Michael Braun's Blog

Twitter has announced a new way for the company to make money. (You will recall that previously they had nothing, so actually adding something should be seen as a good sign.) “Promoted tweets,” or Twitter messages that are paid for by companies, will now show up at the top of a user’s feed if that user follows the company who paid for the message. It’s a good way for companies to get their message out to followers, and it’s a relatively non-intrusive form of advertising. Presumably users will be somewhat interested in the content of the ad, given that they have elected to receive promotional material from that company in the first place. Additionally, the ad can be hidden by the user, and each “promoted tweet” will show up just once per user. So, Twitter must be on the rapid road to profitability and living up to their massive valuation, right?

No. (You knew I was going to say that, didn’t you?) Twitter’s new “promoted tweet” ad model has a serious problem: scalability. That is, their ad platform can’t get much bigger if it should become popular. Other industries supported by advertising don’t have this problem. Take, for example, magazines. If they get more ad dollars, they can put more pages in the magazine. Newspapers, they get more companies buying ads, then they also get more space for news content. Even Google can engage in an effort to role out their third-party ad system to more and more websites. There is room to grow in all these ad-supported ventures. But there’s little room for Twitter’s “promoted tweets” to grow.

The problem is that Twitter can’t show too many promotional messages each time a user logs in, or the ads would become annoying, rather than potentially useful or interesting. And Twitter can’t sell too many promotional opportunities to companies before they run out of space. And Twitter may have issues with competing companies. Let’s say a user follows Pepsi and Coca Cola, and both companies pay for their message to appear on top. Who wins out? Which company wins out first? Can one company pay more to keep the other company’s ads from appearing? These types of conflicts of interest also hurt Twitter’s scaling.

It’s reasonable to suggest that any competition and log jam problems can simply be solved by raising prices for the messages. After all, take a look at television. It’s the perfect example of an ad-supported medium that has the same kind of space problem. Television can’t cram more ads into a program without cutting the length of the program; cut the program length too much and no one wants to watch any more. And television does raise the prices on their ad space, a perfect solution for the limited space problem.

But Twitter can’t do this, at least not right now. That’s because Twitter needs to demonstrate to companies that this new ad platform will really work. They need to offer evidence that users are actually engaging with the messages and helping the company make money. Normally, this would be an easy sell; just offer a few companies cheap rates in exchange for sharing data about how successful (or not) the campaign was. But interested companies already have free advertising to all their followers on Twitter. All they are lacking is the primo spot right at the top of the user’s list of recent messages. So Twitter really needs to demonstrate a good ROI for companies, especially if they hope to solve their scale problem by raising prices.

Now, maybe Twitter will find a way to make this work! Maybe they will become the first web company since Google to successfully monetize a product given away for free. But given that each “promoted tweet” can’t hang around forever (remember, they display just once for each user) and given the problem with scaling, you’ll have to pardon me for being skeptical.

The University of Michigan Press is serializing two novels on their Facebook page. The program started on July 18. They post a chapter each week, such that visitors (i.e. Facebook members who “like” the group’s page) can read the novels in their entirety by the end of the summer. This story has gotten relatively little press, but some organizations have covered the event as if it deserves attention, including the blog of a former classmate of mine. My question: when did a marketing gimmick become evidence of some kind of communication innovation?

This University of Michigan Press announcement is nothing new or revolutionary. Take, for example, the serialization of books. This is something that has been done for quite some time. Perhaps the most famous novel-serializer is Charles Dickens. Millions of high school students (especially those who have little patience with reading assigned books) can curse the serialization of these books, as it provided motivation for Mr. Dickens to make his novels longer and longer. The fact that this is being done again (albeit for free) is nothing but a hark to the past.

The other reason this serialization is nothing revolutionary is that it is clearly being done as a marketing gimmick. The University of Michigan Press is hardly a major publishing house, and the novels being serialized are unlikely to achieve mainstream popularity. But if group can get more people to receive their promotions on Facebook and generate news coverage to boot, then what do they have to lose by allowing people to read these two books without paying? It’s not challenging to see what is being done here. This does not speak to a change in the publishing industry, allowing people around the world easy and free access to quality literature. It’s just a marketing trick (and a relatively poor one – University of Michigan Press has 1,148 “likes” and a Google News search revealed no mainstream news coverage of their effort).

So why are we so easily fooled into thinking that a marketing gimmick actually represents something noteworthy? The reason is that fascination with social networking websites blinds us to actual innovation. Should Facebook be in the news? If the reporting is about a trend or fad, then yes. But if the reporting is about business practices, technology innovation, or communication changes, then the answer is no. This effort by the University of Michigan Press should be considered non-news. But it is positioned with just the right number of hooks (Facebook!, Electronic Books!, Free!) that is is designed to grab people’s attention, and ideally stir up some froth about how our modern world is changing everything. Tech observers and communication scholars alike should take a step back and ask themselves: is this a real change or just more advertising nonsense?

Starting Sept. 1, prices for Netflix are going to rise. My own plan (streaming + 2 DVDs at a time, unlimited per month) will rise from $14.99 to $19.98, a substantial increase. When this increase was announced, backlash was widespread as customers decried the price hikes. Netflix did not relent, and given how much money it costs them to run their business, the prices make sense. Nevertheless, it makes the best deal for content a little less of a good deal, and once again draws into question the feasibility of running an internet business drive by popularity alone. Other internet-based businesses should take note.

