Amie Street is a music website with a difference — the prices for downloading songs are not fixed, but are determined by their popularity. The price for a song varies between free and 98 US cents. Unpopular songs are free, and the price rises the more popular they get. This seems like a very clever idea to me. The biggest problem for new musicians is getting noticed. If your songs are free, that can help to offset (to some extent) the fact that you’re unknown. This pricing model also gives users an incentive to search out the best music, since if you can find a great song before it becomes popular you can get it for free or very cheap. For other users the price of a song is a good guide to its popularity — the price is high only because enough people have been willing to buy it previously. You can also earn credits on the site by recommending songs.

One complaint I have is that the pricing formula isn’t disclosed. This makes the price information less useful. Is a 60 cent song twice as popular as a 30 cent song? As a musician, how popular do I have to get before I receive 98 cents per song? I also wonder if prices are allowed to fall once they have gone up, if a song becomes unpopular, as it is bound to do? Or do prices just go up and stay up forever? Regardless, it looks like a very interesting system, both for musicians and listeners.

It really helps to have influential friends. Apparently, Steve Fossett’s buddy Richard Branson got in touch with his buddies at Google and asked them if their satellite images could be used to help in the hunt for Steve who has been missing for almost two weeks after his light plane disappeared in Nevada. Subsequently, a set of fresh high-resolution satellite images were taken of the relevant area after he disappeared and uploaded to Amazon’s Mechanical Turk service.

Mechanical what? The Mechanical Turk is a website that allows people to collaborate on tasks that only humans are good at (at present). They call it ‘artificial artifical intelligence’. For example, looking for Steve’s plane in thousands of satellite photos. I tried it myself and found that I can check an image in less than 10 seconds. A computer may be able to do it faster, but computerised image analysis is probably not as accurate as human eyes. The Mechanical Turk site facilitates projects like this, where many people can contribute a small part of a larger data-processing project that would be difficult or impossible to program into a computer algorithm.

The Mechanical Turk includes a payment system where you can get paid for each little bit of work that you do. When searching for Steve you don’t get any reward except the warm fuzzy feeling of doing a good deed. But for other projects on the site you can get paid small amounts (usually less than 10 US cents) per small task that you do. Still, at say 5 US cents per task you can make US$10 per hour if you can complete each task in around 18 seconds. For people in developing countries that’s a pretty good wage, and that kind of speed seems feasible for relatively small well-defined tasks.

I have a couple of comments. The first is quality control. Since people get paid by the quantity of work that they do, they have an incentive to do it as quickly as possible. In the search for Steve, they’ve set each image to be examined by multiple people before it’s eliminated from the search set. That’s one way of doing quality control — get multiple people to do the same task and compare the results. However that could be expensive when you have to pay people to work. Alternatively, I’m not sure if Amazon does this already, it might be useful to incorporate some kind of reputation system for workers. If people can establish a reputation for high quality work, and if this could be linked to how much they are paid, then maybe people would have an incentive to work accurately in the first place. It’s a matter of setting the right incentives for speed versus quality.

My second comment is on the design of the site, which I think is bad. As with Amazon’s main website, it’s not very well laid out, and most importantly it’s not optimised for speed. I could review Steve images much faster if the site design were changed a little. For example I had to scroll down to see each image and mark whether it contained anything interesting or not, and then scroll up again to submit the result. The important buttons are just too far apart, and this slows down the work speed. Also the entire page has to reload between images, which is slower than it should be.

Overall, I think the Mechanical Turk is a great idea, provided that the quality control problem can be solved and the site is optimised a bit further. It’ll be interesting to see if this kind of outsourcing (’crowdsourcing’) can be economically viable, or whether the costs and difficulties associated with generating quality work are too great.

If you want to help look for Steve, click here.

I’ve been reading the Google public policy blog. It’s not a very active blog, but it’s interesting to read about the topics that Google cares about in terms of government policy. There’s the controversial issue of ‘net neutrality’, of course, plus recent FCC spectrum auctions, and even IT law in India.

