Concordia University. The program will take 20 months to complete, but only gives me the core credits for the degree (a hybrid of IT and management), so I'll need to collect the difference between the 128 I need and the 40 I already have between additional Concordia classes and the U of M. It's been three years since my last Japanese class, but I really want to take three years at the U of M, so I'll be starting again with a summer class and continue with evening classes during the regular school year. I'll write more as I learn specifics about the Concordia program, and I'll probably annoy the crap out of anyone reading this with inscrutable Japanese language posts when I reinstall the IME on my tablet ;)

I have no idea why VS2008 needs 4 minutes to build a simple ~600kb MSI file for the debug build of my .NET 3.5 project. I also have no idea why it insists recopying all of the prerequisites even when they already exist in the release folder. The "use same folder as my app" option for the 3.5 framework prereq causes it to regenerate a 200mb folder of setup binaries every time. Oiy! Granted, it runs significantly faster on my dual-core desktop with 2gb of RAM and 7200rpm drive, but I'm doing all of this on my 1.1ghz Pentium-M tablet with a home made flash hard drive.