But I think there’s more to Netflix’s move than simply trying to bring in more revenue. Netflix is starting to position itself more strategically in the market. By raising the prices on their plans (specifically on those plans that allow customers to get physical discs), Netflix is raising perceptions of the value of their service and pushing customers toward making a tough choice: you can’t have it all, Netflix is saying, so you have to pick and choose. Netflix is forcing users to choose between Netflix, cable, and other streaming and rental services.

Netflix exists in a competitive and complicated marketplace. First, there’s cable television. Not satisfied with simply delivering a multitude of channels, cable companies have also entered the streaming content business with various “on demand” services. These allow customers instant access to a variety of premium television shows and movies, without having to pay for each show watched. Cable’s saturation in the market is pretty high (and may never get higher), making them Netflix’s number one competitor. Second, there are lesser competitors in other streaming services. Take, for example, Hulu, which offers next-day access to a wide variety of currently airing television shows. Hulu has also started expanding their back-catalogue, allowing them to better compete against Netflix’s streaming service. And finally, there are rental companies like Redbox and Blockbuster that compete against Netflix’s physical disc mailings.

With this level of competition, a low price from Netflix allows them to be an add-on service. Another $10-15 per month is a low cost for many consumers. Even if Netflix becomes that household’s main source of entertainment (as it is in my household), the price isn’t enough to make a consumer seriously consider canceling cable. In other words, Netflix’s prices are too low to force a customer to have to decide between their cable service and their Netflix.

In some ways, Netflix has done very well as an add-on service. Their offering of expanded streaming content, including a streaming-only plan, has caused a dramatic rise in the number of subscribers. And they have successfully expanded to Canada, making up a gigantic portion of streaming traffic in this country. But the market is competitive, and as other companies (like Hulu) look to add pay components to their services, Netflix must recognize that they have to stay competitive in order to stay in business.

Price increases, somewhat paradoxically, may offer the company a chance to become even more popular. Think about it like selling antiques. If an antique seller has a piece of furniture that isn’t selling, she might try to put the piece of furniture on sale. But if price in the antiques world represents quality, then the outcome of such a decision may actually reduce the chances that the piece of furniture will sell. After all, a buyer might conclude, it can’t be truly valuable if it’s priced so low. Raising the price may make the piece of furniture more desirable.

So these types of human decision making processes may actually work in Netflix’s favor. A price increase forces consumers to justify their Netflix subscription. Some, who may not have been committed to the service very much, may cancel when their price increases. Others, myself included, may double-down their support for Netflix. In order to justify the increased price, consumers may begin to reconsider what really matters to them. Do they need cable AND Netflix? Among those for whom Netflix is their top source of entertainment, it may mean canceling cable and sticking with Netflix.

For this reason, Netflix may not need to worry about losing some subscribers. It seems likely that Netflix’s profits come from individuals with higher tier plans (say, 2, 3, or 4 DVDs at a time). These plans make Netflix more money not because Netflix collects more from these customers but because these customers may be slower to return some DVDs. The fewer DVDs that Netflix ships for each account, the more money they make. So when Lauren and I get caught up watching Veronica Mars again on instant stream, we aren’t watching the DVDs we have. (And when that program was removed from instant stream, we went right back to our waiting DVDs!) Canceling cable (a large expense) may prompt some people to opt for more expensive Netflix plans. Netflix’s top plan costs $29.98, but is still just a fraction of what cable costs.

It seems unlikely that Netflix wants to lose subscribers, and their price increases may simply be a way to increase revenue. After all, licensing content is not cheap. But if even a small portion of Netflix’s price rise is done to increase commitment to their service and make them more competitive, then I applaud their price increases. To me, they seem very reasonable for a service that provides me with 90% of the television that I watch on a daily basis. With a price increase looming, I’m doubling my support for Netflix.

As the debate about whether or not to raise the nation’s debt ceiling continues, it is not surprising to see heated rhetoric on both sides of the argument. Some Democrats argue that not raising the debt ceiling is akin to treason, as the nation is obligated to pay its debt. Some Republicans argue that continuing to borrow is like an irresponsible family living far beyond their means, equating raising the debt limit to the problems that caused the great recession in the first place. Especially prevalent in arguments from both sides are positions about taxes or, as some have called them, increased revenue. Democrats say that the deficit cannot be closed without increased revenue, and Republicans say that Washington needs to suffer for a change. Regular readers of this blog likely know where I stand on this issue, but regardless, I think it’s valuable to start to examine the word “tax” that Republicans are so fond of using as pejorative.

What is a tax? A dictionary definition might say something like this, “A compulsory contribution to state revenue.” But how do Republicans define a tax? Somewhat more narrowly, I think, as they seem mostly to be railing against specific deductions from worker paychecks that occur BEFORE the worker receives the money. These types of taxes are called direct taxes. Less frequently demonized by Republicans are indirect taxes, or taxes like sales tax that stores charge customers. Indeed, in the Republican primaries for the 2008 presidential election, Mike Huckabee was advocating the abolishment of the income tax and a huge increase in sales tax. Seldom if ever considered by Republicans are fees and other charges levied against citizens, and considered none at all are the general costs of not paying taxes.

So when Republicans say that they will not support any bill that raises taxes, what are they really saying? Some may claim that the United States already takes in enough revenue to pay its debts, and thus increasing taxes is not necessary; such logic is lucid, but only if one ignores facts. Others may claim that taxes are too high already, and thus raising them further will only hurt people. But not a single Republican is saying that we are playing a zero-sum game, and that the needed money is going to come out of everyone’s pockets one way or another.