Reading the posts, Google tries to portray itself as nice and friendly, with the interests of ‘the Internet’ and the common good at heart. However, we should not forget that they are a successful multi-billion dollar business, and can be expected to behave as such. For example, there’s a post titled More online ad acquisitions = more competition. It’s about Google’s plans to acquire DoubleClick, a firm that specialises in placing advertisements on websites. Google also does basically the same thing, through its Adsense and Adwords services. There has been a bunch of consolidation going on in online advertising, as they describe in the blog post, with Yahoo and MSN and others acquiring advertising providers.

Somehow, Google manages to spin this situation into one in which there is ‘more competition’. Their argument seems to be that the aggressive acquisitions indicate fierce competition to be a leader in this market. As Google says:

It means that each of the leading Internet companies believe that they can position themselves to succeed in the online advertising space — through the free market, and without government intervention. These companies believe that there are many ways to compete in this business.

However, from an economic point of view it’s hard to believe that fewer firms equals more competition. The more independent players in a market (however big or small) the more intense the competition between them. Mergers and acquisitions will lead to higher prices than would otherwise prevail, unless there are some significant cost savings from combining the operations of two firms. Admittedly, as Google says there are other things going on in the market such as entry of new startups, but everything else equal fewer firms equals less competition, not more.

According to a recent study commissioned by Chitika, the top 50,000 blogs earned US$500m in advertising revenue in 2006. Chitika sells online advertising, so I guess they have an interest in knowing how big their market is. You can read the report of their study but it’s a little disappointing. The estimates are based on revenue data from Chitika, and none of the data is revealed, so it’s impossible to verify the results or try to come up with better estimates. I guess the data is commercially sensitive for them, so that’s fair enough.

I think this study is a reasonable first attempt at estimating blog revenues, but there’s some dubious aspects to their methodology though. For example, it seems they only had revenue data from Chitika, whereas there’s many other sources of revenue for bloggers such as Google’s Adsense, so they tripled the Chitika revenue figures to get an estimate of the total. I guess they had to do something, but this assumption isn’t justified at all. Second, the fit of their model isn’t very good and there’s no diagnostics or other statistics presented to help verify that their model is good. So it’s hard for me to have confidence in their headline result of US$500m revenue for the top 50,000 blogs. They could have presented statistical estimates of the reliability of their results too, but they didn’t.

However, I do believe their finding that the distribution of revenues across blogs is highly skewed. What this means is that a few of the top blogs collect a disproportionately high share of revenues. The Chitika study finds that the top 1% of blogs accounted for about 20% of total revenue and the top 5% accounted for 50%. That’s an extremely skewed distribution. If this is correct, and I suspect it is, blogging is an activity where you have to be a superstar to make it work. Being average isn’t going to do it. This is probably because popularity feeds popularity, and the ability to earn revenues might increase more than proportionately with your audience. In economic terms, there’s probably “increasing returns to scale” in blogging. What this means is that the ‘market’ will only support a limited number of profitable blogs, but this doesn’t mean that those that are profitable can take it easy. Since there’s good returns to be had from being successful, there will be fierce competition from outsiders who want to try to break into the market.

In all, there will be a lucky few bloggers who can make good income from it, but it’s not going to be an easy road to riches.

Invisible Engines by David Evans, Andrei Hagiu and Richard Schmalensee was recently published by the MIT Press. All are leading economists and this book takes an in-depth look at software platforms as examples of two-sided markets. You can buy the book from Amazon here. Interestingly, the MIT Press has also made a pdf version of the book available for free download under a Creative Commons license. Therefore, I’m able to offer this book as a free download from my site here. I read the entire pdf version of the book on screen without printing it out. It was the first time that I read a whole book on my computer screen and it was an interesting experience that I’ll talk about later.

