Eric Aderhold's Web Site

The election results are in, and the winner of the Minnesota senate race appears to be Norm Coleman, the first-term incumbent Republican senator. Coleman’s Democratic challenger, Al Franken, finished behind by only 725 votes out of over 2.8 million cast. This razor-thin margin will trigger an automatic recount under Minnesota law. Regardless of the recount’s outcome, however, Minnesota’s next senator will go to Washington with the support of a whopping 42% of the voting public. Former Senator Dean Barkley, the Independence Party candidate, finished in third with 15%.

42%. Think about that for a moment. The will of the people is supposed to reign supreme in our nation, yet we just elected someone who 58% of the voting public voted against. The pundits always try, in elections like this, to figure out what would have happened if the third party candidate had not run. The answer in this case is that there is no way to know for sure. Exit polls suggest that Barkley supporters were just about evenly split between Coleman and Franken for their second choice. Perhaps more interestingly, 45% of Barkley supporters polled might not have voted at all if Coleman and Franken were their only choices. Thousands of Minnesotans who care about our country felt that both major parties were so out of touch with their ideals that they would have rather given up their voice in this election than vote for either Coleman or Franken. That’s an important topic in and of itself, but it is not the subject of this post.

Races with no majority winner have not been at all uncommon in Minnesota recently. Due to the continued strength of the Independence Party, the last three gubernatorial elections have been won by plurality, as have three of the past four senatorial elections and two of the eight U.S. House seats this year. Instead of endlessly rehashing the “what-ifs” in order to convince ourselves that a 42% winner really does (or does not) capture the will of the people, why not switch to an electoral system that guarantees a winner that the majority can live with? Instant runoff voting (IRV) is just such a system. The way it works is simple. With IRV, voters rank all of the candidates from favorite to least favorite instead of just picking one candidate. At first, only the first choice votes are counted (just like our current system). If a candidate has at least 50%, he or she is declared the winner (just like our current system).

Where IRV differs is in multi-candidate races with no majority winner. In that case, the candidate with the fewest first choice votes is eliminated and all of the people who voted for that candidate have their vote automatically switched to their second choice. If a candidate has a majority after this, he or she is declared the winner. If not, the last place candidate is again eliminated until a majority winner is found.

I argue that IRV is much more democratic than our current system. It allows all citizens to vote for the candidate they most want to see elected, regardless of party, without fear of inadvertently helping the candidate they dislike the most to be elected. Here’s an example:

Suppose you have a race where 40% of the electorate really believe that major party Candidate A would do the best job, 30% believe major party Candidate B would do the best job, and 30% believe third party Candidate C would be the best. Supporters of Candidate A all greatly dislike Candidate B, and vice versa, so their second choice would overwhelmingly be Candidate C. Candidate C’s supporters are split. 70% believe B would be better than A, and 30% believe A would be better than B.

In our current system, A and B would scare three quarters (or more) of C’s supporters into pragmatically voting for one of them because “voting for a third party is throwing your vote away.” So the election would result in A winning with 46.75%, B coming in second with 45.75%, and C bringing up the rear with 7.5%, even though the majority believes that either B or C would be better than A.In an IRV system, the first round would show a lead for Candidate A, with 40%, and the other two roughly tied at 30%. Suppose B got a few more first choice votes than C. C would be eliminated, second choice votes tallied, and B would beat A 51% to 49%. If, on the other hand, C got a few more first choice votes than B, B would be eliminated, second choice votes would be tallied, and C would win 60% to 40%. Either outcome would be a much better compromise than what you get with our current system.

Instant runoff voting is not an untested, alien concept. Several democratic governments use it, including Australia, Ireland, and San Francisco. If you believe that IRV would be an improvement over our current election system, send letters to your state representatives. Real ones. With stamps. Tell them that you are sick and tired of a minority of Minnesotans choosing who will represent us, and you want to enact a constitutional amendment to switch to a better system. If they refuse to sponsor the bill because they or their party has benefited from the current system, never forget. Vote against them in the next election. Run yourself if that’s what it takes. This election showed that the people of this great nation want real change in the way they are governed. Make it happen.

I was doing some Christmas shopping on Amazon.com this evening, and my total came out to $24.19, a mere 81¢ short of qualifying for the free super saver shipping. So I did what any self-respecting technologically-literate cheapskate would do, and performed the following Google search: ““$0.81 & eligible for FREE Super Saver Shipping on orders over $25.” site:amazon.com.” That search gave me two options: an ROL Gaskets WO8311-001 Water Outlet Gasket or a Prayer Book for Graduates. However, the prayer book seems to have been discounted to 64¢ since the last time Google crawled that part of Amazon, so I was left with only the gasket to pad my order. I have absolutely no need whatsoever for this gasket, mind you, but purchasing it ended up saving me $6.19 in shipping fees.

The moral of this story is threefold.