Let’s consider a contentious issue – subsidies for growing certain types of crops. When the government wants farmers to produce more of a certain crop (say, for example, corn to make ethanol), it offers to subsidize the purchase of those crops, thus allowing farmers to make more money selling the crop and keeping the price for consumers lower overall. That means that our tax money is going directly to farmers! Cutting these subsidies would seem like a reasonable course of action, but that’s true only if consumers are willing to pay higher prices for meat. Pigs and cows need to eat a lot so that they get big and we can butcher them. If it costs more to feed the animals, then their meat will become more expensive. Is this a tax? It’s certainly an increase in revenue for the government (as less is paid out to farmers), and this change is subsidized by consumers. Consumers could choose to purchase less meat (and no doubt some would), but without a behavioral change, this is amounts to a compulsory contribution to state revenue.

Take less contentious issues like the Social Security and Medicare programs. These two services from the government account for a large amount of money taken out of workers’ paychecks, and the programs are a large chunk of the Federal budget each year. But reducing these services means that older adults have less money overall. That money is going back to the federal government, even if it means mostly the ability to close the budget gap more quickly. I think this too fits under the umbrella of a tax, even if Republicans are reluctant to call it such.

Certainly there are areas of the Federal budget that can be cut without being considered a tax on individuals. Defense spending is the primary example of this. But that too means removing money from the economy. Consider the connections between military and private industry. Reductions in defense spending and the end of two wars means that many companies will find their own revenues severely reduced. This means workers will get laid off, and it means that the money these workers would have had is instead held by the government. It’s not a tax, but it’s getting difficult to separate between the two.

So when Republicans say they want no increased revenue, what they are really saying is that they want no overall rate increases in direct taxes. Their position on indirect taxes and incidental taxes is much more vague. Lower taxes sounds good only when considered from a non-principled standpoint. A fee-based system, which is what lower taxes end up producing, is what Republicans are ultimately going for. And the government that they envision, one that has basically no responsibility for any person, is what they hope to achieve by refusing to raise the debt ceiling. Republicans should recognize the fallacy of their own supposedly economically-driven logic and realize that cuts and revenue increases (yes, including raising taxes!) are the only way to close the budget deficit. It does not matter if the increased revenue comes directly from our pockets or is siphoned off in some other way; increased costs for citizens is the result of Republican plans to hold our country hostage to achieve their ill-conceived ideals.

I had the pleasure of sitting down with a 95 year old woman (who I’ll call Grace) on Friday to help her learn to use her iPad. She was attending Senior Summer School, a program here in Madison (and in select other locations around the country) for seniors looking for some intellectual stimulation. It was a challenge to help Grace learn to use her iPad, and this was somewhat surprising to me. After all, the iPad is supposed to be the computer for non-computer users. Even the Pope uses one so he can use Twitter. What I found was that the design of the iPad mattered very little. The main challenges that Grace faced were simply trying to learn the concepts associated with computer use.

Take, for example, the concept of an email address. You have a home address for mail delivery, so you must need an electronic address for email delivery – seems simple, right? But her instinct was to type someone’s name or home address. Apple has thought of this problem, so if you have your address book set up, then you can just type someone’s name and the app will give you the option to fill the field with the associated email address. But that requires that the address book be set up! Of course, Grace’s wasn’t.

The concept of user-uploaded content was challenging as well. Grace wanted to watch videos of a famous pianist Glenn Gould. So we went to the YouTube app, and I did a search for her. She was able to select a video from the list of results, and the video started playing flawlessly, even over the 3G connection. But it wasn’t clear to her why the video was poor quality, or why the video wasn’t focused on the pianist’s hands. The idea that a website could be useful entirely because of the collective actions of millions of people was totally foreign to her. For those of us better versed with these kinds of collective efforts, we recognize that we may have to look through multiple videos to find one that is just what we want. But that concept is very different from previous ways of getting video (for example, going to a library and checking out a video cassette).

The concept of touch navigation was challenging as well. I can imagine that this might be a disappointment for Apple, as they have pitched touch as a much more logical way to interact with a computer. But the main problem is that touch gestures are not always accurate. Grace, on several occasions, tried to scroll down a website but accidentally touched a link instead. Or she would try to type, but brush against another key, causing input errors. The real challenge then became getting her to recognize that something unintended had happened. It was a difficulty for me because I wanted to correct the problem to continue with whatever I was explaining. But doing so meant ignoring a very practical lesson on why something unexpected occurred.

Even the concept of the home screen and apps was confusing at times. If the entire device is basically a mystery, then it’s not clear when any one particular function is used instead of another. Sometimes it makes sense – mail for email, etc. – but other times the desirability of any one particular app just wasn’t clear. It means remembering a lot of different things just to perform basic functions. The user has to think of her goal, then try to remember what app fulfills that goal, and finally, spend time locating the app. It’s hardly a mindless process for a new user.

Now, that’s not to say that design elements weren’t part of the problem. The text and icons were very frequently too small. And often, the particular options on screen weren’t especially specific about what should be pressed (for example, the send button in the upper right corner of an email looked quite nondescript). In other words, there was a general failure of communication – the interface did not tell the user what the user’s next step should be. Even more problematic, it wasn’t clear why Grace wanted an iPad, other than she generally seemed motivated to learn new things and stay engaged. This made for problems with her email as she didn’t know anyone’s email address. Thus learning to use her email may not have been useful. Furthermore, if reading text off the screen was a problem, then it seems unlikely that getting news online would be appealing.