For now, I wanted to briefly review the book. As I said, it concentrates on software platforms. These are two-sided platforms that are based on some form of software (and often hardware too). Examples include computer operating systems, mobile phones, videogame platforms, and so on. Excluded are two-sided platforms like credit cards, shopping malls and real-estate agents, since these are not really based on software. The first three chapters provide background technical information on software platforms, and set out the basic economic issues. The authors do a good job of highlighting the key economic issues of information goods (high fixed costs and low marginal costs), and the key issues facing two-sided platforms — how to get both sides on board, what pricing strategies to choose, what features to offer, and how many ’sides’ the platform should have. These questions are illustrated with practical examples from various industries that show how different firms have tried to solve various problems.

The next five chapters give very detailed case-studies of five specific industries based on software platforms — Personal computers and operating systems, videogame consoles, PDAs, mobile phones, and digital media players. To be honest, I found the volume of facts about these industries in these chapters overwhelming, and I could only skim-read. It’s impossible to retain so many facts, so I didn’t see the value in reading carefully. These chapters would be a useful reference for anyone who wants to know the detailed history and business strategies that have been used in these industries. The chapters are well organised, but it reads a little like the authors got their research assistants to dig up every possible fact about these industries that they could find.

The book gets better again starting from chapter 9. Chapters 9-11 deal with the basic issues in two-sided platform strategy. The first question is whether to pursue a one-sided or two-sided strategy. A platform can be a one-sided business if it takes care of one side of the market itself. The most common example the authors cite is Apple’s iTunes and iPod system. Apple could have created a digital music platform but licensed the development and production of the portable players to others, but has instead chosen to create a closed system where it produces the hardware itself. Basically everything in the iPod + iTunes system comes from Apple, so it is not a two-sided platform. Other digital media services such as those offered by Real Networks have adopted different strategies, and these are true platforms that bring together hardware producers as well as music listeners and music publishers.

The second question is what pricing structure to use, what prices to charge the two sides, and how to get both sides on board through pricing. The authors discuss the various tradeoffs and constraints that affect these choices and again illustrate with examples from different industries. One theme that comes through here is that it is rarely the case that ‘one size fits all’. As in all markets, unique features need to be taken into account when setting prices in any particular market. The unique possibilities are compounded in two-sided markets, as factors on each side of the market affect decisions made on the other side. The authors do a good job here, although I would have liked to see a more explicit table or summary setting out exactly the implications of different factors.

The third question is the range of features to include in the software platform. The authors address this through standard economic arguments about bundling. Almost all goods and services can be thought of as a bundle of components. For example, a car contains many different components that could in principle be sold separately. The authors use the economic theory of bundling to explain why software platforms might choose to include some features and not others. They also explain why the number of features offered by platforms has increased over time, as a result of competition and technological progress.

The book concludes with a chapter that looks to the future and discusses how software platforms are likely to further shape industries such as retailing and traditional news media. There is also a fairly extensive bibliography of relevant economics and business literature.

I have a couple of minor complaints about the book. First, it seems like it was written in a little bit of a rush, and would have benefited from a bit more editing. I think it’s a bit bloated with industry facts, and there is some repetition in some places. Second, this might seem very picky but I found it annoying, the chapter titles are meaningless. For example, chapter seven is titled “Ba-BA-Ba-BAAAAH”, and is actually about the mobile phone industry. This is annoying because if I want to go back to the chapter about mobile phones later, I have no idea which chapter it actually was.

However, in all I think this is a very good book and I am recommending it. The material in the first three and last four chapters will be of interest to analysts, consultants and businesspeople in all markets that are based on two-sided platforms. Although the book is devoted to software platforms, most of the basic economic principles will apply equally well to other platform industries. If the book was a little more concise and there were fewer trivial facts, I would give it 4.5 stars out of 5, but because of those things and the other minor annoyances I’ll give it 4 stars.

Buy from Amazon here, or download a free pdf version here.