• First, if you find yourself with a total of almost $25.00 at Amazon.com and are looking for the cheapest things available, do a Google search like the one above for something that would get you up to the magic number.
• Second, if your Ford or Mercury vehicle from the mid-to-late 90s needs a new water outlet gasket, ask me the next time you see me and I will gladly surrender mine.
• Last, but certainly not least, if you’re planning on doing any shopping for your family and friends on Amazon.com this holiday season, click on this link to get to Amazon and give me a referral bonus!

So…I haven’t posted for a while. Don’t worry, I’m still alive. Life as a grad student is good so far. Busy, but good. What have I been up to, you ask? I have been learning lots of things, teaching Java for beginners, and taking advantage of any and all opportunities for free food. Career fair week was great for that last item, by the way. I have also been sure to make plenty of time for fun. I’ve gotten to know quite a few other people in the department, and even played a couple of games of field crumpets. Liz is coming to visit this weekend, which is always exciting. Other than that, not too much is happening. Okay, time for class!

After getting done with work at IBM on Friday, I packed most of my possessions into my parents’ van for the move to Madison. On Saturday, my parents met up with me in my car in Rochester, and we drove to Madison. It was raining most of the day (and most of the weekend, for that matter), but the rain luckily let up for a while when we were carrying things into my apartment. On Sunday, we went on a quest for furniture. I called up some people who placed ads in Craigslist, and drove around town scouting out the local thrift stores. We found a well made dresser that almost-but-not-quite matches my desk at a thrift store for only $25, and found a mattress, boxspring, and frame through Craigslist for $50. Not bad deals, if I do say so myself. After all the furniture was inside, my parents drove back to the Twin Cities. They really helped a lot. If you’re reading this, thank you!

I’ve spent the past two days unpacking and running various errands. I got my parking permit, got my UW ID card, found out what textbooks I’ll need (only one if the bookstore is to be believed), got some groceries at the local Trader Joe’s, joined the credit union, and washed most of my dishes (since they had previously been in the garage for quite a while). I’m really liking Madison so far. Much more exciting than Ames, that’s for sure. I’ve got a couple more days to finish organizing everything, then Liz will be visiting this weekend on her way to Michigan, and then after she leaves it’s time to start grad school. Exciting, huh?

Those of you who know me well probably know that I’m a fan of the online webcomic xkcd, by Randall Munroe. I think his nerdly humor is quite amazing, and my friends all seem to agree. Here’s his latest comic:

There are plenty more where that came from.

I haven’t posted for a while. Nearly a month, wow. I’ve been busy, but in a mostly good way. Here’s a brief summary of what I’ve been up to:

Mellies in Chicago II

Near the end of July, I met up with a number of other current and former ISU mellophone players in Chicago for a weekend of fun. We saw the Sears Tower, ate some Chicago-style deep dish pizza at Gino’s East, went to a play, and more. A great time was had by all, and it was so nice to see some band friends again. It might not happen again for a while.

NordicFest and more

The weekend after the trip to Chicago, Liz came up to Rochester. We went out to dinner at Jenpachi on Thursday with some friends to celebrate our friend Elizabeth’s birthday. That was tasty (as always), and the Chinese guy who sings “Happy Birthday” was entertaining (as always). On Saturday we got up early to go to Decorah, IA to see NordicFest. There were a variety of tasty Scandinavian desserts, and I got to meet Liz’s sister and her family. It was a good time. Then after that we drove up to Eagan to celebrate my sister’s birthday. Good food, good cake, good times.

Murder Mystery Weekend

Last weekend I went down to Ames. Liz and I made some tasty spinach and artichoke pizza on Friday night. On Saturday, we got together with some other people for a potluck dinner party with one of those murder mystery games. After some initial problems (the tape in the box was for the wrong mystery), we got into the game and had a good time accusing each other of committing murder.

Work

In between all these great weekends has been a bunch of hard work at IBM. Our summer is almost over. Despite being in a bit of a crunch to meet our deadline, I think our project has been a success overall. Even so, it will be nice to be done with work for the summer and move to Madison for grad school. I’m really looking forward to it. They’re going to have me be a TA for an introductory Java class. That should be interesting, and my classes should be pretty good too.

Well, that’s my recent life in a nutshell. Check back later for whatever random thoughts I feel like putting on the web.

My last few posts have been related to the ongoing MacUpdate/MacHeist bundle sale. This post will be no different. But this time, I would like to focus on the structure of the promotion itself, namely the concept of “unlocking” applications. The way the promotion has worked is that seven shareware applications were available from the beginning. After 2,500 bundles were sold, an eighth application (Little Snitch) would be given at no additional charge to all current and future customers. At 4,000 sales, a ninth application (Intaglio) would be added, and a tenth application (TechTool Pro) would be added after 10,000 sales.

At first glance, this marketing gimmick seems brilliant on two different levels. If the promotion falls short of sales goals, the cost of running the promotion is lower since royalties need to be paid to fewer developers. Secondly, the “unlocking” levels give previous customers a personal incentive to tell their friends, families, coworkers, etc. about the promotion, thus ensuring higher sales levels. It sounds great, but in practice it doesn’t seem to work out so well.