It’s a good experience to try to give instruction to a new computer user. This kind of effort seems especially valuable for anyone working in design. The challenge of making the concepts clear and then orchestrating the design seems somewhat removed from the aesthetic and usability focus of most designers. If designers can figure out a way to communicate the concepts through design, then usability for people like Grace should take a monumental leap forward.

Google’s Eric Schmidt had some harsh words for those who would sue Android handset makers for patent violations. He accused these litigious companies (Apple, really) of responding to Android with lawsuits rather than with increased competition and innovation. It’s a decent storyline, and I have little doubt that Mr. Schmidt truly believes it himself. But in making such claims, he is ignoring what Google owes to Apple for the success of Android, the very real patent violations that are occurring, and the innovations that Apple is continually rolling out.

First, let’s consider Google’s timing of their entry into the smart phone market. Google released Android as free and open software in November 2007, a full 11 months after Apple had announced their iPhone (and five months after the iPhone had gone on sale). The design of Android closely mirrored Apple’s iPhone OS, which itself was a departure from previous designs. Further, Apple’s product was a significant change for the smart phone industry, such that many Apple competitors declared it DOA. The iPhone, instead, would become the biggest smart phone on the market, even as Android now outpaces iOS in total devices. In essence, Apple created a smart phone market for the average consumer, and Android rode on their coattails. Google also followed Apple’s lead with the introduction of a store for selling smart phone applications. Google’s success can thus be credited, at least partially, to Apple’s innovation.

Second, the veracity of Apple’s patent claims must be assessed. The International Trade Commission has ruled in Apple’s favor in regards to two complaints filed against HTC and their Android-powered smart phones. Whether or not these complaints will ultimately be upheld remains to be seen, but once again, we see innovation by Apple being used by other companies. Just as I recommended that Motorola sue its competitors in order to make sure their patents are protected, Apple should do the same. The ITC’s initial ruling further confirms that Apple has valid claims. No doubt this is irritating to Mr. Schmidt, as profit margins for many Android handset makers are already thin; paying money to Apple for each phone sold may cause some companies to look to develop their own operating systems or reduce the number of Android phones that they make.

Third, Apple continues to innovate in their handsets. Consider the hardware changes in Apple’s iPhone 4. The high resolution screen has raised the standards by which other smart phones are measured. The front-facing camera (which Apple wasn’t the first to use) has made this feature standard on new smart phones. And consider the software advancements Apple has announced for iOS 5. Chief among them is the new iCloud service. While it does borrow features from other companies (including Google), its most noteworthy feature is the storage of online music. This feature was rushed out by Google ahead of Apple’s announcement, but Google did so without any agreement from record labels. It’s another example of Apple innovation that is copied by other companies.

If Apple was a struggling company, losing money out of every department (like RIM or Nokia), then suing would likely go hand-in-hand with a lack of innovation. But Apple is innovating and protecting their patent rights. The fact that such lawsuits annoy executives at other companies should be of little concern for Apple. And with a record setting previous quarter, I hope Apple keeps up its aggressive stance against what they perceive to be patent violations. It’s certainly not hurting profits – Apple’s profits were up 125% from a year ago. With return like that, I say keep up the innovation and keep up the lawsuits.

Lauren and I were at the wedding of her cousin this weekend (congratulations again, Kacie!) in Oshkosh, WI. Because there was a substantial time gap between the ceremony and the reception, and because we would be driving back to Madison rather than staying overnight, we spent some time looking for a coffee shop where we could pass the time. Thankfully, it ended up not mattering (we all went to Lauren’s grandma’s house) because there were not many options available. There appear to be just three independent cafes in Oshkosh: New Moon Cafe, Pilora’s Cafe, and Planet Perk. Considering you can find that many cafes on any two block stretch of State Street, it seems like there should be more. (Perhaps there are, but we were dependent on Google Maps for our info. If your cafe isn’t on Google, then it doesn’t exist for me.)

So why the lack of coffee shops? I think there are two forces we can blame. First, there’s competition from chain restaurants. Imagine you run a small cafe that serves decent coffee and some food items. In previous times, your restaurant may have opened early (say, 6 AM) to catch the breakfast crowd and the people getting off their third-shift jobs. You probably stayed open through lunch and then closed sometime in the late afternoon. You didn’t serve dinner, and who wants to go to get coffee after 5 PM? Now imagine a chain restaurant opens down the road. They stay open 24 hours a day. They have an expansive menu. They have reasonable prices. And they have a big sign up on the freeway telling people to exit now. It’s hard to compete against that. You don’t have the space for expanded seating, the kitchen facilities for an expanded menu, or the energy to stay open for longer hours.

These types of restaurants have invaded America’s small towns and made life difficult for these types of small independent restaurants. It doesn’t help that many of the independent places weren’t exactly thriving nor putting much money into improvements. The convenience and familiarity of a brand name restaurant easily wins out over a dingy, local establishment. And the willingness to take a big risk and compete was likely too much for many small cafe owners.

But the second force is actually the opposite kind of force – the Starbucks effect. In this situation, the arrival and popularity of Starbucks franchises in America’s small towns actually creates a market for coffee. It turns people into cafe-loving coffee snobs. Once you’ve convinced people to pay multiple dollars for a coffee-based beverage, they start to think of themselves differently. “Maybe I am the kind of person who really values good coffee.” And thus they start to look for places to get that coffee. There are three Starbucks locations in Oshkosh, and I bet not a single one of them has been open for more than ten years.

