Friday, March 31, 2006

Two Interesting Mobile Ideas

Today's Economist has an article on Hotxt, a company that offers a service seeking to replace SMS. Their idea is to install what seems to be J2ME client software on the phones and use GPRS for sending messages. Since GPRS costs significantly less per 160 characters of content than traditional SMS, they can offer messaging at significantly lower cost - an interesting proposition for teenagers, the most heavy texters of all.

What a nice arbitrage idea: The technology seems almost trivial, but if these guys get the marketing right, they may as well succeed. More importantly, they need to avoid getting blocked or disintermediated by the mobile operators, who would be more than happy to keep their >90% margins on SMS.

Earlier today at ETH, I saw a fantastic demo of barcode recognition on cell phones. The main difference to existing approaches was that it actually worked.

Once this is available on everyone's consumer cell phone, it will be the greatest thing since sliced bread. Barcodes - unlike RFID tags - cost nothing to make and are already on every article sold in every supermaket. Many interesting applications suddenly become possible: Imagine you're in a DVD store and really want to have that latest George Clooney movie. The price tag seems rather excessive, so point your cell phone at the barcode, hit a button, and receive a price quote and a short review from Amazon.

Surely, some existing price comparison sites will want a piece of the action. As so often, my guess is that the winner will be whoever makes the application extremely easy to install and returns results pages with high-quality information. Everyone will be happy - except for the stores, who may start putting black stickers over barcodes for their high-margin goods.

Thursday, March 23, 2006

Google Finance: "More Love" or: The Last-Mover Advantage

Google Finance has just come out. I'm kissing Yahoo Finance goodbye.

This wouldn't have been necessary. In this amazing post, Jeremy Zawodny mentions that all the cool features could already be in Yahoo Finance, as ideas for doing them have been around for years. Ticker search sucked, news could have been overlaid on graphs, and so on.

Seems like it isn't the first time this is happening to Yahoo: The release of GMail was also a wake-up call for the competitors.

Why did this happen?


It's easy for management to stress that "only the paranoid survive," and that all products should constantly improve. In reality, it's often "good is good enough". Yahoo Finance and Yahoo Mail were both state-of-the-art products until the arrival of their Google counterparts. Heck, Yahoo Finance probably also made a lot of money with all those banner ads.

Keeping such a site running takes a lot of work. It's easy to patch on but if you really want to change something core such as charting, there are lots of forces against it: The "you better don't change it because people like it how it is" fraction and the sheer complexity of replacing something at the center of a product.

This doesn't apply just big products from big companies: Even Basecamp (from the popes of agile development) or Reddit are unlikely to change in their core features or appearance.

Revolution ...

There are always going to be young, hungry kids – or still-hungry industry veterans – willing to challenge convention, question authority and re-write everything from scratch. Paul Graham has, in some ways, built a business on his founders' youthful energies.

These guys have two things going in their favor:

Just as the Google Finance people had already seen – or even worked on – Yahoo Finance, they have the last-mover advantage: They've seen your product and can copy the good parts, leave out the bad, and invent cool new features you didn't have time to get around to.

They can also put in more love: They have the luxury of focusing on the user experience, the user experience, and nothing but the user experience. They don't need to worry about links to the conglomerate's Horoscopes product, cannibalizing existing offerings or following the personal tastes of the VP of Product Sameness.

... and how to keep it from happening

So what could Yahoo and others do to avoid such surprises in the future?

Frankly, I think they're doing the right thing: They're buying young and promising companies, when they can still be had for $20 million. Had David Filo said "Dear Larry, here's $20 million for the your PageRank algorithm", the world would be a different place. Yahoo Photos was a good product, but not the leader. Buying Flickr has propelled Yahoo to the front row.

In the long run, even the world's smartest people and restaurant-quality food won't keep you from falling into complacency. I doubt it's possible to launch renegade products inside the company: Taking a handful of talented and hungry employees, and promising them a big reward for creating a new finance product from scratch may have worked. Still, the internal forces of cubicle envy would have crushed any such attempts.

Significant changes will almost always come from new entrants. Today's new entrant was Google Finance. Who is next?

Sunday, March 19, 2006

Christo's Gates in Zurich

Credit Suisse, one of the two large Swiss banks, is currently undergoing a re-branding campaign. To create the fuzzy warm feeling that's needed get people to entrust them with their fortunes, they built an gigantic temporary exhibition hall in a prime location of Zurich to show an exhibition of last year's The Gates installation by Christo and Jeanne-Claude in Manhattan.

I have always been skeptical about Christo's projects, since they seem like such a waste: Thousands of gates and tons of fabric were manufactured for their New York project. Apparently, much of this gets recycled, but only after lots of using up lots of energy.

Still, this exhibition shows that Christo's projects do leave a lasting impression, not by their physical permanence but by becoming a part of the public memory. The images of floating fabric in Central Park are indeed beautiful and the orange color of the gates stands in brilliant contrast to the surrounding winter city landscape, especially in some helicopter shots.

My personal conclusion is that Christo's art is actually "lobbying art": Sure, designing the wrappings and gates is highly demanding artistic work. However, the greatest achievement is getting approval for their projects. For the Pont-Neuf project, it took them 10 years until they got approval from the mayor of Paris (then Jacques Chirac). In New York, they started lobbying the Parks Commission and touring the community boards in 1979 to get permissions.

To have the energy and dedication to convince the 'conservative opposition' of the benefits of their projects requires a blindingly strong vision.