Within the first couple of days of the promotion, sales moved briskly, getting to around the 2,000 sales level. The first couple of unlock points seemed to be a given, and the third was not too far out of reach. However, sales soon slowed to a small trickle. This morning before I went to work, just over seven days after the sale started, the bundle’s sales level was at 3,724. Over the weekend, the sales rate was sitting pretty steady at about 15 bundles per hour. This made the possibility of hitting the 4,000 mark by the promotion’s end at midnight tonight seem iffy at best.

The reason for this? Game theory and economics give some explanation. See, one individual purchase does little to bring the group closer to the 4,000 sales goal. Thus it is in any individual’s best interest to simply wait until others buy up to the 4,000 sales mark before taking the plunge themselves. No individual has any way of knowing how many other people are interested in buying the bundle, so it is best to just wait and see. However, it is better for the group for individuals to purchase as quickly as possible so as to make the deal seem better for future customers who find out about the deal. This explains the standstill. A bunch of people were wanting Intaglio to be unlocked, but they didn’t want to take the risk of being left out in the cold.

What did the promotion organizers do about this? They lowered the final two unlock points to 3,800 and 6,000, respectively. This happened just after I left for work this morning. Within an hour, Intaglio was unlocked, and sales only accelerated from there. Now, about 12 hours after the unlock points were changed, 5,526 bundles have been sold, at a rate of about 150 per hour (ten times higher than before). Keep in mind that at previous rates, the promotion was on pace to only sell about 4,000 bundles. So by lowering the targets, they actually sold a lot more bundles than they otherwise would have sold. Interesting how a scheme that seems so brilliant on its face could backfire so easily.

By the way, if you use a Mac, do look into buying the shareware bundle. It’s a great deal. Only about three hours remain, so act quickly!

GraphicConverter is another Mac OS X shareware application that is being offered as part of the MacUpdate bundle sale, which ends tomorrow night. GraphicConverter is one of the main reasons why I purchased the bundle in the first place. It is an application with a long history in the Mac OS X shareware community. The first version was released in the 1990s for the old 68k Mac platform. As the name implied, it was designed to convert graphic images from one format to another. Over time, the application was improved to include other features, like basic paint functionality, resizing, batch conversion, and more.

I first became a regular user of GraphicConverter with version 4.6 (back in 2003), when it came preinstalled on my old PowerBook G4. I’ve used it ever since as my main application for image editing. However, my old copy has recently started to show its age, as it was written using the Carbon APIs for the PowerPC processor. That’s why I was excited to see the latest version included as part of the MacUpdate bundle, so that I could have a version that runs natively on my MacBook Pro without the need for emulation or any residual Mac OS Classic compatibility.

On that front, I am more than satisfied with the new GraphicConverter. It loads faster, and has kept most of the old interface to which I have grown accustomed over the past few years. But on top of that, even more functionality has been added, including some pretty cool filters that can be applied to images. Just as an example, I’ll show the result of six different filters that have been applied to the picture of me that appears on my home page.

So, now I can not only use GraphicConverter as an image conversion program and as a glorified paint program, I can also use it to apply cool effects to a wide variety of images. All in all, I am quite pleased with the new version.

In an earlier post, I mentioned the Mac OS X shareware bundle promotion that MacUpdate is running until Monday. I finally broke down and bought it tonight, and I have started really trying out some of the programs. The first one was Cocktail, from a Swedish company called Maintain. It seems to be sort of a swiss-army knife of a program, setting all sorts of otherwise hidden preferences and performing various system maintenance tasks. It does things like optimize network settings, view system logs, change hidden interface preferences for various applications, and perform regular maintenance on system files.

I have personally done several of the things offered by Cocktail through the command line, but it’s nice to have these options at my fingertips. For example, in the past I used the “defaults write” command to change my iTunes settings so that those little arrows go to an artist’s or album’s songs in my music library instead of taking me to the iTunes Music Store when I click on them. With Cocktail, I can change that with one click. No googling for the command to run, just one click. It’s handy.

Still, I don’t know that I would pay much money for this convenience all by itself. Maybe $5 or so, but not the $15 that Maintain normally charges for Cocktail outside of the bundle. The program seems to be aimed at what I think is probably a small niche of Mac OS X users who wish to really fine tune their system, don’t wish to deal with the hassle of looking up arcane command line functions, and are willing to pay a premium for a bit of added convenience. It seems to be pretty popular at MacUpdate, though.

Verdict: it’s nice to have, but I wouldn’t have paid full price for it.

I just want to let you know about the folks over at willitblend.com, who recently released a video that pitted a Blendtec blender against an Apple iPhone. Those of you who have seen these videos before know that the blender always wins. They have tons of videos showing everything from a whole chicken to a can of Cheez Whiz (can included) going into a blender.

It’s kind of sad to see a perfectly good iPhone go to waste like that, but at the same time, I as an Apple shareholder don’t really care why people buy iPhones, so long as they buy them.