The expansion of Starbucks into smaller towns means that the town is undergoing a slow transformation. And its independent cafes reflect this. Each (linked above) has a website, talks about things like art and music in their cafe, and brags about their organic, fair-trade, gourmet coffees. These are exactly the kinds of things that Starbucks-raised coffee snobs will seek out. In another ten years, Starbucksification will be complete in Oshkosh, and it will be easy to find a nice cafe for a good cup of joe, a sandwich, and (perhaps most importantly) free wi-fi. The expanded competition into this area of the market has forced out under-performing, under-providing cafes and created a market for high-end, boutique cafes catering to a different crowd. But until this process is complete, Lauren and I will continue to make jokes about being in a cultural wasteland when driving through small town Wisconsin.

On June 11, 2010, I deleted my Facebook account forever. In the 13 months since, I have never had any serious thoughts about re-joining. In fact, it was surprising to me how a website that I had previously considered to be a relatively important part of my internet life could disappear so effortlessly. But what has life really been like without Facebook? Let’s review some of the things I miss and don’t miss about not having Facebook anymore.

First, the things I miss. Topping the list is establishing new social connections with classmates and other acquaintances. I frequently meet people in class, and I liked to connect with them on Facebook. The website made for a convenient living address book, giving me an easy way to contact people. Reactions from new acquaintances toward not being able to “friend” me tend to be ones of confusion. And reactions from friends who are moving away tend to be ones of distress. “How will we stay in touch?”, someone asked me a couple weekends ago. “Email?”, I lamely suggested. These two types of connections – new friends and geographically distant friends – lead me to a general sense of loss as I consider the information being shared on Facebook that I am not privy to. Email does work well, but that means adjusting to a new channel for some people.

So what about the things that I don’t miss? First and foremost, clutter! My number of reason for removing friends was that I didn’t care what they were sharing, and I didn’t like the rate at which they shared it. I had one particular high school acquaintance who seemed determined to post total nonsense multiple times per day. First, I hid his updates, then I removed him altogether. That lead to a general purge of friends, trying to get rid of the people whom I cared little about. Because there was no easy way to filter what I did and did not care about, the “news feed” became increasingly useless. And after Facebook removed the possibility to list favorite music as free text (and replaced it by automatically having you “like” bands), it meant the clutter was only going to increase.

I also don’t miss the privacy concerns. And I had a lot. First, I was concerned about who was seeing what I was sharing. Did I want all those random high school friends seeing my status updates? Second, I was concerned about how my data was being used, especially as Facebook became more and more integrated into other websites. Third, I was concerned about what Facebook would do next; if they are preparing for an IPO, then they need a way to make profits from that massive amount of user data they have. I didn’t want to be a part of that. Not having Facebook helps to put my mind at ease.

So Facebook free for one year, and it feels great. I don’t feel like I’ve lost friends (though perhaps some friends feel like they have lost me), and overall I’ve been just as happy without the service as I was with it. Without viable options for replacing communication over Facebook, one should be wary before quitting entirely. But if the service just isn’t that useful to you anymore, I can testify that my life has not been adversely impacted by my decision.

Today was hot, very hot. Tomorrow will be as hot or hotter. And Tuesday, Wednesday, and Thursday too. We are under an excessive heat WARNING through Thursday night. What does that mean for your week? 1. Do not go into the sun. If you must go into the sun, do not do so without first drinking a gallon of water. 2. Do not sit in your car without the air conditioning on. If you do sit in your car without AC, you will literally be roasted alive. 3. Do not go outside, even when the sun is down. Heat will kill you, whether from the sun or radiating from the ground. 4. Do not breathe. The hot air will scorch your lungs. 5. Do not stay in your house without the air conditioning on. Though rare, cases of human melting are sometimes reported during these types of heat waves.

I do not tolerate the heat well. Neither does Lauren. Between the two of us, we took approximately 42 naps this weekend. Not because we were tired, but because the heat made us pass out. We did exercise today, but it was at the gym. And we had to take two water breaks just to make it across the gym parking lot. The cat is hot too. She can barely be bothered to get off the floor. We do have our second bedroom air conditioned pretty well, but she doesn’t seem smart enough to know that’s where she should be holing up.

There is no end in sight for this heat. The warning ends Thursday, but the heat doesn’t. Friday has a predicted high of 90, and Saturday 88. That means the AC will stay running, and we’ll stay miserable. Tennis lessons, as far as I’ve heard, are not canceled, and I will NOT miss the lessons I paid for just because of the heat. So tomorrow and Wednesday, look for Lauren and me on the court. We’ll be playing and slowly evaporating away.

What to do with a student email that is riddled with typos? Some instructors may react with vitriol. I know some instructors who go over with students, at some length, how to properly send an email. A friend of mine once scared her freshmen students so much that the emails she received from them were overly formal. On the other hand, we can take a functional linguistic perspective and say that if the student is able to communicate the information needed, then the misspellings, typos, and poor grammar are unimportant. For me, I fall somewhere in between. A poorly written email often does make me laugh, but I seldom say something to the student about it.

The fact that instructors can say something to students, though, brings up an interesting point of discussion. How much is teaching like customer service? I can send a sharply worded email to an obnoxious student and suffer no repercussions whatsoever. But the same email sent to an obnoxious customer would result in a lost job. As instructors, we want what is best for our students, just like a great company wants what is best for their customers. So, in general, that would seem to push us toward emails that are polite and helpful.