The exhibition also contained a screening of a video about the artists' Pont-Neuf project. The most memorable part of it is where two middle-aged Parisians are loudly arguing about whether or not the bridge's wrapping constitutes art. One of the men closes the scene by saying: "You and me, we don't know each other. We have never met. If the bridge hadn't been wrapped, we wouldn't ever have talked to each other."

Wednesday, March 08, 2006

Vint Cerf at ETH Zurich

Since Google Switzerland opened its doors in Zurich about 2 years ago, ETH Zurich has enjoyed lots of attention from the company: We had Larry and Sergey visit, and saw presentations by Senior VP Urs Hölzle and former Director of Research Monika Henzinger.

Today, Vinton Cerf, co-inventor of TCP/IP and Chief Internet Evangelist at Google came to hold a presentation. He talked about a wide range of topics: The current Internet adoption rates, the properties of and possible extensions to IP, and his view of where the Internet is going. He closed with a rather far-out topic: The interplanetary extension of the Internet.

The thing I've been liking the most about Google talks is that a large portion of the time at the end is devoted to questions from audience. This time, there were tough questions about Google's censorship in China as well as the privacy dangers of entrusting your personal data with Google.

Overall, Vint seemed like a fun guy. Here are some quips from the presentation (Not sure if these are 100% correct, I'm reconstructing them from memory):
  • "When I started at Google, I had some discussions with Larry and Sergey about my job title. For some reason, they were opposed to 'Archduke', so we went with 'Chief Internet Evangelist'."

  • "When we looked at these statistics of the highest Internet penetration rates per country, we noticed that Scandinavia was very far up there. We couldn't figure out why this was, until we realized they had very long winter nights up there."

  • "Back in the day, MCI, the company I worked for, had a product called MCI Mail which charged $1 for the delivery for each e-mail. Today, this would be inconceivable."

  • "So I talked to the Netflix president and asked him how many DVDs they shipped through the postal service each day. Turns out the number is 1.7 million. So we made a little calculation. It turns out the aggregate bandwidth of the U.S. postal service is pretty high." [By my calculations, it's 739 GBit/s, with each DVD holding 4.7 GB]

  • "It's easy to imagine that in the near future, your fridge will be connected to the Internet. So when you go shopping, it will send you SMS about what ingredients are missing to make that favorite Spaghetti meal of yours. […] However, your bathroom scales will be on the Internet, too, and will record daily weight measurements. Since it communicates with your fridge, it may as well put you on a diet. In fact, your fridge may even refuse to open the door. […] So we may not want to connect our fridges and scales to the Internet, after all."

Tuesday, March 07, 2006

Will Money Make You Happy?

Much like MIT has Harvard just down the street, ETH Zurich has the University of Zurich just across the street. ETH students like myself know that the 'Uni' has cuter girls and better food.

Sometimes, they have more interesting research, too, such as this dissertation (in German) by Alois Stutzer, which contains a chapter on whether higher income leads to happiness. Without going into much detail, I want to quickly show two very interesting charts.

Income vs. Happiness

In our ever-more materialistic society, it seems like having more income should automatically lead to more happiness. This chart shows the responses of US inhabitants when asked to rate their happiness into the categories of "not too happy", "happy", and "very happy".

The number of people answering "not too happy" drops from 23% for the lowest incomes to 6% for the richest households, while the "very happy" answers rise from 16% to 44%.

This is pretty strong evidence for the "money will make you happy" theory, but there is also contradicting data: The dissertation mentions another study for Switzerland in 1992-94 where happiness surprisingly drops from the second-highest to the highest income class.

Today, some economists believe that the relationship between the two is actually concave. I find this assumption very reasonable: When your income is $10'000, you'll be much happier about an addition hundred bucks than when you're pulling a million dollars each year. In jargon, the marginal utility of extra income decreases at higher levels.

Another important point is that individuals often compare their income with that of their peers. People whose reference group has higher earnings than themselves tend to be less happy than those surrounded by equals.

The Hedonistic Treadmill

Between 1946 and 1991, the per-capita income in the US rose by a factor of 2.5, from $11'000 to $27'000. (This is in 1996 dollars, so inflation is factored out.) At the same time, the average "happiness rating" dropped from 2.4 to 2.2.

Why is this? Stutzer argues that it's due to an effect called 'aspiration adaptation': As their income rises, people continuously adjust their ambitions and physically get used to the pleasures of higher income.

When an employee gets a raise, it instantly leads to a higher level of happiness. Over time, this employee will get used to the higher compensation, and return to the former level of happiness.

This phenomenon is dubbed the "hedonistic treadmill" and explains why the continuous rise in incomes and comforts was barely enough to keep happiness levels constant in the US.

It is the reason why people will always want more. When you win a Nobel prize or a gold medal at the Olympics, you'll first be very joyful. After some time, these winners will start comparing themselves to others who were even more successful: Marie Curie, who won two Nobel prizes, or Marc Spitz, who won 7 gold medals at the Munich Olympics. Better watch out for this when you get that call from Stockholm.

The Answer

In summary, I believe the answer to "Will more money make you happier?" is "Yes, but after a while you'll get used to it."

Lots of Pretty Snow

During the night from Saturday to Sunday, Zurich saw an unprecedented amount of snowfall: 110 cm (or 3'7") in a single night, a new record.

This much wonderful snow just had to cause havoc: Even the super-prepared Swiss street cleaning crews couldn't get the streets and tramways cleared, which is why almost the entire local transportation system broke down on Sunday morning. Shocking to locals like myself, who are used to trams arriving exactly on schedule, or train conductors apologizing for even a 5 minute delay.

But it was a lot of snow: Some of the sidewalks are still covered with it, about knee-deep. I wish I had proper boots.