But what exactly is best for our customer, the student? Because we aim to educate, not to simply provide service, the question becomes more complicated. A student who is consistently under-prepared for class would never receive a dressing down from a customer service representative. Rather, that representative would be expected to show patience with the customer. But in a classroom setting, reprimanding the student may improve the student’s education as he realizes that his behavior is substandard. Telling a student that her emails are unprofessional and should be corrected may have a positive outcome for the student as she begins to take more care in her messages.

So ultimately, perhaps teaching is exactly like customer service, except that instructors have a better sense of what is best for their students, and the achievement of that can take a decidedly non-friendly turn at times. The instructor should take heed, though, of helpful advice about actually achieving the desired results. An email can be much more sharp-toothed after evaluations have been completed. True customer service lessons would still advise a kind and helpful tone, even when a reprimand may be the most efficient and effective response.

The situation is all too common. A student has demonstrated solid effort throughout a course. He has come to office hours, completed work early so as to receive comment on a draft, attended optional review sections, and never missed class. But his test scores are low, his paper drafts are terrible and the finished products not much better, and in class, he seldom has anything to say. When the time comes to assign final grades, the numbers say fail but the instructor’s heart says pass. What should the instructor do?

If the low numbers come as part of a class with an overall high average, the instructor’s heart may be overruled. Everyone else seemed to be getting the material, so why wasn’t this student? Indeed, the student’s effort may work against him in this situation, as the instructor asks why all that extra help still couldn’t pull the student up. The conclusion is likely that the student just wasn’t learning on par with other students. But if the class has an overall low average, then the instructor’s heart can play a much bigger role. Many more students are on the bubble between passing and failing, and the numerous tough decisions can be helped by recalling how much work the student put into the course. If not all students can get pulled up, then at least the hardworking students should.

In using effort to raise student grades, we must ask ourselves, what is the value of the boosted grade? Let’s consider this question from three perspectives: the student, the student’s future, and the instructor’s future. The first perspective is the simplest. A boosted grade is likely to produce an immensely grateful student. Unless the student is as daft as his performance would indicate, he is likely to know that he is on the verge of failing. A boosted grade will cause him a huge sense of relief. The instructor can feel good thinking about the service she has done for the student.

But that’s just the short term. How does a boosted grade affect the student’s future? It does help keep his GPA higher and may keep him off academic probation, so in those two regards, it is good. But it may also give the student a false sense of his abilities. Consider specifically GPA requirements that certain majors may have. In my department, students must average 3.0 in their first 3 departmental classes; otherwise, they cannot declare the major. Thus boosting a student may make that student think that his chances for getting into the major are better than they really are. It may cause the student to strive for a goal that is not realistic. False hope given by a boosted grade may encourage the student when his performance should suggest that he develop a contingency plan quickly.

Not only is the student’s future affected, but the instructor may also be impacted when deciding to boost a grade. The problem with boosting a grade once is that it can develop into a consistent pattern of boosting. The instructor may grow committed to raising the grades of hardworking students. This undermines the accuracy of the grading system by lowering the weight of actual tests and assignments. It can also cause problems when classes have overall higher averages than previous semesters. In these cases, the instructor’s desire to stay consistent with past grade raises may skew the class’s average much higher than it should be. Though a grade boost for one exceptional student in a course where the averages are low is not likely to fundamentally change an instructor, it could cause gradual changes that overall have a detrimental effect.

Thus the value of a boosted grade is not always positive. Any short term relief that the student will feel could compound into long term problems as the student finds other instructors less understanding. And, though the instructor may initially feel she is doing a good deed, the long term consequences may be an inappropriate favoring of student effort over student output. For any instructor dealing with the question of grade boosts, these issues should be considered carefully before assigning final grades.

Unexpectedly, I found my way to a Google+ invite yesterday. It was unexpected because it was my advisor who was able to send it to me, and she had not requested an invite, but had instead been sent one by a colleague. Given that several people I know were already using it, it was good to finally get an invite. I spent an hour or so using the site yesterday, including finding new people to connect with, creating some new “circles,” and sharing info (including changing the permissions on some photo albums I already had up on Picassa). Overall, I like the service.

Though I did quit Facebook over a year ago, I am not adverse to the concept of a social networking site. Distrust of Facebook lead me to leave the site, as well as a sense that something better was going to come along. Facebook has become bloated, partly because so many people use the site, and partly because Facebook allowed third-party developers rampant access to various Facebook APIs, leading to lots of clutter. Of course, nothing better did come along. So while I’ve never seriously considered rejoining Facebook, but I have been waiting (somewhat impatiently) for something else. I tried Twitter and found it wildly unsatisfying (same experience I had the previous two times I tried to use it). Diaspora looks promising, but has taken way too long to generate anything. So with one Google+ invite, I was on my way to finding something good.

For me, the key feature of Google+ is the ability to easily determine who can and cannot see information that I share. Now, why is this key? What do I have to hide? Controlling who can see what information is a vast improvement over Facebook not because of the ability to hide information, but rather because it allows me to have control over clutter that others are exposed to. Take, for example, the distinction between friends and “UW Colleagues” – my two biggest circles. Friends should have access to my photos, but colleagues don’t need to be bothered by that. For something like academic articles, my colleagues should get to see that, but friends should not have to. In the case where someone is both a colleague and a friend, then I can put that person in both circles. It makes me feel like I am not spamming anyone when posting something.

A second nice feature is the integration of Google+ into the bar that appears at the top of Google’s homepage. Considering I visit Google more than any other site on the web, often defaulting there if I don’t want to leave my browser open to another website, it is easy to see when new activity has occurred on Google+ because notifications show up at the top of the page. Greater integration with Google search will make the service even more useful. The Google +1 is an interesting feature and giving me the ability to share a page directly from search will make it perfectly integrated.

Still to come will be a Google+ app for the iPhone. This should enable easy photo sharing, something I would like to use more. One benefit of a website like Posterous, for whom I was a professional blogger from fall 2010 to spring 2011, is that I could easily email a photo and have it posted on my page. Google+ integration with my iPhone will enable something similar.

The service isn’t perfect, as it lacks some key features that Facebook has. For example, you cannot add a circle to an event in Google Calendar, meaning Google+ doesn’t yet offer an invitation system. And the concept of “writing on someone’s wall” is absent from Google+. I’m not sure that feature is really necessary, but it could be interesting. Given that Google+ is still in limited beta form, these features (and others) are likely coming.

Google+ probably isn’t worth leaving Facebook over. But depending on how you use Facebook (and why you use it that way), transitioning over to Google+ may make sense for you. So, if you log into Facebook to see if there is anything new from people you care about but find there often isn’t, then switching sites may make sense; if you can connect with just the people you care about on Google+, it will provide you with better value. And Google+ is growing. Some analysts put the number of users at 10 million, with 20 million coming right around the corner. Yesterday, I was able to connect with 17 people that I know. These are small numbers compared to Facebook, but given that I never had more than 160 Facebook friends (and that around 30 of them were from high school), I’m not lagging that far behind.

Let me know if you want an invite (if I know you – sorry other readers!). I’m happy to send one along. Exciting things are happening on Google+.

Of all the types of challenges for an instructor to face, it’s a lot easier to deal with low grades than high. First, let’s consider why it’s so hard to deal with high grades. In some cases, high grades are deserved (a class filled with good students who all have worked very hard). But in other cases, they are high because assessment tools were too lenient or because the weighting of different items was out of whack. For example, last summer, I gave too much weight to writing assignments and not enough to the tests. As the writing had very little to do with the course content, and more to do with an incidental course writing requirement, my grades did not do a good job of reflecting student learning. It’s hard to grade students down, when their own calculations can often show that they are on track for a high grade.

But the challenge of low grades can be vexing as well. It requires two sequential decisions. First, do any students deserve a higher grade? Just like sometimes a higher percentage of As is deserved, due to a lot of great student work, a low percentage of As may be warranted due to a lack of good student work. If it is determined that higher grades are warranted, then the next question is who should be bumped up. Everyone? Only select students?

First things first – how to know if higher student grades are warranted. This requires determining how much of a grade is due to an assessment tool that is unusually hard and how much is due to a lack of appropriate student effort. A too-low percentage can be determined by comparing the actual percentage to the desired percentage – easy enough to do. Determining the amount of effort is a much larger challenge.

Here’s how I have chosen to assess it. First, was there improvement from test to test? Students may naturally have lower scores on a first test because they knew less of what to expect. But when they know better how to study, their scores should improve. Second, did students improve on any repeated assessment? For example, if students complete two papers of the same type, do they improve from one paper to the next? Third, do students show increased engagement after receiving low grades? A hard-working student should panic (slightly) if she receives a low grade. This should trigger the student to make use of all available resources to improve her score.

One additional item of consideration that I am using is perhaps controversial. I specifically repeated items from the previous two tests on the final test. As Tony noted, this does not give me information related to my construct, because it requires students (who were informed that items would repeat and had the past tests to study from) to memorize past answers rather than demonstrate actual learning of the material. I then used the percentage correct on these repeat items to assess if students should have their grades bumped. In this way, my construct became “effort”: if students were putting in effort, then their scores should be high on these repeat items. If effort is evidenced, then it demonstrates that students should be bumped up.

This brings us to the second question – who gets a bump? Given that the behaviors described above may apply to some, but not all, students, it does not seem appropriate to apply a decision rule to all students – bump or don’t bump everyone. Instead, it seems better to create a rule to determine who should be bumped and who shouldn’t be.

Here’s how I have chosen to assess this question. I created a bump index with a total possible value of 5. Students receive a point for each of the following items:
1. Above-median score for test repeated items
2. Correct answer on the item that was repeated on all three tests
3. Above-median score for class participation
4. Above-median score for extra credit earned
5. Correcting past tests for half credit

As you can see, this index is mainly constructed to measure student effort. As doing all these things required the student to put in additional time and work (whether memorizing old tests, extra class participation, coming to office hours, or correcting wrong answers), it is an indication the student has been working harder than other students. An above-median score on the bump index then suggests that that student is a good candidate for having his grade raised. It takes a student who put in a lot of effort, yet may have been hurt by assessment that was overly difficult, and allows him to receive the grade he might have gotten if the assessment had been easier.

The application of such a decision rule is also protective against cases of bias. No matter how good the instructor, there are always borderline cases where the instructor has to confront biases. For example, two students – one very likable and another less likable – who both have the same borderline grade. The decision rule allows the instructor to make a determination based on effort, not on personal feelings. If the less likable student came to office hours, made improvements, and demonstrated effort, then this student should get bumped. If the likable student was not effortful, then this student does not deserve a bump.

Of course, the items used in a bump index are somewhat arbitrary and definitely intercorrelated. This means that the likelihood of earning a point for one item indicates an increased likelihood of earning a point for another item. So, for example, a student earning extra credit for coming to office hours is also more likely to have shown up for those office hours for help in correcting her tests, and with her tests corrected is more likely to get a good score on the repeated items. This intercorrelation gives some reason to be concerned about the use of such a tool.

But, given the challenge faced by instructors with lower grades than desired, this index is a reasonable way to make a tough decision. And it gives the instructor both the power to explain a final grade in an objective way and a rule to rely on that is somewhat sheltered from bias toward or against a student. Tomorrow, I’ll consider ways in which instructors might explain assigned grades to disappointed students.

As my summer course wraps up, I am faced with the challenge of how to deal with scores that are lower than I would like. I didn’t have much of an issue with the first two tests, as I gave students the opportunity to redo their incorrect answers for half credit. It worked like a charm, and the scores rose to a level I was comfortable with. But for the final test, there is obviously no way to give students the same opportunity. And since the scores are lower than I wanted, I decided to apply a curve.

Here’s what I did to curve them. I took the difference between the total possible score and the highest student score, and I added that number to all scores. So basically, I pulled everyone up by the same number of points. This works well if you think of the test as already a good discriminator between levels of knowledge. I didn’t see any items that needed to be tossed out; in other words, there was no evidence that one item was not measuring what I thought it was. But because the average was lower than I wanted, I treated the test as if it was X points too hard. Thus correcting for that number of points makes sense.

But there’s a problem with the assumptions that lead me to that decision. With a small sample size, it’s harder to tell if certain groups of students (say, for example, students performing at C level or worse) were systematically disadvantaged by the test. So maybe there was more misunderstanding for that group than for other groups. So a standard score increase may not help that group as much as it should. Furthermore, should one student be a standout performer, that student is able to wreck the curve for everyone else, leading lower performing students to get a bump that is pretty meaningless.

These problems can potentially be fixed by using a non-linear curve so that lower performing students get a greater increase to their scores than higher performing students. This kind of curve was used in my “correct wrong answers” method. Say, for example, a quiz was worth 30 points. A student who received a 0 has the most to gain from correcting his wrong answers (15 points). A student who earned a perfect score gains the least (0 points). It essentially squashes the distribution of scores by halving the total possible range.

But, as with any curve, the outstanding question is how does the application of the curve affect the ability to make decisions about grade assignment. If low performing students get a large bump, then is my confidence in the grade they earned undermined by the use of the curve? Is it possible to curve up so much that the score no longer reflects the category in which a student belongs? With a set number increase, I have confidence that I am not over-curving. But if that’s 7 points for the top student and 15 or 20 for the bottom, have I artificially bumped a student into passing range when he should have failed?

The challenge of answering these types of questions is something that should make all instructors nervous about curving. Though it can make students feel relieved to know that a hard test will not doom them, it takes away power when ultimately deciding on final grades. I do have hesitations about even my simple form of curving, and my next greatest challenge is figuring out those final grades. Had averages on tests been where I wanted them to be, I wouldn’t be so worried. But with everyone riding a low percentage right now, I need to figure out the next steps to take. More on that issue tomorrow.

Have you had a chance to use Madison Public Library‘s new computer system yet? Though there were some initial kinks to work out and some bugs still remaining in the system, overall it is a huge improvement, and I am a big fan. Here’s why.

The new interface is welcoming and decently easy to use. I like the bright colors and the large text. True, I don’t need the large text to be able to read it off the screen. But I like that it is welcoming for people with many levels of eyesight. The account login link is easy to find in the upper left portion of the screen. This makes my most desired feature easy to find. The LinkCat-managed user accounts are also easy to use and the design of that website fits well with the library’s design. This upgrade alone was worth any pain and difficulty. I like the features offered in this part of the website, including the ability to see pictures of the covers of some items (not available for some older items) and the ability to see my checkout history (not enabled by default to respect user privacy).

The search feature works much better. The old search feature was simply dreadful, returning null results when the search was specific and inaccurate results when the search was general. Now, 9 times out of 10, the item I want is the first or second one listed. And search field is prominently placed at the top of each page. This makes it easy to search from anywhere, rather than having to navigate back to a special search page.

Holds are a breeze to place. For each item record, there’s a helpful link to place a hold. You select what library you want your hold sent to and you’re done! You can even add multiple books to your shopping cart and then place a hold on all of them at once. (At least, in theory you can do this; there seemed to be a bit of a bug that stopped me from doing it the one time I tried. I hope that bug has now been fixed.) Back in the old system, where placing holds was a time-consuming and confusing process, I was frequently likely to skip the hold, especially for books I wasn’t sure I really wanted to read. But now that it is so easy, I’m placing holds like never before!

Holds are even easier to place thanks to the new lists feature, a fun way to find out about books. Lists are collections of books centered around some particular organizing topic. For example, there is a list of Newbery Award winners, making it very easy to place holds on those winners that I haven’t yet read. Though the available lists are rather limited in number currently, users can create their own lists, and I am sure it won’t be long before users will create lists and share them with others. This will add a really useful and fun way to find out about new things to read.

There are a few features that could be improved, like the ability to set a default library for where to have your holds sent. The search feature could also be made even better by returning books as the top result, when relevant, no matter the ranking. Frequently, when the search doesn’t do so well, it’s because some number of audio recordings show up ahead of the books. And a default image should load for book covers that the system doesn’t currently have. Right now, a “no image found” symbol shows up, which is not ideal. And it would also be nice to have a real login name, instead of my library card number.

But aside from these few gripes, the new system is just great. It’s made using the library a lot more fun. I like looking at my list of holds and getting excited about what will soon be available for me to read. Well done, Madison Public Library!